Heat pipes: space age tech to halve roasting times

[Rho]

 

The space race trickled into kitchens in the 60s and 70s, including one curious tool that's faded away in the years since: the thermal pin, a heat pipe skewer that can halve cooking times for roasts:

 

Heat pipes are thermal superconductors, transferring heat 500-1000 times more effectively than solid copper (some people in the sous vide thread have discussed copper pins). They're hollow tubes with the air evacuated and a small amount of working fluid, often water. The usable temperature range is limited by the triple point and the critical point, with additional constraints near the edges. Water is effective from 20C-280C /70F-530F, which comfortably spans most cooking temperatures.

 

Modernist Bread has an excellent section on how bread bakes, including a diagram of the internal heat pipes that develop, summarized here. (click for a good photo!)

Quote

“A Harvard professor once asked me, ‘If I’ve got a roast about this size’ ” – Myhrvold holds up his hands about half a metre apart – “ ‘it takes me a few hours to cook. If I’m cooking the same amount of bread, it’s done in 20–30 minutes. Why?’ It turns out it has to do with heat transport.” As Myhrvold explains, while heat flows into a roast by conduction, in bread it flows more because the bursting bubbles open “heat pipes” through which heat can move by convection. The two have different scaling laws.

 

Sous-vide solves the overcooking side of the gradient problem, but it's still limited by total heat diffusion time-- doubling the size of a cut quadruples the time needed for the center to reach temperature. Heat pipe pins should make larger cuts practical, or normal cuts cook faster. Here's a graph from "The heat pipe and its potential for enhancing the cooking and cooling of meat joints", showing average temperatures over time for 1kg joints of meat convection baked at 190C/375F for 110 minutes (foil removed for the last 30 minutes):

 

Thermal pins were sold commercially from 1956 to about 1990. They're listed occasionally for about $20 on ebay. They even made potato baking racks with heat pipes-- though now you can easily par-cook a potato in the microwave and finish it in the oven.

 

I don't know why production of thermal pins stopped, or what fundamental problems limited their usage. It seems like pans and commercial griddles would be improved by adding heat pipes to spread heat throughout and avoid hot or cold spots. Perhaps roasts fell out of favor as the culture of entertaining shifted away from monolithic centerpieces to smaller, more precisely cooked portions.

[chromedome]

I guess the ubiquitous "potato nail" for baked potatoes is a small-scale version of the same. 

[btbyrd]

Reminds me of Dave Arnold's Bionic Turkey.

[Rho]

9 hours ago, chromedome said:

I guess the ubiquitous "potato nail" for baked potatoes is a small-scale version of the same. 

Somewhat, though note that potato nails -- even the aluminum ones -- conduct heat >100x slower than a heat pipe. In a heat pipe, the liquid vaporizes at the hot end, travels through the tube (by convection/density), and condenses at the cold end, depositing its thermal energy. This moves heat inwards faster than metallic conduction (atoms colliding).

 

 

[dcarch]

In theory the heat pipe sounds great, but in reality no.

 

I have been playing with heat pipes for cooking for many years and have found that there are issues.

The heat collecting end is too small to collect heat, and the contact surface between the pipe and the meat is also not big enough to be of any significance. 

The biggest problem is without the heat pipe, you always insert your thermometer into the center to check temperature,  but if you cook with the heat pipe, where do you insert the thermometer to make sure that you have reached safe temperature?

 

Heat pipes have been in use for many years, in practically every laptop computer to evacuate CPU heat. They are very inexpensive.

 

dcarch

 

 

Edited January 10, 2021 by dcarch (log)

[Daniel-J]

Maybe it's really useless, but I didn't know that way before. It's interesting, maybe if you work it out right, something awesome will come out of it.

[AlaMoi]

get an aluminum potato nail.

hold onto the pointy end, hold the other end over a gas cooktop flame.

start counting.

 

boil a pot of water.

hold onto the pointy end, stick 2-3 inches of the other end in the boiling water (212' vs oven at 350-500'....)

start counting.

 

then decide if a potato nail will move heat fast enough.

 

 

[DocD]

I think the true heat-pipe thermal pins mentioned by the OP (Rho) were ahead of their time and therefore mostly unappreciated by their target audience and their copycat competitors with the aluminum nails, plus the results were a little strange too - your medium-rare roast would have the more-done gray on both the outside AND in the center around the pin, with a donut of pink between those; slices would certainly look a bit "space age" IE Martian! That does answer one question above - the meat thermometer should ALWAY be put in the slowest-heating part of the cooked item; for normal roasting, that would be the center, but with the pin it goes halfway between pin and surface (the instructions enclosed with the pin explicitly state this). 

 

The aluminum potato nails may have also been part of the thermal-pins downfall because to the unaware eye, they look similar - but function SO much more poorly as Rho said. The aluminum can get just as hot, as AlaMoi mentions - the difference is in the total heat flux (NOT temperature), the aluminum carries only 1% of the heat energy, is slower, and looks to be of little use. For a potato, this isn't a huge deal. For a 25# raw turkey, it is. Competitors could pump out aluminum nails for pennies, while the heat-pipe pins actually require somewhat serious engineering due to the forces involved (a computer heat pipe is nowhere near as extreme). The consumer, however, wouldn't see this on the product page, only the price difference - so I suspect they'd buy the cheaper one, find it didn't work well, and write the whole thing off.

I own 2 heat-pipe pins, and they work exactly as advertised - half the cooking time: 20# turkey can be cooked in 90 minutes with a heat pipe pin in each side. These days I actually see them referred to more in COOLING large cuts of meat to stop cooking and/or expedite storage, they halve the cooling time also. 

 

The little experiment AlaMoi mentions doesn't mention a heat pipe. An aluminum nail takes at least 10-20 seconds for the heat to get to the other end. Using my heat pipe pin and boiling water, it gets too hot to hold in 1-2 seconds. 

 

(I ran across this thread looking for info on the pins, and wanted to contribute to the searchable history.)

[dcarch]

Heat conductivity is a function of two factors, i.e. conductivity of the materials in contact and the surface area in direct contact.

 

1. What is the total area of the heat pipe in direct contact with meat? four square inches? How fast can you cook a turkey if your frying pan is only four square inches big?

 

2. The heat pipe indeed can conduct heat fast, however, speed of conductivity being a constant (assuming that the meat is not moving), how fast can meat conduct heat where is in contact with the heat pipe even the pipe can conduct heat super fast?

 

Also, thermal specific heat of air is extremely small for the heat pipe to collect with such a small heat collecting area.

One cubic foot of air weights 0.0807 lbs, and one lb of air has only 0.24 BTU per degree.

Just my unscientific mind trying to reason this one out.

 

dcarch

Edited December 28, 2020 by dcarch (log)

[rotuts]

I think this ( low ) tech idea is interesting

 

but probably not very practical and useful when

 

looking over various cooking techniques.

 

there was a Snappy looking set , came w a hardwood box etc

 

I can't find any ref to them.   I imagine there were ref'd here some time ago

 

perhaps a long time ago

 

the idea that you want to  move heat into the core of you meat makes sense

 

theoretically.

[dtremit]

On 12/27/2020 at 8:25 PM, DocD said:

I think the true heat-pipe thermal pins mentioned by the OP (Rho) were ahead of their time and therefore mostly unappreciated by their target audience and their copycat competitors with the aluminum nails, plus the results were a little strange too - your medium-rare roast would have the more-done gray on both the outside AND in the center around the pin, with a donut of pink between those; slices would certainly look a bit "space age" IE Martian! That does answer one question above - the meat thermometer should ALWAY be put in the slowest-heating part of the cooked item; for normal roasting, that would be the center, but with the pin it goes halfway between pin and surface (the instructions enclosed with the pin explicitly state this). 

 

It seems like a pin or two could be very helpful with something like a bone-in rib roast; one could put them between the bone and the meat to achieve similar results to a boneless roast.

I imagine the donut effect would also be much less of a concern for something you're not trying to cook medium rare — thinking of say a big pork shoulder.

 

On 12/27/2020 at 8:25 PM, DocD said:

The aluminum potato nails may have also been part of the thermal-pins downfall because to the unaware eye, they look similar - but function SO much more poorly as Rho said. The aluminum can get just as hot, as AlaMoi mentions - the difference is in the total heat flux (NOT temperature), the aluminum carries only 1% of the heat energy, is slower, and looks to be of little use. For a potato, this isn't a huge deal. For a 25# raw turkey, it is. Competitors could pump out aluminum nails for pennies, while the heat-pipe pins actually require somewhat serious engineering due to the forces involved (a computer heat pipe is nowhere near as extreme). The consumer, however, wouldn't see this on the product page, only the price difference - so I suspect they'd buy the cheaper one, find it didn't work well, and write the whole thing off.

I own 2 heat-pipe pins, and they work exactly as advertised - half the cooking time: 20# turkey can be cooked in 90 minutes with a heat pipe pin in each side. These days I actually see them referred to more in COOLING large cuts of meat to stop cooking and/or expedite storage, they halve the cooling time also. 

 

Curious what brand you have and/or what other brands you're familiar with. The link in the original post to eBay is long since broken. Looking around on eBay I can find something called a "King-Pin" but it doesn't reference heat pipes or anything similar. Also uselessly it has a thermometer built in — which would always be in the wrong place, of course.

[dcarch]

May be this can explain things in a different way:

 

If a very hot pin (or rod) of limited surface area can be effective in getting sufficient heat into a large piece of cold meat to make cooking better, would n't it be easier to have a pin (a rod) that is electrically heated? Wouldn't a cooking pin like that be widely available and already in everyone's kitchen?

 

Another consideration, a heat pipe conducts heat both ways. Imagine what that will do to your timing concept of resting meat for carryover heat to work?

 

dcarch

 

 

Edited January 11, 2021 by dcarch (log)

[dtremit]

On 1/10/2021 at 2:32 PM, dcarch said:

If a very hot pin (or rod) of limited surface area can be effective in getting sufficient heat into a large piece of cold meat to make cooking better, would n't it be easier to have a pin (a rod) that is electrically heated? Wouldn't a cooking pin like that be widely available and already in everyone's kitchen?

 

Maybe if we all had electrical outlets inside our ovens 🤣

 

On 1/10/2021 at 2:32 PM, dcarch said:

Another consideration, a heat pipe conducts heat both ways. Imagine what that will do to your timing concept of resting meat for carryover heat to work?

 

I think you're overlooking the phase change aspect of heat pipes. They're not just conductors — you are boiling a liquid at one end and then condensing its vapor at the other end. That ends up resulting in an effective thermal conductivity of several hundred times that of solid copper.

The minimum temperature is a factor of the pressure inside, of course, but it's unlikely that one designed for cooking would function "backwards" as you describe — the temperature of the meat once removed from the oven wouldn't be sufficient to boil the water inside the heat pipe.

 

 

[dcarch]

4 hours ago, dtremit said:

Maybe if we all had electrical outlets inside our ovens 🤣

Entirely possible. There is always electric connections in all ovens, the light, the convection fan, and we have electric ovens. A 500 Watt pin will give you 1,700 BTUs. But still a problem because meat is a very poor heat conductor. Thank goodness meat is a poor heat conductor, otherwise browning meat will be a disaster.

 

 

4 hours ago, dtremit said:

I think you're overlooking the phase change aspect of heat pipes.

 

very good comment. But no. I have not. All heat for phase changing heat still have to come from the size of the available heat collecting surface. The surface area of the "heat Pipe" shown will be able to collect no more than a few BTUs.

 

dcarch

Edited January 21, 2021 by dcarch (log)

[DocD]

I collect these pins, and have the Hot Rod, Space Age Cooking Pin, Thermo Pin Potato Baker, and & the aforementioned King Pin, ALL of which are heatpipes. Note that unlike the others, the King Pin has a large heatsink on the outside end to increase heat flux into the pin. The thermometer in it doesn't just report the heat around the outside end - it very much responds to the inside temperature; perhaps it's tied to expansion of the internal heatpipe. 

 

These pins are far more effective than you might imagine, despite the issues mentioned like small surface area, low air density, etc. The proof is in that graph initially posted by the OP / @Rho - the pin decreases the time to reach temperature by almost threefold, regardless of those issues. That's real data that supercedes any supposition about those issues. Note also it addresses the comment by dcarch - the cooling is indeed also shortened, as seen in the second part of each curve, though it's less effective in this direction for the reason identified by @dtremit. This cooling is in fact used for industrially-processed meat, to reduce the time to bring it closer to refrigeration temperature by "more than 25%". 

Edited April 21, 2022 by DocD (log)

[dtremit]

@DocD out of curiosity, what's your favorite?

[DocD]

Easy - having someone else cook! 

 

I haven't tested them side by side, but I probably should, given my collection. Thermodynamically, the fins on the King Pin make sense... but a very simplistic, informal test hinted that heat didn't transfer as fast as the others. But I should cook 3 roasts to test them... 

 

Fun fact - the same principles help conduct heat into porous foods, too, like bread - evaporation at the hot surface then condensation at the cooler interior, whether that's a thousand small gas bubbles in bread or the interior pocket of a larger thing like a popover. It just doesn't happen as much as in the pressurized tubes in the King Pin etc. 

 

Edited April 28, 2022 by DocD (log)

[Bernie]

Just an outsider view, not about cooking.

A few years ago i was investigating heat pipes as a way to cool small chambers for medical research. Basically we needed very low noise (electrical) so having heating/cooling elements away from the chamber is a good isolating method.

After investigation, heat pipes would sort of do the job, BUT they have to be specifically designed for each job. Basically the liquid to gas phase change of the fluid (which is how the heat is transferred) is the most important design consideration.

Whatever the fluid used it must undergo phase change within the hot & cold ends of the tube. So the fluid is super important . BUT more important is the temperature of each end. The "cold" end for instance must be at or below the temperature for phase change, the hot end must be at or above the phase change temperature. So a heat pipe is restricted to specific temperature ranges. The way to get around this is to mix several different fluids each fluid operating in specific temperature ranges. If you think that is easy then you need to consider what happens when a liquid turns to gas, the pressure increases and hence so does the subsequent phase change temperature.

It can be done. For cooling or heating in specific heating ranges they work fairly well (like cooling semiconductors where the final temperature can be set and the cooling power can be continuously adjusted to keep the amount of cooling happening.

I am pretty sure the heat pump would work pretty well to heat from cold to a specific temperature but that temperature would usually designed into the device and is probably not alterable.

 

Having said that, if you desired your roast to reach say 60C then you could design the heat pipe device to operate to 50C and the roast would get there much quicker but above its design temperature it would only heat up in the normal way, the heat pipe would just not be doing anything above its design temperature.

 

(I actually didn't use the heat pipe, i went to peltier cells since they are effectively fully adjustable by means of current control)

 

As a final note, there is probably a reason why they are not common place and widespread, more to do with the tight tolerances needed in manufacture and the expectation of consumers that they operate over wide temperature ranges.

[paulraphael]

On 1/15/2018 at 4:10 AM, Rho said:

It seems like pans and commercial griddles would be improved by adding heat pipes to spread heat throughout and avoid hot or cold spots. 

 

 

A griddle would seem like the killer app for this, if there was a fluid that could work at the range of temperatures you'd need. Even with commercial power, it's hard to get a griddle to heat evenly and to have quick temperature recovery.

[DocD]

On 4/27/2022 at 9:51 AM, paulraphael said:

A griddle would seem like the killer app for this, if there was a fluid that could work at the range of temperatures you'd need. Even with commercial power, it's hard to get a griddle to heat evenly and to have quick temperature recovery.

 

What type of griddle do you mean, panini-press type, stovetop, electric or gas countertop, or what? 

 

I've looked into this myself a little before, I have a love/hate relationship with my aluminum Calphalon stovetop griddle LOL The practical aspects of the engineering make it increase rapidly in cost, depending on what you require from it - this is a much more difficult range to work in compared to cooling electronics. What would people be willing to pay for a griddle that heats a lot more evenly? Even just a 1/4" sheet of plain copper 12x24" would cost $750 - and that's without handles or any particular finish, and it still wouldn't be as evenly heated as something with heat pipes. 1/2" aluminum would cost $250 and not work as well as copper. 

 

Edited April 28, 2022 by DocD (log)

[paulraphael]

20 hours ago, DocD said:

 

What type of griddle do you mean, panini-press type, stovetop, electric or gas countertop, or what? 

 

I've looked into this myself a little before, I have a love/hate relationship with my aluminum Calphalon stovetop griddle LOL The practical aspects of the engineering make it increase rapidly in cost, depending on what you require from it - this is a much more difficult range to work in compared to cooling electronics. What would people be willing to pay for a griddle that heats a lot more evenly? Even just a 1/4" sheet of plain copper 12x24" would cost $750 - and that's without handles or any particular finish, and it still wouldn't be as evenly heated as something with heat pipes. 1/2" aluminum would cost $250 and not work as well as copper. 

 

 

I was thinking of of big commercial griddles, or the similar ones built into fancier consumer ranges. But if the technology could scale down, I'd welcome it in a stovetop version. I recently bought 3/16" thick 2-burner stovetop griddle. It's about a 20 lb slab of steel that's lots of fun to cook on, but it doesn't actually conduct heat very evenly. Between its huge size and relatively week conduction of steel, you can't use the whole surface for things like pancakes. 

 

I probably wouldn't pay $250. Unless it were an upcharge on a $7,000 range, which would make that kind of money seem cheap.