Joule-Thompson Intercooling?

[DR1665]

Curious thought. Any thermodynamics geniuses in the house to tell me why this isn't worth exploring? (I'm sure there's math involved and math, especially on this level, scares me.)

Joule-Thompson effect basically says that, "At room temperature, all gases except hydrogen, helium and neon cool upon expansion by the Joule–Thomson process." This is similar to how your air conditioner and refrigerator work. High pressure liquid is forced through an orifice and then allowed to expand into a lower pressure environment, resulting in cooling.

Okay. So you run a 14b at 25psi. You run outlet-diameter plumbing to your intercooler, where there is a sort of merge collector at the inlet, reducing the diameter of the piping. The inlet/outlet of the cooler are larger sized, maybe 4". I dunno. The super-heated 25psi charge is forced through a smaller opening only to expand on the other side.

The charge is cooler on the other side, right? How much cooler? And how much strain does this place on the turbine and compressor? What if you ran a larger turbo? Suppose you had a 35R pushing 40psi into the IC, but only realized 10psi at the TB?

Just a curious thought. Random. I know. I should have gone to bed hours ago...

 

It is possible, but the amount of restriction that you would have to place on the IC inlet piping would cause so much back pressure that it would negate any gains I would think. The reason being is that the principle is based on expansion based off of pressure differentiation. I don't remember any of the thermodynamic equations off the top of my head, but have 40psi at turbo and having 10psi at TB would only be a 30psi drop. Factor in IC and piping restrictions that caused part of that pressure drop and your probably only seeing 25psi drop immediately post orifice. Not enough to allow any dramatic expansion.
Oh, and your going to have to funnel all that air back down at the throttle body. Thats going to cause an increase in pressure throughout the whole system making the pressure drop even less.

 

[prove_it]

Some ideas make sense, but in the end don't work. Like using the ac compressor to generate cold air to chill the intake air, which just takes power to gain power. I'm not a scientist, but this idea sounds like it would be complex and put more strain on all the componets for a little gain, if any. Just spray the IC with CO2 or nitrous and call it a day.

 

[Rausch]

Quote:
posted 09/30/09 04:20 AM

Hmm.... /ubbthreads/images/graemlins/rofl.gif

 

[prove_it]

i missed that part, /ubbthreads/images/graemlins/rofl.gif /ubbthreads/images/graemlins/applause.gif

 

[DR1665]

Quoting Rausch:
Quote:
posted 09/30/09 04:20 AM

Hmm.... /ubbthreads/images/graemlins/rofl.gif






(Although it was actually closer to 0120hrs, since I don't pay too much attention to post time.)

 

[Rausch]

And that is no time/date fooling.

 

[Hertz]

Quoting DR1665:
Curious thought. Any thermodynamics geniuses in the house to tell me why this isn't worth exploring? (I'm sure there's math involved and math, especially on this level, scares me.)

Joule-Thompson effect basically says that, "At room temperature, all gases except hydrogen, helium and neon cool upon expansion by the Joule–Thomson process." This is similar to how your air conditioner and refrigerator work. High pressure liquid is forced through an orifice and then allowed to expand into a lower pressure environment, resulting in cooling.

Okay. So you run a 14b at 25psi. You run outlet-diameter plumbing to your intercooler, where there is a sort of merge collector at the inlet, reducing the diameter of the piping. The inlet/outlet of the cooler are larger sized, maybe 4". I dunno. The super-heated 25psi charge is forced through a smaller opening only to expand on the other side.

The charge is cooler on the other side, right? How much cooler? And how much strain does this place on the turbine and compressor? What if you ran a larger turbo? Suppose you had a 35R pushing 40psi into the IC, but only realized 10psi at the TB?

Just a curious thought. Random. I know. I should have gone to bed hours ago...



Whoa.

Yeah, you'd gain AT LEAST a tenth of a degree. /ubbthreads/images/graemlins/devil.gif

Truth is there would be very little expansion because the job of the compressor is to feed more gas into the intake system, increasing the pressure in a constant volume, therefore increasing the temperature. click

I think the effect of J-T would be more apparent in the system during blow-off.

 

[curtis]

This only works when going from super high pressures to super low ones like 300psi to 30psi or 25K to -20psi

Think of a gas cylinder like nitrous its sitting on 1K and dumping into the atmosphere or into a vacuum in the engine. Big difference between 1 atm and -1 atm.

Lots of variables going on mass of the intercooler saturation of the air with water vapor, is it in a super heated state or at just the verge of going super heated, what pressure before and after then theres the pressure inside the plenum, radiant heat in the intake pipes and plenum. Then you have to look at the rate of change and time between temperature variations, pressure waves, heimholts ram effects on the plenium and differentail of flow between ports at a given rpm, change in entrophy and the whole BTU mess going on. Figure all day worth of calculations if all variables are known or one pressure and one given system at a certain load. Change it by.01psi or change the Temp .1 degree calculations all change as well as rpm changing one revolution...........You should sleep more. /ubbthreads/images/graemlins/grin.gif

My thermo professor gave a take home part of the final that had to be dropped back off. Don't remember the exact question was 4 years ago but it was a ac system at a steady state and 75% of the variables known the others had to be derived out with calculus then you had to give pressures temp, rates of change at like 30 points along the way taking account for the flow of the system and roughness of the insides of the pipes, bends etc. cooling effects of the fan and ambient temp. Took all day Saturday and half of Sunday for just that. I got the figures correct down to 4 decimal points and all work shown. It got to the point as long as I got the final problem correct didn't care what I got in the class as long as I beat his ass on it and was able to tell everyone I did. /ubbthreads/images/graemlins/grin.gif

 

[curtis]

Oh I'll add one more thing. Google peliter coolers. Electrons are excited in a heated object and not as much in a cooler object so imagine a IC pipe with fins inside and out. then along the pipe are huge 00 gauge cables running to a larger heat sink under the car the hotter active electrons move away. Plastic industries use these to rapid cool pvc pipe. It can come out of an extruder at 350 degrees or so and be -50 in a few millaseconds.

click

 

[DR1665]

Quoting Rausch:
And that is no time/date fooling.


F-f-f-foolin'! lulz

Quoting Hertz:

I think the effect of J-T would be more apparent in the system during blow-off.


One more reason not to VTA. Word.

@Curtis - Peltier sounds interesting, but I guess we're just going to have to stick to the traditional methods of charge cooling. I feel like I just proposed Coke II, but we all decided to go back to Coke Classic before the release.

 

[Rausch]

Quoting prove_it:
Some ideas make sense, but in the end don't work. Like using the ac compressor to generate cold air to chill the intake air, which just takes power to gain power. I'm not a scientist, but this idea sounds like it would be complex and put more strain on all the componets for a little gain, if any. Just spray the IC with CO2 or nitrous and call it a day.



Pretty sure there are a few cars running this set-up. Bear in mind it may not be as efficient as another method, but when taking available space into consideration, most likely the best option. There's an MR-2 that comes to mind in that circumstance.

 

[prove_it]

Running the AC to cool on coming air? I would like to see how they did it. I always thought that the compressor added too much drag to benefical.

 

[Rausch]

You have to factor in design and availability of space. Using the A/C system to cool the incoming air may not be the most efficient, but without the physical space to install a tradtional system, the vehicle would benefit from what would normally be less effective. Simply put A/C intercooling > no intercooling. For example.

 

[Hertz]

...and then you forget that your Keydiver chip has the A/C clutch cutout at WOT... /ubbthreads/images/graemlins/grin.gif

 

[prove_it]

good point rausch. If that were the case though, wouldn't it be better to design a liquid to air intercooler setup that uses a fluid pump and front cooler like the old Celica GT4's used?

 

[Rausch]

Perhaps, but looking at something liek an MR-2, or any other mid/rear engine vehicle, you are either limited by available airflow through the back, or have to plumb cooling up front. Using the A/C would potentially be more compact (It's jsut about all already in there).

 

Actually the paltier setup for a car has been done and is patented for that use even. I read about it in one of my turbocharging books. I believe the guys name is Micheal Perkins or something like that. Major draw back is the large voltages needed for Peltier type setups if this scale. His was 250ish volts I do believe.

 

[prove_it]

There's always more than one way to kill a bird. I'm going to look into these theories.

 

[Andy_S]

Best bet for car with limited airflow (mr2 etc.) would be methanol injection. Numerous guys are running fast mr2's with no fuel control and stock fuel system just spraying the crap out of it. In theory, pure methanol injected at the optimal ratio could yield around -200*F charge temps.