Sous Vide: Recipes, Techniques & Equipment (Part 6)

[infernooo]

Does anyone have a good guide on tuning the PID parameters on the auber/svmagic units? Auto-tune is OK but gave me some pretty wacky parameters (using this 7qt slow cooker with ceramic insert, heated from all sides except the top: http://www.breville.com.au/products_detail.asp?prod=354)

p.s. I own the WS-1500A, WS1500-B and WS-1500C, and yes I have the manuals/instructions but I am looking for a foodies/EgulleterSousViders guide instead of the manufacturers guide/explanation :-).

Thanks!

Just a little patience, as an addendum to the 1500D user manual http://www.freshmealssolutions.com/ will soon be coming out with a PID-tuning document which is coauthored by three of us EgulleterSousViders. Associated with it will be an Excel sheet to evaluate tuning results, you find a preliminary version here: http://peter-gruber.gmxhome.de/SV/TESTING_PID-SETTINGS.xls . Any suggestions are welcome, I might refine this spreadsheet.

BTW with your Banquet Meal Maker BSC400 which seems to be a rectangular side-heater, I should strongly recommend using forced circulation with an aquarium bubbler (for higher temperatures) or an indoor fountain pump (mine works fine up to 58°C). For better insulation, you might cover the water surface with ping pong balls.

Regards

Pedro

Thank you, I will be eagerly waiting for this guide!

As for the ping pong balls - the slow cooker has a lid, how would these help? I will look into the bubbler... any suggested types/models as an example?

Thanks again!

[PedroG]

The lid helps against evaporation, but is not a very effective insulation. Ping pong balls will add an insulating layer of air.

@ aquarium bubbler: use the air stone as a weight to pull the tube to the bottom of the bath, but cut a sideward hole in the tube just above the air stone; the much larger bubbles will produce less cooling by evaporation than the fine bubbles from the stone.

[infernooo]

Thanks Pedro - much appreciated! I will check out the aquarium shop tomorrow and see how I go.

[PedroG]

Does anyone have a good guide on tuning the PID parameters on the auber/svmagic units? Auto-tune is OK but gave me some pretty wacky parameters (using this 7qt slow cooker with ceramic insert, heated from all sides except the top: http://www.breville.com.au/products_detail.asp?prod=354)

p.s. I own the WS-1500A, WS1500-B and WS-1500C, and yes I have the manuals/instructions but I am looking for a foodies/EgulleterSousViders guide instead of the manufacturers guide/explanation :-).

Thanks!

Just a little patience, as an addendum to the 1500D user manual http://www.freshmealssolutions.com/ will soon be coming out with a PID-tuning document which is coauthored by three of us EgulleterSousViders. Associated with it will be an Excel sheet to evaluate tuning results, you find a preliminary version here: http://peter-gruber.gmxhome.de/SV/TESTING_PID-SETTINGS.xls . Any suggestions are welcome, I might refine this spreadsheet.

BTW with your Banquet Meal Maker BSC400 which seems to be a rectangular side-heater, I should strongly recommend using forced circulation with an aquarium bubbler (for higher temperatures) or an indoor fountain pump (mine works fine up to 58°C). For better insulation, you might cover the water surface with ping pong balls.

Regards

Pedro

Thank you, I will be eagerly waiting for this guide!

As for the ping pong balls - the slow cooker has a lid, how would these help? I will look into the bubbler... any suggested types/models as an example?

Thanks again!

You find Bob Jueneman's "radically condensed version" of the above-mentioned PID-tutorial here: http://chowhound.chow.com/topics/678636#5304760

When googling "PID-Controller" you will find many descriptions and tutorials, e.g. http://en.wikipedia.org/wiki/PID_controller or http://www.learncontrol.com/tutorial/index.html .

Please note that different authors define parameters differently.

The proportional band P is in degrees in SousVideMagic controllers; in 1500A/B/C P=10 is 1.0°, whereas in 1500D P=10 is 10.0° (at least in the pre-production model it was so). Other authors define P as gain, where Proportional Band = 100/Gain.

The integral parameter I may be defined as repeat time in min or sec or as repeats per min or repeats per sec. In all SVMs I is repeat time in seconds; the higher the value, the weaker the action, a zero value means infinitely weak action.

The derivative parameter D is in seconds in all SVMs. The higher the value, the stronger the action, and zero value is zero action. Other authors use gain which is reciprocal.

All clarities cleared?

[KennethT]

SV marrow bone hasn't been much discussed, so I've been experimenting.

First, as a control, I tried the conventional methods of roasting and boiling. Roasting in a hot oven (450F/230C) for 15 mins is associated with a marrow-on-toast dish by Fergus Henderson. It yields slightly browned bone (which looks better sitting on a plate than a beige bone) and very soft marrow, suitable for slathering on toast. But: so much of the delicious fat melts away to waste! Boiling in soup or stew is traditional and it too gives a soft marrow, yet the fat and juices remain to fortify the pot. Tradition has much going for it.

Then, I tried two SV ways. One was relatively hot and quick (175F/80C for 1 hour), the other was low and slow (135F/57C for 8 hours). Neither rendered much fat, though some juices were given off (much as happens when SVing flesh meat). The hot+quick marrow was quite firm, the low+slow was jelly-like -- better texture for my palate.

The bottom line: SV marrow is an alternative to conventional methods, but it is not as clearly superior as SV-cooked flesh is to conventionally-cooked flesh.

Thanks Martin! Very interesting! I wonder if a few more experiments are in order.... maybe 175F for 2, 3, 4 hours, etc to see the effect of more time... Bone is a pretty lousy heat conductor, so I wonder if the heat transfered through in only 1 hour. Also, a quick torching to the outside of the bone is a good solution to the beige coloring...

[MartinH]

Bone does insulate and on a thick marrowbone there's 1/4" or even 1/2" of it for the heat to penetrate. I wonder if anyone knows what the actual thermal conductivity of bone is compared to flesh.

[MartinH]

Or, is bone in fact an insulator? I've heard this said before, but I'm now wondering what is the science of bone conductivity. Fat certainly insulates (hence the blubber layer of sea mammals); but bone perhaps not.

[KennethT]

I would think that bone is definitely an insulator... that's why meat roasted on the bone is always juiciest right by the bone - because it never reach as high of a temperature.

I do chicken legs/thighs on the bone SV all the time - and it always takes a LOT longer to get that spot right next to the bone to be fully cooked, compared with the rest of the meat.

Also, just thinking of the makeup of bone - basically a hard mesh of calcium with lots of air spaces - makes me think it's a good insulator.

I guess the only way to really tell is to put a thermocouple in the middle of the marrow in the bone, and see how long it takes to get to full temp....

Edited January 8, 2010 by KennethT (log)

[tomdarch]

What you're describing sounds more like the bones having high thermal mass, rather than being a good thermal insulator. It's possible for bone to be both. Either would be an issue when SVing.

For SV marrow, are you SVing long pieces of bone, or are you cutting the bones into "disks" to make them easier to bag? It would probably be overkill, but the "best of both worlds" approach that came to mind would be to cut the bones, remove the marrow, SV the marrow, oven roast the bones, and then re-assemble. Probably crazy outside of the French Laundry or Alinea's kitchens...

Edited August 13, 2010 by LindaK (log)

[PedroG]

See

Authorities differ on the thermal conductivity of bone. McGee states that bone conducts more than the meat, Wolke says it less then the meat. The measurements I have been able to find also vary widely, with values for k from half to double those of the meat.

and http://books.google.ch/books?id=iRlnToZbRxEC&pg=PA274&lpg=PA274&dq=thermal+conductivity+bone+meat+fat&source=bl&ots=0nFgEkkMjN&sig=-3jV-JMz2lAmdgfu050tAHKiHP8&hl=de&ei=kbZHS8uZMYvJ-QbHr_1w&sa=X&oi=book_result&ct=result&resnum=1&ved=0CA4Q6AEwAA#v=onepage&q=thermal%20conductivity%20bone%20meat%20fat&f=false

which states the following thermal conductivities in W/m/°C:

Lean meat 0.49, fat 0.21, compact bone 0.56, spongy bone 0.26, bone marrow 0.22

In marrow bones which are mainly from the shaft (diaphysis) of the bone, the major part of the bone is compact bone with thermal conductivity similar to meat; cuts approaching the joint (epiphysis) contain more spongy bone which acts as an insulator. So both McGee and Wolke are right, it depends on what kind of bone is concerned.

[MartinH]

For SV marrow, are you SVing long pieces of bone, or are you cutting the bones into "disks" to make them easier to bag? It would probably be overkill, but the "best of both worlds" approach that came to mind would be to cut the bones, remove the marrow, SV the marrow, oven roast the bones, and then re-assemble. Probably crazy outside of the French Laundry or Alinea's kitchens...

Love the thin disks of marrow idea, sounds just right for a perfectionist chef. But at home, I'm wondering if the spouse will allow a big band saw into the kitchen alongside the SV setup (I used longish pieces about 2"-3").

[infernooo]

Interesting,

I ran auto-tune on my WS-1500B with my slow cooker (see above post) and got these values:

P = 5 I = 154 D = 288

(Keep in mind this is the "precise" controller and it is in Celcius, not Fahrenheit, multiply P by 1.8 to get Fahrenheit values).

The manual suggests the following settings:

Slow Cooker 7 quart: P = 100 I = 700 D = 40

(Multiply P by 1.8 to get Fahrenheit values, so recommended values are P = 180 I = 700 D = 40)

Will run it from cold for 12 hours using a datalogger to see how well it does in terms of minimal overshoot and how steady and accurate it can hold the temperature to my set value (my rare beef favourite temp - 54.9c).

[MartinH]

and http://books.google.ch/books?id=iRlnToZbRxEC&pg=PA274&lpg=PA274&dq=thermal+conductivity+bone+meat+fat&source=bl&ots=0nFgEkkMjN&sig=-3jV-JMz2lAmdgfu050tAHKiHP8&hl=de&ei=kbZHS8uZMYvJ-QbHr_1w&sa=X&oi=book_result&ct=result&resnum=1&ved=0CA4Q6AEwAA#v=onepage&q=thermal%20conductivity%20bone%20meat%20fat&f=false

which states the following thermal conductivities in W/m/°C:

Lean meat 0.49, fat 0.21, compact bone 0.56, spongy bone 0.26, bone marrow 0.22

In marrow bones which are mainly from the shaft (diaphysis) of the bone, the major part of the bone is compact bone with thermal conductivity similar to meat; cuts approaching the joint (epiphysis) contain more spongy bone which acts as an insulator. So both McGee and Wolke are right, it depends on what kind of bone is concerned.

This chimes with my experience that marrow from the middle of the bone was about as done as marrow from near the ends. There must be another reason for the common observation that meat near the bone is more tasty and juicy - perhaps it contains more collagen, and of course is usually furthest from the exterior, or perhaps "thermal conductivity" is just more complex than these bare measures.

[e_monster]

Interesting,

I ran auto-tune on my WS-1500B with my slow cooker (see above post) and got these values:

P = 5 I = 154 D = 288

(Keep in mind this is the "precise" controller and it is in Celcius, not Fahrenheit, multiply P by 1.8 to get Fahrenheit values).

The manual suggests the following settings:

Slow Cooker 7 quart: P = 100 I = 700 D = 40

(Multiply P by 1.8 to get Fahrenheit values, so recommended values are P = 180 I = 700 D = 40)

Will run it from cold for 12 hours using a datalogger to see how well it does in terms of minimal overshoot and how steady and accurate it can hold the temperature to my set value (my rare beef favourite temp - 54.9c).

It is not surprising that the settings vary from what is mentioned in the manual. The manual (at least the one that comes with the Auber Instruments version) makes it clear that the settings represent ones that worked for the particular devices that they used and that you are likely to need different settings if you use different equipment. Devices that heat from the side are especially challenging because the weak convection currents results in irregular latency. Adding a bubbler (you don't even really need the airstone) will help. And you should re-auto-tune once you start using the bubbler because it significantly alters the dynamics.

Other people's settings may be of limited use unless they are using near identical equipment. The settings they give as the default work almost flawlessly on my Presto Multicooker (which only heats from the bottom) but are far from optimal from any of the 'side-heating' devices that I have tried.

[infernooo]

FWIW my slow cooker heats from all sides INCLUDING the bottom, so perhaps convection is not as big an issue as with most peoples slow cookers... might even be better than a rice cooker?

[MartinH]

Sous vide and hash: they were meant for each other!

True, sous vide is advanced cooking, while hash is about as primitive as it gets. But the two complement each other perfectly. Sous vide gives perfectly done meat. Yet, SV meat isn't browned, and it doesn't have any unctuous, mouth-coating, liquid fat. Make a hash a couple of days later, and you can get everything. I've found that the old-fashioned hash method needs to be slightly modified to yield a superior SV-hash. First, I prefer to avoid canola or other seed oils and to use instead a tasty grease, like butter or duck fat when I have it. Then, instead of throwing everything into the pan together, I briefly brown the meat chunks alone over high heat then remove them from the pan. Keeping it brief is of course important to maintain the rareness of the meat. Doing it in chunks is better, I've found, than browning one big slab of meat since the chunks have more edges, and edges tend to brown most quickly. Finally, into the pan go the vegetables. I keep a supply of blanched onion in the fridge because I think blanching takes away the harshness of onion. I tend to use little cooked potato and substitute cauliflower, which browns nicely. Some cooked cabbage, a few capers, roasted red peppers, all go well.

Anyone else an afficionado of SV-hash?

[PedroG]

Does anyone have a good guide on tuning the PID parameters on the auber/svmagic units? Auto-tune is OK but gave me some pretty wacky parameters (using this 7qt slow cooker with ceramic insert, heated from all sides except the top: http://www.breville.com.au/products_detail.asp?prod=354)

p.s. I own the WS-1500A, WS1500-B and WS-1500C, and yes I have the manuals/instructions but I am looking for a foodies/EgulleterSousViders guide instead of the manufacturers guide/explanation :-).

Thanks!

Just a little patience, as an addendum to the 1500D user manual http://www.freshmealssolutions.com/ will soon be coming out with a PID-tuning document which is coauthored by three of us EgulleterSousViders. Associated with it will be an Excel sheet to evaluate tuning results, you find a preliminary version here: http://peter-gruber.gmxhome.de/SV/TESTING_PID-SETTINGS.xls . Any suggestions are welcome, I might refine this spreadsheet.

BTW with your Banquet Meal Maker BSC400 which seems to be a rectangular side-heater, I should strongly recommend using forced circulation with an aquarium bubbler (for higher temperatures) or an indoor fountain pump (mine works fine up to 58°C). For better insulation, you might cover the water surface with ping pong balls.

Regards

Pedro

You find Bob Jueneman's "radically condensed version" of the above-mentioned PID-tutorial here: http://chowhound.chow.com/topics/678636#5304760

When googling "PID-Controller" you will find many descriptions and tutorials, e.g. http://en.wikipedia.org/wiki/PID_controller or http://www.learncontrol.com/tutorial/index.html .

Please note that different authors define parameters differently.

The proportional band P is in degrees in SousVideMagic controllers; in 1500A/B/C P=10 is 1.0°, whereas in 1500D P=10 is 10.0° (at least in the pre-production model it was so). Other authors define P as gain, where Proportional Band = 100/Gain.

The integral parameter I may be defined as repeat time in min or sec or as repeats per min or repeats per sec. In all SVMs I is repeat time in seconds; the higher the value, the weaker the action, a zero value means infinitely weak action.

The derivative parameter D is in seconds in all SVMs. The higher the value, the stronger the action, and zero value is zero action. Other authors use gain which is reciprocal.

All clarities cleared?

To download my spreadsheet "TESTING_PID-SETTINGS.xls" please use the following link: http://peter-gruber.gmxhome.de/SV/TESTING_PID-SETTINGS.htm which will link you to the actual version of the file. The filename will now contain a version number.

Pedro

[howsmatt]

Could I use my SVM to control let's say 2-4 of the same rice cookers (same water etc.) if I plugged a power bar into it and had my rice cookers plugged into that? Obviously the cookers that aren't directly connected may not be perfect. Or would this cause my SVM to explode?

[Paul Kierstead]

Wel, I received my Sous-vide magic on Friday and so far have made poached eggs, steak and fish. I have to say, I am enjoying the experience (and food) and really love the added simplicity to meal preparation. The steak was fantastic, the fish was fantastic (and much stronger flavoured then expected, considering we used Haddock). The eggs were .... different. The yolk was lovely, but the custardy white might take a little getting used to.

I'm now looking forward to lots more. Books remain an 'interesting area'; I am a big fan of cooking from Keller, but a survey of his Under Pressure book shows that I would have great difficultly in sourcing the primary ingredient in probably a large majority of the recipes. Time to start looking around. So far I am working off the the very excellent work so kindly provided to the community by Douglas Baldwin.

[e_monster]

Could I use my SVM to control let's say 2-4 of the same rice cookers (same water etc.) if I plugged a power bar into it and had my rice cookers plugged into that? Obviously the cookers that aren't directly connected may not be perfect. Or would this cause my SVM to explode?

That wouldn't work. You need one temperature controller per cooking device. Even with the same brand/model of cookers the response is going to be different enough that the temps would not be reliable -- in complex systems like this small differences in input can result in large differences in output. Not to mention that the power draw would be too much -- even if power draw weren't an issue -- it wouldn't work very well (if at all).

[jk1002]

>>I would have great difficultly in sourcing the primary ingredient in probably a large majority of the recipes. Time to start looking around<<

That is true, one can simplify here and there but in general the style of food in that book is part of a multi course menu, I don't think that can easily be scaled up to become a full entry for lets say a family meal. Sometimes for example a "starch" component is missing etc. Over a multi course dinner that is no big deal, if you spend 6 hours or more to assemble that single plate it is a different story.

What I have been doing is, I take the recipes I usually cook but toss the protein into the bath. For example I took the Pork Loin Spiedino out of the A16 book, SVed the loin and just finished off in the end for a few seconds under the broiler.

Recipe is here:

http://www.seriouseats.com/recipes/2008/09/cook-the-book-pork-loin-spiedino-with-pine-nu.html

For pork, chicken breast and tougher steak cuts SV is my goto technique. I haven't yet tried vegetables or fish. Lobster was a failure and even though I know where I goofed I would probably stick to my old method of parboiling and broiling.

The connection one has to make is, how do you merge the SVed product with the original recipe. To me it seems that often this is not such a big problem.

[PedroG]

Interesting,

I ran auto-tune on my WS-1500B with my slow cooker (see above post) and got these values:

P = 5 I = 154 D = 288

(Keep in mind this is the "precise" controller and it is in Celcius, not Fahrenheit, multiply P by 1.8 to get Fahrenheit values).

The manual suggests the following settings:

Slow Cooker 7 quart: P = 100 I = 700 D = 40

(Multiply P by 1.8 to get Fahrenheit values, so recommended values are P = 180 I = 700 D = 40)

Will run it from cold for 12 hours using a datalogger to see how well it does in terms of minimal overshoot and how steady and accurate it can hold the temperature to my set value (my rare beef favourite temp - 54.9c).

It is not surprising that the settings vary from what is mentioned in the manual. The manual (at least the one that comes with the Auber Instruments version) makes it clear that the settings represent ones that worked for the particular devices that they used and that you are likely to need different settings if you use different equipment. Devices that heat from the side are especially challenging because the weak convection currents results in irregular latency. Adding a bubbler (you don't even really need the airstone) will help. And you should re-auto-tune once you start using the bubbler because it significantly alters the dynamics.

Other people's settings may be of limited use unless they are using near identical equipment. The settings they give as the default work almost flawlessly on my Presto Multicooker (which only heats from the bottom) but are far from optimal from any of the 'side-heating' devices that I have tried.

SousVideMagic 1500A and 1500C do autotuning with a hysteresis of ±0.5° (C or F), whereas the 1500B does autotuning with a hysteresis of only ±0.1°C, which in my experience gives less reliable results. Try autotuning with the 1500C in Celsius mode and use these values on the 1500B!

[PedroG]

Does anyone have a good guide on tuning the PID parameters on the auber/svmagic units? Auto-tune is OK but gave me some pretty wacky parameters (using this 7qt slow cooker with ceramic insert, heated from all sides except the top: http://www.breville.com.au/products_detail.asp?prod=354)

p.s. I own the WS-1500A, WS1500-B and WS-1500C, and yes I have the manuals/instructions but I am looking for a foodies/EgulleterSousViders guide instead of the manufacturers guide/explanation :-).

Thanks!

Just a little patience, as an addendum to the 1500D user manual http://www.freshmealssolutions.com/ will soon be coming out with a PID-tuning document which is coauthored by three of us EgulleterSousViders. Associated with it will be an Excel sheet to evaluate tuning results, you find a preliminary version here: http://peter-gruber.gmxhome.de/SV/TESTING_PID-SETTINGS.htm . Any suggestions are welcome, I might refine this spreadsheet.

Regards

Pedro

Thank you, I will be eagerly waiting for this guide!

The "Guide to PID Control for Sous Vide Cookery" is now on-line at http://www.freshmealssolutions.com/downloads/PID1500Dtuningguiderelease1.pdf

Pedro

[jk1002]

I found by accident few SV demonstration movies with Daniel Humm of Eleven Madison Park in NYC. These are also on the Eleven Madison Park Homepage

Presentation

Consistency

Texture

Flavour

[Robert Jueneman]

Just to come back to the issue of pan-searing vs. torch-searing meat.

In the past, I used to use grape seed oil or rice bran oil for high temperature pan searing. Then we moved (to south-central Colorado), and in the process I had to have someone come in to clean the apartment. It took several hours to wash the kitchen walls. In the meantime, the movers had packed all of the pots and pans and the stainless steel racks, and it wasn't until I unloaded them that I realized how filthy greasy they all were. I must have done 20 loads in the dishwasher in the first two days! Blame it on an inadequate over-the-stove range hood, I suppose, but the grease from pan-searing spread throughout most of the house. Yuck!

Anyway, I don't do that anymore.

Instead, I use a Le Creuset ribbed grill pan, and put it on the stove at a medium high heat without any oil or butter (and certainly no Pam). I'm sure a well-seasoned lodge pan would work equally well, but the Le Creuset is easy to clean, and didn't require seasoning.

I carefully dry the steak (or chicken, or fish) with a paper towel, then fire up my Iwatami butane torch (Douglas Baldwin's recombination, available at Williams-Sonama). It uses the same kind of butane cylinder that is used for table-top cookers, and it puts out a flame that even my plumber would be impressed by. (Way better than my dinky Creme Brulee torch, which isn't even very good for that. No, I haven't tried oxyacetylene!)

Flip it over, and do the other side, and then rotate it 90 degrees to cross-hatch the meat if you are really, really finicky.

In comparison to using smoking hot oil, the torch method seems to sear the little bumps on the surface of the meat, more than searing everything, but maybe I don't sear it long enough. The searing is mostly for looks, and a bit for the maillard reaction, but I don't want too much crust, and I don't want to overcook the meat.

Once it warms up a bit (in the winter, this area of Colorado drops about 30-40 degrees after sunset), I might try searing on the outdoor gas grill with a very hot fire, or even over charcoal, but that degree of preparation takes away from the simplicity and ease of of cooking a steak SV. Now, if I had a proper salamander and a professional range hood, or even a fireplace, that might be a different matter. Sigh...maybe when we finally build my dream kitchen/house.