adrianvm

The text you linked to is kind of ambiguous, but if you click through and pick a specific pan, for example https://www.copperpans.com/fusion-frying-pan-24-cm-94-in then you get this explicit breakdown: Substantial 1.4 mm copper content for even heat distribution Pan lining of 0.6 mm stainless steel for practical durability Induction ring on the bottom that is ferritic SS and 0.5 mm thick Surely if there was more copper they'd brag about it. It does indeed appear that the bimetal part is 2.0 mm, of which 0.6mm is stainless. An added 0.5mm of stainless on the bottom brings the total thickness to 2.5mm. A lot more steel than the Coeur line.

I'm not seeing a pan like you describe. They've introduced the "Fusion" line with a disk base, total thickness of 2mm comprising 0.6mm stainless on the inside, what they refer to as a "substantial" 1.4mm copper layer, and then a 0.5mm layer for induction compatibility on the outside.

Why wouldn't I want the answer? I like answers. (I might not want to use the proposed method, but that's another matter.) I did not attempt to measure temperature of the (oil in the) pans for this test. But my previous observations with heating cast iron were: "Tried measuring a very large cast iron skillet with the Control Freak set to 250 on slow intensity. After about 20 minutes it is in steady state with the center at 260, a point 2" off center is 260, a point 4" off center is 230, and 5" the temperature is 200. It took quite a while to reach steady state and overshot by a lot (up to 290) on the "slow" speed. " If it overshoots by a lot that gives it a browning advantage. It also means that if I preheat longer I'm not waiting for heat to conduct vertically to the pan top but (hopefully) sideways. Really the pan is 5mm thick, whereas width is 300 mm. I don't know what the maximum distance from a coil is, but much more than 5mm. Of course maximal evenness, by which we mean no temperature variation across the pan surface, requires a uniform heat source. But what about good-enough or practical evenness? The aluminum base pan browned evenly enough that I noticed no difference in browning across the entire pan with regards to how the meatballs browned. That's good enough. My previous testing showed about a 15 deg temperature drop from center to edge, which is pretty even. For cast iron we have a 50 deg drop. I don't know what the drop looks like when you heat to 400 instead of 250, but clearly with cast iron it was enough to affect browning performance noticeably and with thick aluminum it was not.

Uh, yes, I did a simple test of browning where I browned some stuff in pans under the normal use pattern likely to occur if I actually cook stuff, and observed how things browned, namely evenly and well in the aluminum base pan, likewise, though slower, in the copper pan, and poorly and unevenly in the cast iron. Why should I think the pans aren't "fully preheated". As shown by thermocouple measurements, the Coeur pan overshoots the control freak target temperature generally, and it's smaller, so why wasn't it "fully preheated"? It doesn't take long to heat that pan and if anything I expect it was hotter than the target. The predicted behavior of the Coeur pan was for it to deliver the worst browning performance, because it has the least thermal mass of the three pans; it did better than predicted. The cast iron I heated on low and I didn't time how long it took, but something between 10 and 15 minutes. I think that's longer than I generally have preheated it when using it on gas. The cast iron did not grow more even over time in my previous observations of pan temperature. It seems like preheating in the oven is the only way to "fully preheat" cast iron, but that's no longer stovetop cooking and not something I'm likely to do just in order to fry something, so it has no practical relevance. If you have the flexibility to "choose the cooking mode that works best" that's great. But most people have just ONE range, not three, so they have to work with the cooking mode that is available. This thread is about choosing pans for induction, not about choosing your stove to fit your pans. But if I were to use the pan on gas, is there a way to make cast iron deliver superior performance using only the stovetop? Certainly I have the same uneven heating issues with the large cast iron pan on the gas heat source.

Usually people have the cooking technology they have and have to choose pans to use with it. This thread is about what pans to pick for induction. But my observation isn't just about EVEN browning but about browning at all. The cast iron, even in the center, did not brown as well. Is there some reason I'd expect to get different results if I repeated the test with gas heat?

I did a simple test of browning where I made the meatball recipe that previously gave me browning trouble. It has 2 lbs ground meat and 2 cups of water, so the high moisture content makes browning a challenge. Previously it would not brown well in the Proline. I tested it on the Control Freak where I preheated to 400 using the Breville Thermal Pro 12.5" skillet, a Le Creuset cast iron 12 inch pan, and a Falk Copper Coeur 9.4" pan. The Breville browned very effectively and rapidly and produced a good crust on the meatballs. The cast iron dropped in temperature and stayed cold and while it did brown, it took much longer, and the crust was thinner. Furthermore, it didn't cook well around the outside part of the pan whereas the Breville cooked well over its whole surface. The Falk pan also seemed to lose temperature rapidly and it took longer to brown, but it did recover and produced a nice crust eventually. (I realized belatedly that I should have tested the proline, not the falk, but too late to change that.) Based on my observations I don't see that cast iron is a good choice for induction.

Yes, the water test is a fair measure of the heat stored by the whole pan but I actually care about browning food. Is heat storage the full story for browning power? Is heat in the pan side walls available to brown food that is only touching the pan bottom? Actually another thing occurred to me with regards to the water test. Browning requires that the temperature be high but transferring heat to water does not. Maybe this difference isn't fundamental, but it seems unlikely that in induction stovetop cooking you can get the pan sides over 275 so that they can reheat the pan center after it cools. The fissler pan I have with the cookstar base is I believe 6mm thick, so it's thicker than proline, which is 4.7mm.

Wouldn't the water test be unfair for disk based pans because the side walls are thin? Cast iron is thicker than proline, but only by 0.3mm, and yet when I cooked wet meatballs only the cast iron could brown them, not proline. Perhaps I need to do a wet meatball test.

I tested a couple more pans for temperature uniformity using the Control Freak induction cooker. One is a Paderno Grand Gourmet 28cm (11 inch) fry pan. This pan has a thick aluminum base which the manufacturer told me is 6mm of aluminum. (Apparently in the past it was 7mm.) It has 1mm total of stainless steel in the base. The base is a bit over 8.75" diameter. I measured the thickness of the base and found it to be 7mm total. The side of the aluminum on this disk is exposed for those that worry about that sort of thing. Also this is obviously not flat: it rocks quite a bit on its base, so not the best quality either in design or manufacture. In my testing with the Control Freak set to 250 on low intensity, the pan heated to 259 in the center and stayed at that temperature. At 2" from the center it was 257 and at the edge it was 242. (I didn't test the side wall.) The other pan I tested is a Breville Thermal Pro frying pan. The pan is 12.5" diameter. This pan has has tri-ply clad sides with an aluminum base bonded to the bottom. The aluminum disk is covered by stainless steel and its bottom is about 8.5" diameter. I couldn't find advertised thickness data. I found the total thickness of the pan bottom to be 7.3 mm (presumably including 3 layers of staniless and 2 layers of aluminum). Pan sides are quite thin, maybe 2mm. This line is discontinued and the pans are cheap on amazon at the moment. The base of this pan is quite significantly concave. This pan heated to 254 in the center, 258 at 2" and 248 near the edge. The sidewall temperature was 215. One thing I've been wondering about is what makes cast iron good and can a pan with better heating uniformity deliver the same benefits. It seems like people are excited by the searing power of cast iron. I've seen cast iron work better for browning wet meatballs than my demeyere proline skillet. (They browned in cast iron; they did not brown in demeyere.) So what makes cast iron work better? Presumably it's the thermal mass. So what does that actually mean? It seems that the number of most interest is the volumetric heat capacity. Cast iron 3.6 J/K cm^3 Aluminum 2.4 J/K cm^3 Copper 3.4 J/K cm^3 Stainless 3.8 J/K cm^3 (depending on alloy) My cast iron is 5mm thick, so its total areal heat capacity would be 5*3.6=18 J/K cm^2. Falk Coeur has 1.9mm copper and 0.6 mm stainless, so 1.9*3.4+0.6*3.8 = 8.74 J/K cm^2. So I predict that the Falk holds a lot less heat than the cast iron. What about Demeyere proline? Reportedly it is 3.7mm aluminum a 1mm stainless. So that gives 12.68 J/K cm^2. So better than falk but not as good as cast iron. I'll pause here to note the weight issue. Of these materials, aluminum actually has by far the best heat capacity by mass at 0.9 J/g K compared to 0.46 J/g K for cast iron and 0.39 J/g K for copper. This means we can get much more heat into less weight if we use aluminum. That is, to match the 18 J/K cm^2 of 5mm of cast iron we need only 7.5mm of aluminum, and the weight of the aluminum is 50% that if the iron. So that would suggest you could get the same searing capability as cast iron from a much lighter weight pan. Considering Paderno, at 6mm aluminum and 1mm stainless it's areal heat capacity is 18.2 J/K cm^2, which suggests it should perform similarly to cast iron. I don't have data on composition for the Breville, but it's base is 0.3mm thicker and it seems likely it has more stainless steel, so its score will be similarly high. I haven't managed to find a way to clearly test searing or heat holding power of the pans, however, to make a comparison between the cast iron and the aluminum disk pans. Does anybody have any thoughts on a way to test this? And have I overlooked anything? Aluminum has higher thermal diffusivity which suggests heat should move out of it into food faster than from cast iron. That would be good, right? By the numbers, copper seems to offer remarkable little gain for all the hype associated with it as the best pan material. Its thermal diffusivity is 111 mm^2/s compared to 97 mm^2/s for aluminum. Seems like barely a difference and Regarding other notable pan options, there is Volrath Centurion which has a 5mm core of aluminum and 1.9mm total of stainless steel (19.22 J/K cm^3). There is (maybe) Cristel Casteline (which has vanished from the Cristel web site) which has (had?) a similar construction to the discontinued Breville line but a very high price tag and an interchangeable handle system whose usability isn't clear. Scanpan Impact has a total base thickness of 6.4 mm.

Sidewall heat on induction is going to be negligible no matter the pan, is it not? When I tested my Demeyere Proline skillet the temp was 205 on the sidewall, and that's a pan with a conductive sidewall. If I use a clad pan, which seems to mean higher Delta T, then food that's anywhere close to the edge---not just the food that's touching the edge---will receive a lot less heat than foods at the center. That's not an improvement. It's a bit unclear to me what the benefits of sidewall heating are. Cast iron seems to be better at browning food than anything else. I recall a situation where I cooked half the meatballs in the Demeyere Proline and half in a giant cast iron skillet and the ones in the cast iron browned while the ones in the Demeyere did not. And the Demeyere was on the more powerful burner. They were very wet meatballs that were hard to brown. So what is the explanation for cast iron's ability to brown? It seems like aluminum has a 5x higher thermal diffusivity I would think would enable it to transfer heat to the food faster. Aluminum also has double the heat capacity of cast iron, so it holds twice as much heat per mass. So what's the magic with cast iron? Can I get the same browning with something that heats more evenly? Is it simply about the mass of the pan?

I tested the Fissler pan again and this time measured the center, the inside corner, and a spot an inch up the sidewall on the inside. I heated the pan on slow to 250 as before and the center temperature came to a steady state around 265 F. The corner of the pan was 250 F. That was pretty surprising. The temperature up the side wall was 201 F. I got to wondering whether that sidewall temperature was actually measuring convective heating from the inside bottom surface, so I moved the thermocouple to the outside of the pan side and there I got the reading of 168 F. None of this seems particularly surprising. There's no indication that heat is "accumulating" at the corners or in the sidewall as suggested above. I also tried another caramelization test, this time with the Freak set at 375 F, where I let it come up to temperature and then sprinkled sugar into the pan. The fissler gave a very nice result with a very even brown across its surface. The demeyere dutch oven (which is the same diameter) left a ring of white sugar around its edge, showing that it is definitely not heating as evenly as the fissler. Another bit of information: I finally got Cristel to give me a number for the thickness of their disk based pans. They report that the pans have 4mm of aluminum in the disk bases in the Mutin and Strate lines. My understanding is that Fissler and Paderno have more aluminum, so the Cristel may outperform clad pans in terms of uniformity, since they typically have 2.3mm of aluminum or less, but it won't be as uniform as Fissler or Paderno.

You're saying a better algorithm would adaptively adjust the PID parameters to decrease overshoot in the center? It's not obvious that this is always the desired behavior. Can you do this without making it slower? And will it adjust quickly enough when you add food and the thermal characteristics of the pan/food system change? The extra complexity could make it harder to use the system.

Actually that's incorrect. In order for it to automatically adjust power based on how the pan is reacting it needs to know how the pan is reacting. It does not know this. I have just done a bunch of temperature testing (which you can see in the thread about induction friendly cookware) and have seen that the pan temperature can be 70 F higher at a point away from the center compared to the temperature reported at the CF probe. You only know how much overshoot is occurring at the center point, but heating is away from the center, so the temperatures can be significantly different, and the overshoot can be much larger than you might think. I cooked in cast iron this morning and butter was threatening to burn in a 6" ring while the pan reported 350 F and outside that ring the butter hadn't even started to brown. It would appear that heating this pan evenly (if even possible) requires a slower speed than "low" but there's no way the CF can tell because at the center it was measuring 350 F.

I got a message back from Chefsteps support claiming that Demeyere cookware is not compatible with the Control Freak because it draws too much current and the Control Freak therefore thinks the pan is not compatible. This doesn't really make sense to me, as I've never gotten the "no pan" message when using the two Demeyere pans I have. Has anybody had issues using Demeyere with the Control Freak? Or, can anybody report that they use Demeyere without obvious problems (and if so, is it the Atlantis or proline with their special induction layer, or one of the lower lines without it)? I'm trying to decide what followup questions to ask them about this.

Do ordinary induction cookers have a PID controller? If so, what are they controlling? I suppose induction systems probably do use cycling for at least some of the different settings, which is a time nonuniformity. Whether that's significant would depend on the cycling frequency and time constant of the pan. Discreteness of settings on induction has nothing to do with time uniformity.

I definitely do not understand why gas as a heat source would be more uniform over time than induction. The control freak obviously varies its output because it's running a control loop, but a normal induction system should be producing constant power output at a given setting. What source of time variability exists? Temperature dependence of electromagnetic properties of the cookware?

Yeah, in a theoretical sense, thermal equilibrium will exist---when the pan is at the temperature and distribution where its loss to the environment equals the input from the heat source. But in a practical sense, if the equilibrium occurs at 700 F that's not a useful, practical equilibrium point. I'm not sure what you mean about the output of gas being "more constant". More constant over time? Over space? It seems like heat output from gas is pretty concentrated in a ring. The difference is that we have convection to distribute the heat. But maybe you count that as part of the gas system's heat output.

If you aren't using a control freak then thermal equilibrium may not exist, or may not be a reasonable place to be, for an empty pan. I noticed that the control freak has "low" marked at 130 F. I tried setting a pan on my gas stove on low and monitored the temperature, which continued to rise until I discontinued the test, at which point the pan was I think somewhere between 200 F and 300 F. Not equilibrium yet. This makes me wonder if I can get a pan in equilibrium on the gas stove. If I recall correctly on centurylife he tests by heating the pan until the center reaches a target temperature. It seems like this might favor thicker pans that get more time and clearly it's not an equilibrium measurement. I don't know what affects how much heat the pan loses to the environment, which would determine where the equilibrium is in an open loop system where you are setting the input power. I agree that you can think of a clad pan as a large disk. But you didn't explain how this helps us to understand pan heating. On the Demeyere proline, the 5" point was only at 205, so heat is not all rushing out to the periphery of the disk.

You think the pressure cooker measurements are anomalous? They seem like the least strange data set of all the ones I've collected. Maybe I should try the caramelization test with that pan. Are you saying that if I compare two pans that both have 3mm of aluminum but one has it as a disk and one is fully clad the clad pan will be more responsive? It seems like it might be easier to cool down a clad pan because you would radiate more heat out the sidewalls, but also harder to heat it up for the same reason. I don't know what the time scale is for that effect, though. In my other tests it doesn't appear that heat is rushing to the outside edges of pans across the bottom, so I don't know why it would do it up sidewalls. Your suggestion that heat "accumulates" suggests that we should see a higher temperature at the corner of the pan, which is not what I'd expect. I was curious about whether the thick disk construction would result in a more even temperature across the pan surface compared to thinner clad and disk construction. It seemed like it did do that---so a thick disk may be better for uniformity. Clearly a thick disk is going to sacrifice responsiveness, so it depends on the relative importance of these two metrics.

Why would I want to measure the sidewall? I don't understand. Presumably it will be much lower. Do you think the temperature discontinuity at the sidewall transition would affect the temperature at the 4" radius mark, where the temperature was 247? Note that my primary motivation for testing the pressure cooker was that it's a representative of a pan with a very thick base that goes all the way to the edge of the pan. It showed more uniformity than Demeyere Atlantis.

When you say there is only one coil you mean the entire coil---both parts on both sides of the gap---just always fires up? I realized I had a compatible pan I hadn't tested---a Fissler pressure cooker with a very thick conductive disk base. So I repeated my temperature test with the thermocouples. I set the Freak to 250 F as before and after give it quite a while to settle (on low intensity) I got readings of 260 F at the center, 260 F at 2" off center and 246 F at 4" off center. I waited several minutes more and got 254 F center, 254 F at 2", and 247 F at 4". So that is the most uniform measurement I've seen. I do again have the readings higher than the temperature reported by the Control Freak, which is a bit strange. I wonder if hot air could be staying in the pan and heating the pan top surface. (It's not clear why it wouldn't just rise out of the pan, though.) If I blow into the pan for a moment the measured temperatures drop closer to 250 F. This pan has tall sides.

I'm not sure what you mean by "these are hot plates after all". Like therefore they cut corners? I measured power while heating at it was around 1680W. Once I got to a boil and turned the temperature down to 225 it bounced between 1300W and 1500W. It crossed my mind that if the coils were given equal power then the outer coil, which covers a larger area, would be delivering less power per area to the pan than the smaller inner coil. You didn't answer the question of whether there's any reasonable test that could reveal if the outer coil is energized.

So it's actually possible that I put a 12.5" skillet on the cook top and it only energizes the inner ring? That seems like a pretty bad idiosyncracy. Tried a Demeyere Atlantis dutch oven that is 9.5" diameter and it also seems like visible boiling is confined to a 6" ring. I mean, could something be wrong with my unit? Is there any way I can tell if the outer ring is energized?

I tried Demeyere proline with the boiling water test and also got a small ring. I don't understand why I don't at least get two rings given the double ring design of the Control Freak. Does the induced field really fall off as 1/sqrt(r)? That's much slower fall-off than I was expecting.

Fair enough. Is there something wrong with the boiling water test used in that video? I tried it on a couple pans. On the Heston nonstick I got a totally even bubbling surface of water. On the Falk I got a ring of bubbles, which the video would argue is inferior.