Curry Recipes Online
Curry Chat => Lets Talk Curry => Topic started by: livo on February 07, 2019, 03:39 AM
-
Can you expand a bit on that x1.5 per doubling? Is it that if you double, then rather than doubling the chilli you only increase by 1.5? And if doubling again, and again ... then what?
There's an implicit, logical dead-end to that rule. If we continue to increase the size of the cooking pot at some point there will be no need for any spices at all! It's the cooking equivalent of perpetual motion. It's just not logical.
First of all SS, obviously no one is going to increase the size of a pot as to the point of approaching infinity, and at which the functional incremental spice increase would approach zero.
To expand a bit, yes. This is a theorized and often utilized practice pertaining to the increased quantity of spicing in relation to the increased volume of primary ingredients in a dish. For each doubling of the primary ingredients there is an applied increase of only 1.5 X the amount of certain spices (as listed above) but certainly including chilli. This may or may not be necessary in the first instance, ie; in going from a single serve to a double serve in the case of BIR style cooking.
So as not to insult anybody's intelligence, I won't do the sums for us. Obviously there is no need to be FOAM (fixated on accurate measurement) and rounding up or down to an appropriate closest measure would be more than acceptable. This is a fairly general cooking practice and I have in the past found several sources confirming the application of the adjustment method. It may be related to the lack of need to use extra ingredients when they are not really required (economy) or it may be related to the actual fact of spice overload of linear scaling, or possibly a combination of both. My reading certainly indicated it the be to prevent over-spicing of dishes. I've posted some links in relation to this theory in other related threads in the past. I guess I could trawl the internet again to see if I can re-find references to this material.
-
In addition to previous links to supportive dialog. Here http://www.curry-recipes.co.uk/curry/index.php?topic=13774.msg121623#msg121623 (http://www.curry-recipes.co.uk/curry/index.php?topic=13774.msg121623#msg121623)
See Tip # 3 here: https://shop.mybluprint.com/cooking/article/doubling-recipes/ (https://shop.mybluprint.com/cooking/article/doubling-recipes/)
(Also interestingly tip #4 on fats and oils.)
Here https://civilizedcaveman.com/meal-plans/double-a-recipe/ (https://civilizedcaveman.com/meal-plans/double-a-recipe/)
http://www.canadianspiceassociation.com/spice-industry/322-2/spice-usage/ (http://www.canadianspiceassociation.com/spice-industry/322-2/spice-usage/)
On these pages, Chilli is actually recommended to be increased by a factor of only 1.25. This is only a couple that explain this recommendation in detail. There are many more although the Universtiy of Missisipi worksheet on the subject is clearly wrong. After stating this;
" Increase the herbs and spices by 25% (0.25) ......
Heat builds in recipes quickly, especially when adding hot seasoning such as red
pepper (cayenne), mustard, cloves, and peppercorns."
the worksheet actually goes on to show the amount actually increased by 125% for double, using a conversion factor of 2.25 instead of 1.25. OOOPs https://theicn.org/resources/354/no-time-to-trainshort-lessons-for-school-nutrition-assistants/105432/lesson-34-adjusting-a-recipe-with-herbs-and-spices.pdf (https://theicn.org/resources/354/no-time-to-trainshort-lessons-for-school-nutrition-assistants/105432/lesson-34-adjusting-a-recipe-with-herbs-and-spices.pdf)
-
Is this related to the fact the chilli heat is transmitted to our palate via the layer of oil which coats the curry, rather than the bulk of the sauce itself? If the surface area of a curry increases in proportion to the square root of its increase in volume, then should the chilli content increase by 1.41 (square root of 2) each time you double the volume?
In other words, scale the chilli powder in proportion to the increase in surface area of the curry ( the bit where the oil "hangs out"), rather than volume. And before someone starts taking about different shaped dishes, obviously "all else must be equal", i.e. same shape.
.
-
Root two is certainly what I have always thought was probably the optimal ratio, but in real terms the difference between root two and 1.5 is sufficiently small that the latter can serve as a good approximation.
** Phil.
-
Is this related to the fact the chilli heat is transmitted to our palate via the layer of oil which coats the curry, rather than the bulk of the sauce itself? If the surface area of a curry increases in proportion to the square root of its increase in volume, then should the chilli content increase by 1.41 (square root of 2) each time you double the volume?
In other words, scale the chilli powder in proportion to the increase in surface area of the curry ( the bit where the oil "hangs out"), rather than volume. And before someone starts taking about different shaped dishes, obviously "all else must be equal", i.e. same shape.
Pretty much no. On all accounts.
Firstly, the chilli heat is not just transferred by the oil. Don't believe me? Throw a raw green chilli in you mouth and start chewing. You know what the effect of that is, right? And where's the oil there?
Secondly, the oil is distributed throughout the curry as an emulsion. So there's no "surface" effect.
Frankly I'm bored with the subject as I don't believe it has any merit. If one chilli is needed in one curry then n chillies are needed in a curry n times the volume to achieve the same chilli-heat effect. Anyone who claims otherwise is bonkers.
Otherwise tell me what the mechanism is that allows less of an ingredient to be used to achieve the same effect purely because the bulk quantity is increased. Or maybe we're to believe that it's some sort of homeopathic mechanism whereby you reduce the active ingredient to increase the ingredient's effect. Or some other undefined witchery.
-
@livo
Regarding your links. You'll note none of them are in any way scientific and specifically they offer no scientific explanation for their scaling figures. It all smacks very much of; well that's how my granny used to do it so that's how I do it. Yeah, ok.
In particular, from the Canadian Spice Association:
Hot red pepper (typically 1,000 to 2,000 ASTA Heat Units in foodservice packs) builds up even more quickly than the herbs. For the first doubling, the red pepper can still be doubled. But, after that, use only
-
The person who could have shed real scientific light on this vexed topic (Professor Jack Pridham, founder of "Chemophilia", a web site devoted to debunking chemical myths) sadly died a few years ago. I have asked someone else (a friend who is a professor of human nutrition) who may be able to shed some light.
** Phil.
-
I'd be happy to be proved wrong Phil ... as long as it's backed by good science and not pixie logic. ;D
-
I might go in for pisky [1] logic occasionally, but my scientist friends most certainly do not !
** Phil.
--------
[1] Cornish spelling of pixie.
-
Right, I just had a quick trawl through six food-science books I have to see what they have to say about this. Amazingly there is barely any information at all. One didn't even mention chilli or capsaicins at all - even allowing for variations in the spelling of chilli!
Significantly though none of them made any allusions to a rule-of-thumb for how to deal with spice quantities when multiplying-up recipes. I really would have expected that information as it's such a common question. Anyway, I have a couple more weightier tomes that I can't locate at the moment but I'll report back if I find anything.
-
Little or no chilli (or much of any spice) in base gravy is what allows me to conclude that it is directly scalable SS. It isn't out of synch with the notion that chilli is non linear. As I've said previously, the scalability of chilli is possibly a situation in which the direct linear conversion factor is simply not required in order to attain the flavour or heat level sought. Although, my reading actually says that you should not scale it directly. It is something that I have only tried in mild dishes that I wanted to remain very mild. I went along the path of caution to ensure success for the task at hand.
I had read before that direct doubling on the first iteration was possible for single serve dish, but thereafter, subsequent iterations should have the ratio of 1.5 applied. I had not previously seen the ratio of 1.25. I guess that if you directly double the first and then use 1.25 you will soon end up with the same as 1.5 all the way. I don't think there is any scientific link to the mathematical square root of 2 but a scientist may have evidence to support this.
I have just provided links to some examples of where this "rule" has been proferred
Also worth note is the fact that some of my links are referring to large batch cooking of 50 - 100 - 200 serves. I wish I could find one of the sites I came across 4 years ago as it was presented by an accredited food scientist and it was relevant to smaller quantities. It is worth noting that batch cooking does not use the same ingredient ratios as single serve cooking. This is the same principle in practice.
Maybe you should adopt the University of Mississippi method of reducing chilli by doubling it then adding an extra 25%. ;D
-
I'd like to add this comment to the thread in order to provide some leeway to both sides of the argument and allow for reasoned debate pending scientific input, if any can be found. Possibly someone with direct experience in large volume cooking of curry can contribute to this discussion.
The BIR method recipes are given using accepted (sometimes even uniform) measures, but anybody cooking them is likely to modify to their own taste, or not use these measures exactly, and so differences will occur from one person to the next and one session to the next. I rarely use a tsp or TBSP any more as I just sort of know approximately what they look like and realise that it isn't critical. Different batch, brand, variety or freshness of spices will be another contributing factor in variation to flavour and heat level. In terms of the BIR cooking method, anything more than a double serve quantity is probably not going to be prepared in an aluminium frying pan, and so by that reasoning a direct linear doubling of spices is the normal and acceptable approach. The thing is that using the X 1.5 approach would probably provide a dish with very little, if any, discernible difference, given the above listed variability factors. After you'd cooked a dish once, either way, you'd simply adjust it for your next visit. I would,
However, once you start to get to the second iteration of doubling to 4 serves, and certainly doubling again to 8, you are looking at a different cooking mechanism. The ability to flash fry over high heat is reduced significantly. As SS has already pointed out, this change in the cooking dynamic may have something to do with the flavour and heat transfer from the spices. Note: may have! I don't know definitively. What I can say is that when I cooked large batches of mild curry in a modified BIR style, the reduced spicing provided the dishes with adequate and acceptable spicing. I don't know what the result would have been had I scaled linearly. They were mild curries to start with and remained mild. A Madras or a Vindaloo may be a whole different scenario. I would possibly consider doing 2 X double serve to get 4 serves but I wouldn't do 4 X to get 8 and I certainly wasn't going to do 10 X to get 20 serves of a single dish.
Many non-restaurant takeaways out here have their main curries batch cooked and sitting in heated bain maries. I have witnessed the counter attendant call out to the cooks that more of a particular dish is required and then seen a new container brought out to replace the old. Not added to it by the way. They usually have a blank space in the rack to facilitate this procedure. I would doubt that this style of cooking is done in single or double serves. I do not know if these large batch dishes are prepared using linear or non-linear scaling. I will ask my man K if he is familiar with this process and his "chef's view" on levels of incremental spice scaling.
One of the links in my previous post instructs to add chilli in 3 stages when you are trialing a new increased quantity recipe. I'm not sure how adding raw chilli to a dish at the end of cooking, or even in the middle, would work out though.
-
I believe I have at least most of the answer to this brain teaser now and it's blindingly obvious once pointed out.
When true bulk cooking is done, so we're talking fifty or one hundred or several hundred portions in one go, the chef is not going to use a teaspoon to add the spices because of the sheer quantity involved. And the spice, specifically chilli, will be in a bulk container, not the little bags most of us at home buy.
What's the difference? Well, the bulk supplies are, relative to the small packets, heavily compacted (are you having a light-bulb moment at this point like I did?). So when the chef scoops out the spice with a large implement, maybe a measuring jug, or purpose-made spice scoop, there is likely to be further compaction. This means the density of the added ingredient is different to that which would be obtained from a teaspoon lifted from a small packet of spices.
So if a recipe called for one teaspoon of chilli for one portion and we bulk cook one hundred portions and work out what 100 teaspoons are in volume (this doesn't work if done by weight), let's say it's five large scoops, then because the density of the bulk-container chilli is greater those five large scoops might be in reality equal to about 120 teaspoons chilli as measured by the single-portion teaspoon.
So in a nutshell that's it. In true bulk cooking if you scale up by volume you have to reduce the total quantity of chilli to account for increased packing density in bulk-container spices. And I don't think most of the rule-of-thumb scaling factors specify whether it should be by weight or volume. Nor do they specify that this bulk scaling factor only really applies to true bulk cooking and not home bulk cooking that might run realistically to scaling up a recipe by no more than ten portions and where the same packet of chilli is being used for one teaspoon as for ten.
Therefore, for home bulk cooking, if you make one curry and use one teaspoon of chilli then you should use ten teaspoons if you make ten, twenty teaspoons if you make twenty etc., as long as you use the same scooping implement and the same pack of spices each time where the packing density will be relatively constant.
And of course no scaling should be necessary for any quantity of bulk cooked recipe if you scale by weight because then a true linear scaling of the ingredient is achieved. Cooking method differences between small quantities and bulk quantities would still have to be taken account of though.
And also, by the way, I have found scientific backing for the fact that the perception of chilli heat or pungency is on a exponential scale. But that does not in any way alter the above.
-
Purely on the basis of my own empirical experience (but supported by the large number of authorities who maintain the same), I continue to believe that a two-person curry should not contain twice the chilli (by volume) of a one-person curry. Sadly my hoped-for advice from my friend the professor of nutrition did not materialise; she says that the question is too remote from her own research interests and expertise [1].
** Phil.
--------
[1] No longer true
-
she says that the question is too remote from her own research interests and expertise.
Or, put in South London street urchin speak, dunno innit!
-
I believe I have at least most of the answer to this brain teaser now and it's blindingly obvious once pointed out.
When true bulk cooking is done, so we're talking fifty or one hundred or several hundred portions in one go, the chef is not going to use a teaspoon to add the spices because of the sheer quantity involved. And the spice, specifically chilli, will be in a bulk container, not the little bags most of us at home buy.
What's the difference? Well, the bulk supplies are, relative to the small packets, heavily compacted (are you having a light-bulb moment at this point like I did?). So when the chef scoops out the spice with a large implement, maybe a measuring jug, or purpose-made spice scoop, there is likely to be further compaction. This means the density of the added ingredient is different to that which would be obtained from a teaspoon lifted from a small packet of spices.
So if a recipe called for one teaspoon of chilli for one portion and we bulk cook one hundred portions and work out what 100 teaspoons are in volume (this doesn't work if done by weight), let's say it's five large scoops, then because the density of the bulk-container chilli is greater those five large scoops might be in reality equal to about 120 teaspoons chilli as measured by the single-portion teaspoon.
So in a nutshell that's it. In true bulk cooking if you scale up by volume you have to reduce the total quantity of chilli to account for increased packing density in bulk-container spices. And I don't think most of the rule-of-thumb scaling factors specify whether it should be by weight or volume. Nor do they specify that this bulk scaling factor only really applies to true bulk cooking and not home bulk cooking that might run realistically to scaling up a recipe by no more than ten portions and where the same packet of chilli is being used for one teaspoon as for ten.
Therefore, for home bulk cooking, if you make one curry and use one teaspoon of chilli then you should use ten teaspoons if you make ten, twenty teaspoons if you make twenty etc., as long as you use the same scooping implement and the same pack of spices each time where the packing density will be relatively constant.
And of course no scaling should be necessary for any quantity of bulk cooked recipe if you scale by weight because then a true linear scaling of the ingredient is achieved. Cooking method differences between small quantities and bulk quantities would still have to be taken account of though.
And also, by the way, I have found scientific backing for the fact that the perception of chilli heat or pungency is on a exponential scale. But that does not in any way alter the above.
Absolutely Classic ;D Your roughly saying you would add 100 tsp chilli to a 50L pot SS? Profit and costs in the bin.
-
I don't think that's a light bulb moment. Some references are made to "bulk" cooking, but others are clearly referring to recipe adjustments on a more likely home scale. Either way, the underlying principle is consistent with reference made to the non-linear increase in spice with each subsequent increase in portions.
Increase portions by 100%, means increase spices by 50%, or in some cases only 25%. (in some this is done after the first instance)
-
I don't think that's a light bulb moment. Some references are made to "bulk" cooking, but others are clearly referring to recipe adjustments on a more likely home scale. Either way, the underlying principle is consistent with reference made to the non-linear increase in spice with each subsequent increase in portions.
Increase portions by 100%, means increase spices by 50%, or in some cases only 25%. (in some this is done after the first instance)
Good theory Livo and you can
-
Absolutely Classic ;D Your roughly saying you would add 100 tsp chilli to a 50L pot SS? Profit and costs in the bin.
1. Show me where I said that.
2. If you knew how to edit your posts to show just the salient portion you are referring to I wouldn't have point 1
3. If you disagree, show me the science. ;)
-
Therefore, for home bulk cooking, if you make one curry and use one teaspoon of chilli then you should use ten teaspoons if you make ten, twenty teaspoons if you make twenty etc., as long as you use the same scooping implement and the same pack of spices each time where the packing density will be relatively constant.
50L pot roughly 100 portions. You say Therefore, for home bulk cooking, if you make one curry and use one teaspoon of chilli then you should use ten teaspoons if you make ten, twenty teaspoons if you make twenty etc., as long as you use the same scooping implement and the same pack of spices each time where the packing density will be relatively constant.
Did you see it?
-
Therefore, for home bulk cooking, if you make one curry and use one teaspoon of chilli then you should use ten teaspoons if you make ten, twenty teaspoons if you make twenty etc., as long as you use the same scooping implement and the same pack of spices each time where the packing density will be relatively constant.
50L pot roughly 100 portions. You say Therefore, for home bulk cooking, if you make one curry and use one teaspoon of chilli then you should use ten teaspoons if you make ten, twenty teaspoons if you make twenty etc., as long as you use the same scooping implement and the same pack of spices each time where the packing density will be relatively constant.
Did you see it?
What? The only person who wrote "50L pot roughly 100 portions" is you!
So again, show me where I said it and show me the science for your theory.
-
Nice try SS, unfortunately your lightbulb moment was not so illuminating.
-
The mathematics.
Assumptions.
- 1 tsp chilli powder per serve starting point.
- 1 tsp = 5 ml.
- 5 ml (1 tsp) of chilli powder = 2.7 g.
- 1 metric cup = 250 ml (50 tsp)
- Linear scale applied to initial double from 1 to 2 serve (commonly acceptable to both theories).
For the sake of mathematical convenience, enabling direct comparison, let us consider a bulk cook volume of 128 serves instead of 100, this being 6 iterations of doubling after the initial occurrence, ie; from 1 serve to 2. Also a combination of 32 serves added to 64 serves. 96 serves. The combination of 32 and 64 is not exactly consistent to the theory but it is closer to 100 serves and allows examination of the numbers.
128 serves
Linear scaling.
1 X 128 = 128 tsp chilli powder = approx 2.5 cups = approx 345 g.
Non Linear scaling factor of 1.5 after initial double.
2 X 1.5 ^ 6 = 22.78 tsp chilli powder = approx 0.5 cup = approx 62 g.
Non Linear scaling factor of 1.25 after initial double.
2 X 1.25 ^ 6 = 7.63 tsp chilli powder = approx 20 g.
96 serves
Linear scaling
1 X 96 = 96 tsp chilli powder = approx 2 cups = approx 260 g.
Non Linear scaling factor of 1.5 after initial double.
2 ( 1.5 ^ 4 + 1.5 ^ 5 ) = 25.3125 tsp chilli powder = approx 0.5 cup = approx 68 g.
Non Linear scaling factor of 1.25 after initial double.
2 ( 1.25 ^ 4 + 1.25 ^ 5 ) = 5.49 tsp chilli powder = approx 15 g.
Conclusion
As to be expected, there is considerable difference in the amount of chilli required and this difference would increase with further iterations applied. Even I would have to think that a scaling factor of 1.25 is a little bit weak on credibility, but I haven't tested it so I don't know? Possibly not so on scaling already bulk food quantity to even bigger bulk, ie; 100 to 200 serves etc.. If (please read IF, because I haven't actually done it) the 1.5 factor of scale does provide ample spice levels, then there is a reduction in spice usage of between 74 % and 82%. Not insignificant.
Disclaimer: This is purely to illustrate the difference in amounts of spice that would be used in the application of the different scaling theories. I have not ever cooked in these quantities and do not offer this as proof of anything.
-
This is another thread I have been following closely as I cannot get the 'heat' when I bulk cook that I do from 1-2 serving dishes, trust me I have tried. I used to be able to make a minced beef chilli that had you sweating before you tasted it (but it didn't make your mouth numb) so I guess my taste senors are just desensitised after so long.
However, I do get what SS is saying, in the same way I get that we use 'Double concentrated' tomato puree in a large number of the recipes provided. I can't argue whether single concentrated has less oil but more water or whatever else might be in the production of double concentrated tomato puree but it is different to a tin of chopped tomatoes.
So, given compaction or whatever, I can understand that a measure by volume of 5ml 'loose powder' might be different in strength to 5ml 'packed powder'.
I have lots (a dozen, maybe two at least) of jars, filled with powders and whole grain spices and they hold either 200g, 400g or 800g but when refilling, I often have to 'hammer' them quite repeatedly on my work top to make room for the last few grammes (except the Cassia Bark).
This is a very interesting debate, thank you to all that are contributing.
BAJ
-
128 serves
Linear scaling.
1 X 128 = 128 tsp chilli powder = approx 2.5 cups = approx 345 g.
Non Linear scaling factor of 1.5 after initial double.
2 X 1.5 ^ 6 = 22.78 tsp chilli powder = approx 0.5 cup = approx 62 g.
Non Linear scaling factor of 1.25 after initial double.
2 X 1.25 ^ 6 = 7.63 tsp chilli powder = approx 20 g.
...Even I would have to think that a scaling factor of 1.25 is a little bit weak on credibility...
You have doubts about the quantity for 1.25 scaling factor livo but consider the 1.5 scaling factor. 62g of chilli powder divided between 128 single servings is 0.48g per portion or a bit less than a fifth of one teaspoon of chilli powder per portion. So even with this better case of scaling we are expected to believe that one fifth of a teaspoon per portion when bulk cooked has the same chilli heat as 1 teaspoon in a single portion! :o
I'd be prepared to concede some ground if the figure was, say, 4/5 teaspoon per bulk cooked portion based on some as yet unknown effect that extracts more capsaicin into the bulk. But 1/5 equating to the effect of 1. No. Just no. And of course that would be worse for the 1.25 scaling factor (equating the effect of roughly 3/50 of a teaspoon to 1).
It also occurs to me that, specifically regarding curries, bulk cooking wouldn't require a linear scaling of oil as you can cook spices, onions, etc. in far less oil at the start of the cook than a linear scaling would suggest. And so, the very ingredient which is acting as the major carrier of the capsaicin is being reduced. If anything the quantity of chilli should be increased in bulk cooking to offset this not reduced.
-
...so I guess my taste senors are just desensitised after so long.
It's a well established fact. If you're a chilli monster you will have desensitised the pain sensors that give us that chilli-heat buzz.
So, given compaction or whatever, I can understand that a measure by volume of 5ml 'loose powder' might be different in strength to 5ml 'packed powder'.
Pretty basic isn't it. So why aren't people getting it?
I often have to 'hammer' them quite repeatedly on my work top to make room for the last few grammes (except the Cassia Bark).
Exactly. Even at the domestic level we have this packing density problem. A measure of spice taken straight from the packet would differ considerably to the same measure from a compacted jar.
-
No offence, SS, but I think you are going in the totally wrong direction with this compacted spice nonsense.
I have followed this discussion with some interest as it is, unusually for here, relevant.
I have no answers but I am interested in what I see are the three main points to this discussion.
1/ Is the use of chilli when scaling-up recipes linear or non-linear?
2/ If it is non-linear, what should the multiplier be?
3/ and most importantly (for me), why?
I have difficulty getting my head around the fact, as SS put it, if you cook two exactly the same curries, mixed them together, they would have the same heat level as each individual curry. If you cooked the same curry, with double the ingredients, the heat level would increase. And from purely anecdotal evidence, I believe that the non-linear theory is true.
I would love to know the science behind it, so keep it up guys :)
-
...I think you are going in the totally wrong direction with this compacted spice nonsense.
Sure I am. Because we all know that compacted spice, which contains more of the spice than uncompacted spice, for equal volumes, has exactly the same effect. Right?
Doesn't really encourage me to read anything more you have to say on the subject if you believe that.
This discussion is like trying to convince religious adherents to give up their beliefs. No matter what logical argument you put they'll always override you with their faith, which, by definition, is an unbreakable belief based on no evidence.
-
This discussion is like trying to convince religious adherents to give up their beliefs. No matter what logical argument you put they'll always override you with their faith, which, by definition, is an unbreakable belief based on no evidence.
I think that the exact opposite is true, my friend. You are the one sticking to a ridiculous belief despite all contrary evidence, and are making a mockery of the whole discussion with this absurdity.
-
You are the one sticking to a ridiculous belief despite all contrary evidence
Yes, that must be it. Would you help me out by summarising that evidence for me please?
-
I would be happy to, if you would clarify your 'argument' about compaction. If your argument is true then the resultant bulk curry would be hotter than the non-compacted spice one. Isn't that the exact opposite of what this whole debate is about?
I have made curries in one portion and then doubled all the ingredients to make two. The double portion has always been hotter using the chilli from the same jar.
-
From trying to scale up on a number of occasions i have settled on the idea of two portions at a time as a max then add to the pot until you have got enough, i feel the problem is making sure the spices are cooked out, easy when doing a portion at a time, but more difficult with a large mass.
Regarding chilli which is a dominant spice if scaling up and the rest of the spices have not been cooked out then in my humble opinion the chilli will dominate,
If you are going to attempt a large scale up you have to know what you are doing and have the burners capable of proving the heat.
-
I would be happy to, if you would clarify your 'argument' about compaction. If your argument is true then the resultant bulk curry would be hotter than the non-compacted spice one. Isn't that the exact opposite of what this whole debate is about?
No, that is exactly what this whole debate is about. The hypothesis, with which Santa disagrees (for perfectly good reasons), is that when one makes a curry $n$ times larger than a 1-person curry (for $n \in 2
-
Thanks for clarifying that, Phil.
Still makes very little sense to me since we are, I assume, using the same spices in each dish.
-
By the way, I am not ignoring your "if I cook two identical curries ..." gedankenexperiment, it's just that my brain hurts every time I try to reconcile it with my own hypothesis ...
It has stopped hurting :) All is now blindingly clear. No-one is suggesting that if you make $n$ 1-person curries you will need less than $n$ portions of chilli, regardless of whether you are planning to serve them individually or pour them all into one huge bain marie. The suggestion is, rather, that if you try to cook the $n$ portions in a single pan. all at the same time, then you will not need $n$ portions of chilli but rather less. The exact amount needed may be of the order of $n^{n/{n+1}}$ {1,00 1.58, 2.28, 3.03, 3.82, 4.64, 5.49, 6.35, 7.22, 8.11, ...} but no-one really knows. So, since you have already told me it is not a gedankenexperiment but a real one, all you have to do is to cook three curries : two 1-person curries, each with the normal amount of chilli, and one 2-person curry, with twice the amount of chilli. Then pour the two 1-person curries into a large pot, and get as many people as possible to tell you whether your 2-person curry is the same heat as, less hot than, or h0tter than, your two 1-person curries that have been poured together. Oh, and report back !
** Phil.
-
Still makes very little sense to me since we are, I assume, using the same spices in each dish.
Yes, the same spice (one has to assume or there is no hope of a resolution), but not the same density of that spice. Bulk curries should be made with less of the spice compared to home-made single portions because as I and Phil and Bob-a-Job have already said, the packing density of a tub or large jar or bag of catering quantity chilli will be greater than a much smaller packet of chilli used at home. So any measure of catering chilli will contain more chilli than an equal measure of home chilli and thus you need to use less of it when scaling from single dishes to bulk. It's a really imprecise relation because of the variability of packing density, and probably only part of the overall answer to bulk scaling. but it would apply to any powdered ingredient, not just chilli.
-
I will say this only once.
Same spice, same jar, same spoon, same density......does not give the same heat when multiplied on a linear scale.
How's the search for the Loch Ness Monster going btw?
-
It has stopped hurting :) All is now blindingly clear. No-one is suggesting that if you make $n$ 1-person curries you will need less than $n$ portions of chilli, regardless of whether you are planning to serve them individually or pour them all into one huge bain marie. The suggestion is, rather, that if you try to cook the $n$ portions in a single pan. all at the same time, then you will not need $n$ portions of chilli but rather less.
I'm tempted to say ... well, Duh! (But I'm too polite so I won't ;D )
...all you have to do is to cook three curries : two 1-person curries, each with the normal amount of chilli, and one 2-person curry, with twice the amount of chilli...
Nope. Unless you are going to proffer the bonkers notion that two one portion curries when combined somehow change character compared to the single portion curry, then you only have to make two curries. Those being the single portion and the double portion (with everything scaled linearly by a factor of 2).
And, which I haven't stated before, I have done this with a madras or vindaloo in the past (too long ago to remember which though). That has been the only time I've done a domestic "bulk cook" recipe. And at that time as now I didn't have the notion that anything other than linear scaling was required. It was a four portion cook in one pan, all ingredients scaled linearly, bar the oil and I have no recollection of thinking, goodness me this is blowing my head off, compared to my normal vindaloo/madras. So anecdotally I still believe linear scaling is appropriate at home.
-
Same spice, same jar, same spoon, same density......does not give the same heat when multiplied on a linear scale.
Ok, and I say it does ... and ne'er the twain shall meet!
I'm still waiting for some logical reasoning for your statement though.
-
Apart from the spice density factor, which is a real consideration (Garp!), I'm wondering if this simply doesn't come down to the factor I mentioned earlier, which is that the oil, being the main carrier of capsaicin, does not need to be and almost certainly won't be scaled linearly. It will in fact be reduced to greater degrees as the bulk is increased and so the amount of capsaicin that can be released is inherently restricted.
-
Out of interest how many members discussing this topic have any experience of bulk cooking curry or intend to cook curry in bulk? Say 40 portions +? I
-
i feel the problem is making sure the spices are cooked out, easy when doing a portion at a time, but more difficult with a large mass.
Exactly. I mentioned early on in the discussion that cooking method will change significantly with bulk cooks.
This isn't a single factor problem but I feel we are a good deal more informed than at the start of this thread. Well, I am anyway. ::)
-
As you know LC, I have and will soon be again cooking large portions, hence my interest.
Oil is definitely not increased linearly. That's a given. It's affect on capsaicin transfer and relationship to this anomaly ? unknown.
Spice compaction? A known occurrence but not in the same scale as the apparent magnitude of spice to main curry volume ratio reduction. Plus as Garp has pointed out, within a portion conversion exercise, we would be scaling from the same batch whether compacted or not. It may have some relevance but does not explain away the basis of this hypothesis. I have stated in the Assumptions that the density of chilli is 2.7 g / tsp (5ml) ie; uncompacted.
Having in my calculations above allowed for commonly accepted linear scaling in the first instance the minimum amount to cook side by side would be a 4 serve quantity for side by side taste testing against a single serve dish. This is consistent with some information from the referenced reading. When dealing with only single to double serve there is normal error in measurement and differences may be indiscernible. Although, this is incongruous with the notion and we should really expect to notice a difference.
In practice the difference, if there is any, would become very noticeable at 8 serves.
1 serve - 1 tsp
2 serves - 2 tsp
4 serves - 3 tsp (still measurement error region)
8 serves- 4.5 tsp.
If chilli heat were going to weaken noticeably it would be here.
I think as LC has alluded to above, this really is more so applied to larger volume, however there is no mention of this in the so far referenced material. Can anybody find information that is contrary and simply recommends linear scaling? I haven't looked and won't have time today.
There are 2 people I wish to ask about this. A chef and a spice merchant. I'll get to it asap.
-
Out of interest how many members discussing this topic have any experience of bulk cooking curry or intend to cook curry in bulk? Say 40 portions +? I
-
I received The Spice and Herb Bible Third Edition by Ian Hemphill (Herbies Spices) last week as a birthday gift from my mother-in-law. In the side note on page 86 he simply states that 1 tsp, 5 ml , 2.7 g of chilli (chile) is recommended for 500 g of either red or white meat. This is based on using chilli rated 6 - 10 along his own heat scale being reduced to a 1 -10 scale. Kashmiri chilli being rated at around 7. So far I have found no mention of scaling procedure. I will email him directly to see if has any knowledge in these matters.
-
I will email him directly to see if has any knowledge in these matters.
The absolute best person to email if at all possible would be Heston Blumenthal. His team are actual, well-qualified scientists as well as well qualified chefs.
-
This thread is a fun read, interesting though is when a group of people discuss the subject /use of chillies it still produces heat :), as SS say's the result is the readers being more well informed, cool.
-
This thread is a fun read, interesting though is when a group of people discuss the subject /use of chillies it still produces heat :), as SS say's the result is the readers being more well informed, cool.
But the real question is, "Does the heat increase linearly as the number of participating forum members, or does the same fractional-power law apply as in the case of chilli content w.r.t dish size ?" !
** Phil.
-
I've had 2 contacts today. In person with my Indian chef acquaintance (K) and Mr Ian Hemphill called me directly to discuss the point upon reading my email. It appears there is a little bit right from everybody. No scientific formula but you do not scale in a linear conversion.
Chef K said when asked about increasing 1 tsp of chilli per serve to a dish of 8 or 10 quantities.
No. You don't use 8 or 10 tsp. I was always told to reduce by 20% every time you increase servings.
I clarified to per double quantity. He replied, Yes.
I clarified to reduce by 20% after doubling spices. He said, Yes, that's right.
So Chef K's formula is not X 1.5 but X 1.6. Double it and reduce by 20% equals X 1.6.
When I asked him why he said: It is because the stove cannot burn or cook the heat out of the chilli as much. The cooking is different.
This is directly in line with SS's point about the change in cooking dynamics from bigger quantities. It is also a point raised by Mr H in our phone call.
My email was quite specific in relation to the linear scale or X 1.5 theory and even mentioned the X 1.25 method. During a lengthy chat he said that you do not scale chilli in a linear way, but rather a factor of between 1.5 and 1.8 depending on the type, quality and freshness of the chilli, the amount you are cooking and the size and shape of your cooking vessels. He added that it is also to do with surface area for evaporation, heat input from your particular stove, the style of dish you are making and the other spices and ingredients in the dish (some also being non-linear but probably by a different conversion factor). So scaling a dish is not just a matter of multiplying everything by an integer being the number of serves, or a random figure of 1.5, or the scientifically aesthetic square root of 2.
So this brings me to some sound advice provided here, which is to cook out your spices first in oil after having gone with approximate scaling at reduced quantities using a factor of X 1.5. Add in your tomato and salt etc, followed by pre-cooked meats or veggies and your base gravy. I would be using a thickened form of base gravy and my past experience showed me to reduce it's quantity per serve as well. Simmer. While this is simmering, and to finish the dish if necessary, final adjustments can be made using incremental, BIR style additions taking sauce from the main pot and putting the extra spiced result back in until you have the balance you need. The addition of a tarka or bunjarra at this stage could be beneficial.
This is good advice and the person who gave it can take credit if they wish because it turns out they were pretty much spot on in explaining a good bulk cooking process. Using this method it is best to be under spiced so addition can be made. Difficult to reduce over-spicing.
Note: You can't just throw extra spices in at the end, They needs to be cooked out and added.
-
So this brings me to some sound advice provided here, which is to cook out your spices first in oil after having gone with approximate scaling at reduced quantities using a factor of X 1.5. Add in your tomato and salt etc, followed by pre-cooked meats or veggies and your base gravy. I would be using a thickened form of base gravy and my past experience showed me to reduce it's quantity per serve as well. Simmer. While this is simmering, and to finish the dish if necessary, final adjustments can be made using incremental, BIR style additions taking sauce from the main pot and putting the extra spiced result back in until you have the balance you need. The addition of a tarka or bunjarra at this stage could be beneficial.
This is good advice and the person who gave it can take credit if they wish because it turns out they were pretty much spot on in explaining a good bulk cooking process. Using this method it is best to be under spiced so addition can be made. Difficult to reduce over-spicing.
Note: You can't just throw extra spices in at the end, They needs to be cooked out and added.
;) Ohhhh! Ok I
-
That all makes total sense to me Livo, especially having enough heat to cook out the spices, also your chef threw in a few other points like chilli freshness, surface area of the vessel, points all born from experience. Personally for me i find cooking curries therapeutic and enjoyable, if I'm going to be cooking say four portions of a dish i will do them as a max of two at a time as i would be more in my comfort zone, i do not have the experience like your chef to scale up and have control of all those parameters and would hate to waste the ingredients by gettting it wrong.
-
I want to be very comfortable with the process before I go making 3 big pots of curry, each using 4 or 5 kg of chicken or lamb. The last thing needed is to have AUD$200 in ingredients wasted because they are too spicey to eat. This is crucial to my upcoming event due to the target clientele involved. I will be doing reduced (but still bulk) trials of each dish to test this procedure. I'll use 8 portion tests and eat some myself, give some to family and freeze the remainder.
Big thanks to all for your input. :) I now feel quite comfortable with the job.
-
Chef K's formula is not X 1.5 but X 1.6. Double it and reduce by 20% equals X 1.6.
OK, so let's list the easy ratios first :
1 : 1
2 : 1.6
4 : 2.56
8 : 4.096
So by the time we reach an 8-portion serving, we are adding only half of the spices that we would add for a single portion. I confess that that feels a little spice-light to me. Do you think you might bounce the 8-portion figure off Chef K. when you get the opportunity and ask him if that is what he would expect ?
** Phil.
-
Phil it
-
I will try to see him again.
It appears to be normal practice to not include the first iteration Phil. This doesn't prevent your point from becoming pertinent at the next iteration to 16. With rounding it would be:
1 serve : 1 tsp
2 : 2
4 : 3
8 : 5
This is how I interpreted his answers anyway. There is a slight language barrier issue although his English is quite good. It was quite busy where he was working, and I was in a hurry as well, so I didn't have the time to really press him. He may have meant to only reduce the increment by 20 % or it is even possible that he misunderstood my question about every instance of doubling and just did the math in his head for the amount I asked about. He did think for a second or so before answering.
So that would be either (with rounding)
1 serve : 1 tsp
2 : 2
4 : 3.5
8 : 6.5
or
1 serve : 1 tsp
2 : 2
8 : 6.5
10 : 8
My initial interpretation of his answers though, was consistent with my discussion with Ian, although he did say to use a factor of between 1.5 and 1.8 depending on previously mentioned criteria.
I think it is important to remember LC's recommended method to be used for final adjustments. If I was to make a batch for 32 or 40 serves at a conversion factor of X 1.5 and taste that it was OK, I'd leave it alone. If it was clearly under-spiced I'd use his method of adjustment.
As I've said, this is new ground for me, having only done it once before and I cannot remember exactly what I did last time (4 years ago) although I'm positive I did use the 1.5 reduction factor because of my target consumption group. I now wish I'd kept notes.
Posted at the same time as you LC. I think you are raising the most important point here. Increasing portion quantities is not directly just multiplying anything by a number. Increasing from 1 to 2 serves will be different to increasing from 10 to 20, and this will again be different to increasing from 100 to 200. The chef must know how to make those adjustments.
-
We need to remain aware of an important thing here. It's not possible to just apply a number. If you continue to apply a constant reducing factor in conversion, the point will come when the incremental quantity approaches zero, Your point I think Phil. When talking to Ian yesterday we discussed this exact point and his statement about it was that if you were to take that approach it would be like the Grains of Rice (Wheat) on a Chessboard problem but in reverse.
This is a very good analogy and I don't think anybody is suggesting that a singular magic number is the answer, but the discussions would clearly indicate that direct linear conversion is not the right way to go.
-
It's getting late but I have some findings to post tomorrow along with photos. Short story is that today I turned 2.4 kg of raw chicken into 10 serves of chicken madras. I cooked a double serve for comparison, in the usual BIR method with full spices and then a bulk cook of 8 serves using reduced quantities of nearly everything except the chicken. Oil, Mixed Powder, chilli, base gravy and some other ingredients were all reduced in proportion to the amount of chicken per serve. The double quantity took about 12 minutes to cook and the 8 serves took about 20 minutes. The dishes are so close to being the same that there is little to no chance of deciding which is which, other than the fact that i deliberately kept the bulk cook mild at wife's request so the kids can eat it. I could easily have made the two identical without using the full 8 tsp of chilli. 5 would have been enough.
2 tsp of Kashmiri chilli in 2 serves should have gone out to 4.5 tsp for 8 serves at X1.5 per double conversion. I only used 4 and it is fine. To a person eating either dish the difference is negligible.
It's going to take 2 posts to put up the photos. I'll list the ingredient quantities in the second half.
The first 4 photos show:
1) 2.4 kg of raw chicken chunks pre-cooked down to 2 kg. Boiled for 15 minutes in water, 1 cup Latif's Base Gravy and a heaped TBSP of Curry Powder (Mother's Recipe Madras)
2) Double serving ingredients using CT's basic Chicken Madras video recipe. (400 g of the pre-cooked chicken)
3) Double serving of Chicken Madras cooked BIR style. Now gone. Wife and my dinner. ;D
4) Bulk ingredients for 8 serves, reduced ingredient ratios and the remaining 1600 g of pre-cooked chicken.
-
Pictures cont'd
5) Bulk ingredients all added to the pot and cooking.
6) Some sauce removed for reduction in the aluminium curry pan. Spice adjustment possible here.
7) Chicken removed to prevent over-cooking.
8 ) The 2 and 8 serve dishes side by side. (As stated I deliberately kept the 8 serve tray milder by actually using less spice than calculated. To be honest there was very little difference even after having done this.)
-
The ingredient quantities.
I based the experiment / practice cook on Chewytikka's Chicken Madras video and the only thing different was the addition of a small squirt of Worcestershire Sauce. The quantities were guessed from the video single serve but kept simple to allow for easy scaling to increased quantities.
Double serve. (exactly 2 X single serve). In order of addition to the pan and cooked in normal BIR method as per CT's video.
1 TBSP butter ghee
1 TBSP vegetable oil
1 TBSP garlic and ginger paste
1/2 cup dilute tomato puree
2 tsp mixed powder
2 tsp kashmiri chilli powder
1 tsp kasoori methi
Splash of Base Gravy
2 tsp lemon juice (added a squirt of W' sauce)
1/2 tsp salt
400 g pre-cooked chicken
1/2 TBSP coriander leaf and stem
400 ml Base Gravy (I used Latif's and diluted 3 : 1) I reduction cooked the dish considerably to suite Australian expectation.
8 serve bulk. Bracketed number shows approximate conversion factor per iteration of double. Modified method as described above.
2 TBSP butter ghee ( X 1.5)
2 TBSP vegetable oil (x 1.5)
3 TBSP garlic and ginger paste (X 1.75)
1 1/2 cup dilute tomato puree (X 1.75)
5 tsp mixed powder (X 1.6)
4 1/2 tsp kashmiri chilli powder (X 1.5)
2 tsp kasoori methi (X 1.5)
1/2 cup base gravy
1 1/2 TBSP lemon juice / W' sauce (X 1.5)
1 tsp salt (X 1.5)
1.6 kg pre-cooked chicken (direct linear scaling)
2 TBSP coriander (direct linear but to suite taste)
1 litre base gravy (X 1.6) Note: not diluted and this is in total including the half cup already added. It took a fair bit of reduction to get to the same as the double serve.
Note: As stated I reduced the quantity of Kashmiri chilli powder even further to only 4 tsp (possibly less) and there was minimal difference to the taste or heat level when compared to the full ration ingredient double serve dish.
Conclusion: Non linear scaling is necessary Had I used linear scaling with the 8 serve quantity the spicing levels would have been too high..
-
Now all the debates are concluded ;D Comit big and get yourself practice a 50l pot and earn some money! ;)
-
Good afternoon Livo, my hours are not your normal '9-5' so yeah, up late some nights and out early the following day.
I am new in tonight and going to have to work through the numbers, just had to try to explain tax law (on his variable earnings) to my son and he doesn't recognise or agree that it is not a linear scale based on his earnings as they vary, as do mine, on a weekly basis, he thinks tax should be the same every week irrespective of earnings... scaling debate with different thresholds springs to mind!
MB6 look different (lighting), MB8 look the same. I think you have it, visually, now if you could send me a portion of each, I can taste test it for you. No fee. ;)
-
Now all the debates are concluded ;D Comit big and get yourself practice a 50l pot and earn some money! ;)
Wishful thinking perhaps LC. I'm now convinced having performed the operation. Some may argue my results are tainted. I can however, assure the naysayers that my experiment was legit. The proof will be in the pudding when my son comes home and reports on his lunch. If he complains about it being still too spicy from the batch of 8 serves then I will know for sure that linear is out.
I have no idea where to even start to get rid of 100 serves of curry. ;D
-
Good afternoon Livo, my hours are not your normal '9-5' so yeah, up late some nights and out early the following day.
I am new in tonight and going to have to work through the numbers, just had to try to explain tax law (on his variable earnings) to my son and he doesn't recognise or agree that it is not a linear scale based on his earnings as they vary, as do mine, on a weekly basis, he thinks tax should be the same every week irrespective of earnings... scaling debate with different thresholds springs to mind!
MB6 look different (lighting), MB8 look the same. I think you have it, visually, now if you could send me a portion of each, I can taste test it for you. No fee. ;)
I have a friend who works in hospitality and since he started I rarely see him. He sleeps when I'm awake and works when I'm asleep.
Aha. Marginal tax and bracket creep. I recently had to try to explain to my son, in his first year of employment, that he can't simply make up deductions so he gets all his pre-paid tax back. His response. Everybody else does!
The 2 pans in MB6 are the same sauce but the curry pan is being used to rapidly reduce the sauce from the bigger pot. This is also providing the usual caramelization on the circumference to provide the Maillard reaction flavour. I did this twice, dumping it back into the big pot and repeating. It is at this stage where you could add extra spicing if you choose.
MB8 shows the 2 preparations together. There really was no discernible difference between the 2.
-
Conclusion: Non linear scaling is necessary Had I used linear scaling with the 8 serve quantity the spicing levels would have been too high..
Did you actually do that too? Unless you did you can't draw that conclusion.
-
MB8 shows the 2 preparations together. There really was no discernible difference between the 2.
Based on how many testers?
-
I had anticipated both questions SS and they are valid.
No, I didn't prepare 8 serves with linear scaling as it is too much food to waste in the likely event that all information I have been given is correct. Let's just say then that linear scaling is not necessary and therefore wasteful, (also undesirable for my purpose).
There were 3 testers and there can be no more as the comparison sample is gone. As the cook and being more tolerant of spice, my own judgement may be questioned. Both my wife and son agreed they were the same dish with no discernible difference.
Also keep in mind that I deliberately reduced the spice in the X8 serve dish even further than the X 1.5 conversion factor calculated amount. Probably down to as little as 1.3 - 1.4. This would suggest that the spice levels are easily matched in larger cooking preparations by using considerably less than linear scaling. Are the dishes "exactly" the same? Probably not!. Is it acceptable and very close? Definitely!.
You might repeat this experiment and draw a different conclusion. However, for my needs and purposes, this test has convinced me that I will not be using linear scaling when I bulk cook soon to feed 40 - 50 people curries that need to be mild.
Some people really enjoy spicy heat and willingly use extra hot chilli. For such people linear scaling is most likely quite acceptable, even desirable. I on the other hand, enjoy some spiciness in food but find the extreme heat of chilli overload to be ruinous of otherwise enjoyable food. I don't enjoy the burning sensations and I don't need perspiration dripping from my brow to assure me I'm eating nice food.
I will again reinforce the reason I have performed this test. I need to do bulk preparation of mild curry dishes for a target consumer group. This is a very specific test and inquiry for that purpose. It is not intended to be the be all and end all of this discussion.
I do have 1.5 kg of Chicken Breast remaining. This would be sufficient to do another test comparing a single serve to a batch of 4 - 5 serves using linear scaling. I'll put it in the freezer as I won't have time over the weekend. I still have to finish the Andy2295 Base Gravy.