Automotive Air Conditioning Information Forum (Archives)

Provided by

We've updated our forums!
Click here to visit the new forum

Archive Home

Search Auto AC Forum Archives

Big pulley for A6 Pages: 12Last

Jim Blackwood on Fri December 04, 2009 10:28 PM User is offline

I'm looking for a large enough A6 pulley to be able to cut serpentine grooves in the body. I know there was one used on big GM trucks and Case & Massey tractors that used a 7-7/8" pulley. That would work but Everco wants $200 for one with the clutch and I don't want to pay that for something I'm going to cut up if I can help it. A used one should work just fine. Can anyone direct me to a junkyard or other source that would be likely to have something that would work? The root of the V belt groove in the common 5" pulley is too deep and I get into the cement around the clutch plate before I can cut the grooves. Thanks.


bohica2xo on Sat December 05, 2009 2:42 AM User is offline

I usually just turn them down far enough to shrink a serpentine ring on the outside. The Sanden / Seltec 125mm pulley can usually be bored out to fit properly. Many other boneyard pulleys night be used as well. You can tack weld it in a couple of spots if you are worried, but a good interference fit on that large of a diameter has always been enough for me.


"Among the many misdeeds of the British rule in India, history will look upon the act of depriving a whole nation of arms, as the blackest."
~ Mahatma Gandhi, Gandhi, An Autobiography, M. K. Gandhi, page 446.

Jim Blackwood on Sat December 05, 2009 10:25 AM User is offline

Thanks, that may be the way I go in this. First I'll have to get another pulley since the one I have is now scrap. Also I wanted to decrease the OD as much as I could and I can go smaller gutting the grooves in the A6 pulley.


ice-n-tropics on Sat December 05, 2009 4:05 PM User is offline

Hi Tim's Band of Brothers, long time no see.
To expand on some details related to Brad's comments for shrink fit, I recently turned down still another, 1/2" groove clutch pulley for a shrink fit with a Sanden poly-vee (PV) sleeve as follows:

1) Chuck the inside female trepan (where the coil winding outer housing clears the pulley under the grooves) of the A-6 clutch pulley with a 4 jaw male chuck and use an indicator to center within 0.0015" total indicator reading (TIR). I don't like to chuck on the male hub which contains the bearing because the thin, flimsy pulley friction surface may distort due to the transfer of torque as the cutting tool digs in during machining. Take light cuts to reduce forces.

2) Chuck the poly-vee clutch pulley on it's OD and indicate to 0.0015" TIR
3) Target a FN2 class shrink fit (see Machinist Handbook), e.g., total diametrical interference between the A-6 inner part and the MPV outer part of 0.0039" to 0.0016".
Caution: Too much interference can bow the final friction surface of the shrink fit assembly and the clutch will burn up due to slippage friction heat.
Deburr sharp edges. The PV sleeve should be have a min. radial thickness between the groove OD and the sleeve ID of 1/4" for strength.

4) Machine the 2 parts of the assembly so that they have a common front surface in the axial direction (gage line/GL). KISS: This way during the shrink fit assembly the outer sleeve can be just dropped over the hub till they both rest on a flat surface (like a hand press bed). Keep in mind that PV clutches should align axially with the crankshaft grooves very closely or the misaligned belt will squeak.

5) Put the inner A-6 hub in a ice chest with dry ice to shrink it's OD. It's OK to leave the bearing in the hub.

6) Put the outer PV sleeve in a 450 F oven to expand the ID.

7) With the cold hub positioned in a hand press, drop the hot outer PV sleeve over the hub until they both rest on the press base and then let the temperatures normalize. If the outer sleeve doesn't go all the way on, the back up plan to maintain the intended GL is to use a little press force while there is still a max. temp. delta to overcome any resistance between the 2 parts due to distortion or burrs.
Caution: Never apply force which causes the bearing's inner and outer race to resist each other. This will result in brinelled ball bearing races and make running noise and shorten bearing life.

8) On a flat surface like a surface plate lay some fine grit sand paper and buff the friction surface. It will be evident if it is still flat. If flatness is not within 0.001" it may be necessary to recut the friction surface in a secondary lathe operation repair. The machining grooves in the friction surface should match a new clutch, not rougher or finer.

9) Before releasing the vehicle, burnish the clutch on an operating A/C system for a minimum of 100 cycles to increase the static torque of the clutch by 10 to 30%.

8) Use 3 or 4 3/8" long tack weld on the rear side for insurance.

9) Clean and paint the assembly with Rust-O-Leam.

I haven't been very for much help around here lately.... busy with military HVAC and a book "HOW to Air Condition Your Hot Rod" due out by mid 2010 by Motor Books.
Got a 17" inside white tail antler spread and built a fishing lake these last 3 weeks on a pine tree farm in Montalba, TX.
Old IV Guy

Isentropic Efficiency=Ratio of Theoretical Compression Energy/Actual Energy. How To Air Condition Your Hot Rod

TRB on Sat December 05, 2009 4:16 PM User is offlineView users profile

Had the Big O ask if I knew how to contact you a few months ago. My email address for you was at the compressor company! Not sure what they wanted but if you want send me an email. I'll get it over to the person I was speaking with.

Glad to see your still having fun!


When considering your next auto A/C purchase, please consider the site that supports you:

Edited: Sat December 05, 2009 at 4:16 PM by TRB

Jim Blackwood on Sat December 05, 2009 11:06 PM User is offline

Appreciate the tips, sounds on the money.
But, I want to reduce pulley diameter as much as is reasonable and the shrink fit approach adds about 3/4" and likely more to the diameter of a poly-v cut into the cast iron body of an A6 pulley. Everco does sell the pulley I'm looking for as part of a clutch package minus coil. Part number 48275, or GM # 6551129, Case/IH # A44935 and Massey # 533719-M91. Used in GM heavy and medium duty trucks from '80 to '84, it has a pulley diameter of 6-7/8" and a belt width of 7/16". There is plenty of meat to cut off the pulley down to the OD of the clutch and then cut the grooves without getting into the cement around the clutch plate. Only thing is, Everco wants $200 for it, which seems a bit much to pay for something which is just going to be cut up. So I'd like to find a used one, and then if the system works out as expected I can consider new parts once everything is proven, if needed.

Incidentally, how much rpm will the A6 tolerate? Will it handle occasional excursions north of 12 grand without damage? Thanks,


HECAT on Sun December 06, 2009 8:26 AM User is offline

Originally posted by: ice-n-tropics
Hi Tim's Band of Brothers, long time no see.

I haven't been very for much help around here lately.... busy with military HVAC and a book "HOW to Air Condition Your Hot Rod" due out by mid 2010 by Motor Books.

Got a 17" inside white tail antler spread and built a fishing lake these last 3 weeks on a pine tree farm in Montalba, TX.


Old IV Guy

Man, it is GREAT to hear from you, and as always, with fantastic info to share! I have often been wondering what you were up to. Hunting, fishing, writing, consulting, and hot rods; is just like what I had hoped and pictured you were doing. I too had only your former business contact info. Please give me a call if you can, I would like to catch up and hear more about the fun you have been having.

[email protected]


HECAT: You support the Forum when you consider for your a/c parts.


NickD on Sun December 06, 2009 12:18 PM User is offline

Incidentally, how much rpm will the A6 tolerate? Will it handle occasional excursions north of 12 grand without damage?

Last vehicle I had with an A6 in it was a 78 Fleetwood that came with a 2.2 rear end ratio, could only drive the thing at 55 mph at the time but cooled extremely well at 1,700 RPM driving through the Dakotas in a 125*F ambient. Was a 1:1 ratio from the engine to the compressor. Trying to visualize what you are putting this thing in that hits >12,000 RPM. With that hunk of iron, would think you would see pistons flying out at the speed.

Jim Blackwood on Sun December 06, 2009 12:59 PM User is offline

Now try not to be negative Nick. It's going in a convertible.


NickD on Sun December 06, 2009 1:45 PM User is offline

A V-2 Honda 500 CC engine did come to mind, can be modified to output, 17,000 RPM, 13,000 is stock.

Would that be a convertible before or after the compressor blows? A6 doesn't really produce any extra cooling above 2,000 rpm, system capacity is already met. Maybe you are thinking about turning the compressor pulley down up to a 1:4 ratio that would also increase the toque load on the belt by that much as well. Definitely would need a much larger belt, and even a larger high torque engine to drive it. But won't gain anything by doing so.

Jim Blackwood on Sun December 06, 2009 2:31 PM User is offline

Interesting. So what's it look like when an A6 blows if I might ask? Anyone ever see that? Inquiring minds want to know!

But to elaborate a bit. I find it a little odd that you'd have a 5" crank pulley driving the compressor, but certainly not impossible. And, running that slowly you must have had every detail nailed down tight so I find your professionalism in the install remarkable and somewhat unusual but perhaps that's more the norm for this group. I am not entirely that optimistic regarding my own skills in this area but plan to do the best I can with what I have available and initially accept a certain level of inefficiency as a consequence. But that's a fairly small part of the picture. My crank pulley is 6-3/4" diameter. I didn't go out and measure it, but simply estimated the clutch OD for the purposes of calculating speed at 4" diameter. It may be a little larger than that. But, that means at redline the compressor would be running about 11812 rpm. I simply wanted to know the design limits of the unit. There were factory SBC's that would exceed 7 grand (and may have had a larger diameter crank pulley) so I would think I'm not entirely out of line here. Considering the axial piston design I wouldn't think that speed should pose much of a problem for it, certainly the stroke length and piston speed is not severe enough to cause material failure provided the construction is reasonably robust. But if nobody has the relevant data that's OK and I don't mind being the guinea pig here. If it blows up, it blows up. Just please try to refrain from telling me I'm being stupid to try it if you can, since unless you do have the relevant info you really aren't any closer to the answer than I am. Maybe the design limits of the Sanden are easier to obtain, and it does have a similar construction. Also, perhaps the poly-v belt-on-damper construction*is* larger in diameter than the typical bolt on single or double V pulley and therefore justifies a larger diameter compressor pulley. These are details I need to resolve obviously.

But you seem to think it's mass overkill. OK, maybe so. But it isn't a typical A/C installation. There are other reasons why I want to maximize compressor output and maybe we'll get into that and maybe we won't. Let's just say that for the time being the concern is A6 rotational redline and go from there if you don't mind. I realize output is directly related as well as other factors, but I have an abundance of power to drive the unit and the only other limiting factor at this point is sizing of the condenser, and that has not yet been determined so there is some leeway there as well.


bohica2xo on Sun December 06, 2009 3:45 PM User is offline

Ok, so you want to turn parts from a 25 year old medium duty truck @ 12,000 rpm? I recomend Kevlar undies for that one.

Without more information about the specfic capacity requirements, it is hard to make a specfic recomendation. The A6 is a fantastic compressor, but it will not deal well with extreme high rpms - no compressor will really. An RPM cut-out switch would be the logical answer if the drive spins too high.

A more reasonable apporach would be to use a V7. They are used on the corvette, and will take 8500 rpm (upshifts, not continous) according to Delphi. A variable displacment compressor with evaporator pressure control. Available as a new unit, serpentine pulley already installed.

It does not matter which compressor you choose, the V/E drops off as the compressor speed increases. An A6 spinning at 10,000 rpm does not put out twice the volume it did at 5,000 rpm.

Take a look at the V/E vs speed:

So, how much cooling do you need?


"Among the many misdeeds of the British rule in India, history will look upon the act of depriving a whole nation of arms, as the blackest."
~ Mahatma Gandhi, Gandhi, An Autobiography, M. K. Gandhi, page 446.

NickD on Sun December 06, 2009 4:11 PM User is offline

Just glanced over the entire thread, don't recall seeing a specific vehicle model listed, just turning down a V pulley on an A6 to assumingly be driven off a serpentine belt and 12,000 rpm. Not really much to go on.

Jim Blackwood on Sun December 06, 2009 6:30 PM User is offline

Well, you guys have been real helpful so I guess it wouldn't hurt to get into the details a bit more. Maybe you'll stick with me on this long enough to at least point out why it'd be a bad idea. Looking at that capacity chart and sort of extrapolating the curve based on what is there, it would seem that the V7 tops out on capacity between 4500 and 5000 rpm. Whether or not the output might even decrease beyond that speed is a valid question I think, but the only thing to be gained beyond that point is area under the curve, which will have a limit. Let's say 5 grand and starts back down, then right around 8 grand we'd be into diminishing returns where additional speed does not increase area under the curve any longer. So *IF* the same curve were applicable to an A6 there would probably be little or no point in spinning it above 6 grand. If it were the same. Fair enough, no real way to know that without the same chart for the A6. But, the shrink fit poly-v pulley begins to look a lot more attractive. Plus, being probably steel as opposed to cast iron it should strengthen the hub. So apparently, (and reasonably so) it seems most cars are set up with some overdrive to keep the compressor in the center of the curve under "most used" engine speeds. Modern cars probably also cycle the compressor off above a certain engine speed. I won't be doing that.

So then. The car is a small convertible British sports car with a big engine ( ) which is in the process of getting a new and still larger engine. On the stand is a Buick 340 with an Eaton M112 blower which should go into the car this winter if I can just get past this compressor issue and get back to finishing the intake. The engine is a relatively mild build but is designed to take advantage of the blower output to increase the rpm range by improving the top end breathing and an integral part of this is an effective intercooler. The last engine (Buick 215 with Eaton M90) used an experimental ethanol charged intercooler based on heat-pipe technology which was effective enough to allow boost levels of 16psi without destroying the engine, and I think using A/C can be even more effective. But it isn't simple, and in addition to running the intercooler, the system will be set up with cabin air. I believe the compromises involved in doing this can be overcome, the immediate methods involving a pair of solenoid valves, TXV's, probably one POA, an oversized drier to act as an accumulator, and possibly an accumulator of some sort on the suction side. The objective is to achieve a coolant "dump" into the intercooler that will last 15 seconds by a combination of maximizing compressor output (area under the curve above) and a suitable drier volume. (beyond 15 seconds would be OK, but I'm not sure it is feasible)

So there you have it. No worries about freezing, as the IC will be operating from ambient up to around 300*F under full boost and for too short of a time to form ice, and will not operate except under boost conditions. As for cabin air, the TXV/POA combo sounds like an effective way to keep that under control, simultaneously allowing recharge of the drier.

So here is your opening. Why won't it work?

Jim Blackwood, Esq.
Blackwood Labs, LLC
Prototype Development

bohica2xo on Mon December 07, 2009 4:13 AM User is offline

BTDT, several ways...

I owned a Sunbeam Tiger. No room for anything. At least you stretched the body a bit.

I have done the refrigerated intercooler, although not quite as you described. Actually my flushing operation is similar to your plan. I use an evacuated 30LB recovery cylinder as a receiver, and blow liquid R600 through the item under flush. This works well, but the receiver will only take a 10 second blast before flow goes to hell - and you need to recover the tank to continue.

The first problem I see with your plan is vessel volume. It would take a rather large evacuated tank to make it cool for even 15 seconds. A large tank of cooled liquid would also be necessary. Finally, you will need a lot of condensor capacity.

The V7 is designed for passenger compartment cooling, so the built in POA might be a detriment for your application. A Seltec HX21 might be a better choice, especially if you want to run lower suction pressures.

The other possibility you might consider is a dead loss CO2 system. Carry a 20 pound CO2 tank in the trunk, and blow it through the heat exchanger you are using for the intercooler. This of course does not give you cabin A/C. I did this in my 900SPG for a while when working out the high boost issues.

My final system in the 900 was more elegant. I installed a 30 quart ice chest in the cargo area. In the chest was an aluminum tank that fit the inside of the chest, that held an evaporator. The tank held 25 quarts of coolant / water mix as well. I tried several controls, but the dodge H valve worked as well as anything else. I adjusted it for 20f or so. The balance of the plumbing was the usual stuff. RV water pump (make sure it has the santoprene diaphram for the cold), liquid /air intercooler assembly & a lot of insulated plumbing.

The refrigerant was tapped off of the vehicle A/C. The initial pulldown from a 3 day hot soak was of course somewhat longer, mostly because the OEM system was never designed for this. Once the box was cold, the A/C worked just fine. The stored energy in the box was enough. I could run 19+ psig of boost for much longer than I had road to eat up. Since the intercooler was designed for water & a heat exchanger in ambient air, there was a big advantage to the below freezing mixture.

Of course I was dealing with a 2L engine, but the system could be enlarged. Your trunk could hold an 80 quart cooler - and improve traction at the same time. Nobody ever thinks the "ice chest" is anything else. Make sure the top of the internal tank is just far enough below the lid to lay a pop can down on it's side & close the lid.


"Among the many misdeeds of the British rule in India, history will look upon the act of depriving a whole nation of arms, as the blackest."
~ Mahatma Gandhi, Gandhi, An Autobiography, M. K. Gandhi, page 446.

NickD on Mon December 07, 2009 4:55 AM User is offline

Since we are not talking just about cooling the passenger area, but the intercooler as well. Brother-in-law had an MGB like that, was very fun to drive, but I needed a shoe horn to get into it. And you probably noticed the British with their more tepid climate forgot to add any compartment insulation to that thing. Sure picked an engine with a heavy valve train, must have very stiff valve springs to get any kind of rpm, prefer a dual overhead cam setup. Isn't oversteering also a problem with all the extra weight? Then we have Homeland Security that doubled the size of our police force, really hate taking my Supra Turbo out, get speeding tickets when parked. It's not fun anymore and those guys would be the first to hide if any terrorists were around.

Bohica has given some very good suggestions, V-7 is a variable displacement compressor that would vastly simply your passenger cooling and works just like a POA, but without all that extra stuff, looks like you have a space problem and more suited to your speed requirements. I am really not negative, just practical and realistic.

Jim Blackwood on Mon December 07, 2009 10:31 AM User is offline

That's OK, you guys have been real helpful so far. Based on how the MG ran with the 215 (around 300 hp and 2400 lbs) I'd say 15 seconds is going to be more than enough time at max boost under any but very extreme conditions. Normally it'll be more like maybe 5 seconds or less. And you are right, I loved the crack about speeding tickets when parked. The 340 (with alloy heads) will up the power a bit and the weight penalty is only 82 lbs over the all aluminum 215. (which, minus blower, is 30+ lbs lighter than the stock engine) So handling is not affected. Engine weight will be just a bit over 400 lbs, roughly equivalent to a SBF with alloy heads. The valvetrain is pretty standard for a pushrod motor so it should be capable of 7 grand with the hydraulic custom grind and the right springs.

I'd prefer to avoid the complexity of a trunk mounted system, though I realize getting enough thermal density to support the 15 second blast is going to be a challenge. Ultimately it may simply come down to limiting the boost. (it will be less than the 16+ lbs seen with the 215 but static compression is higher) Prior to that I want to see if I can come up with a workable system. I have an evaporator (BMW) which is 3" x 8" x 13" and is a good fit under the blower, and a "H" type TXV (Jaguar I think) which is a 2-1/2 ton unit with a 3* preheat. If I can determine how much liquid freon will be needed to support flow through that for 15 seconds I'll be closer. Then of course, where is the vaporized freon going to go? The A6 would be working at full capacity but very likely would not be able to keep up. Since the solenoid valves would be at the inlet of the evaporators the freon could backflow into the cabin evap as well so that adds some volume but that still may not be enough. It seems to me that the greater the suction side volume the better the system will work. Regardless of how much liquid volume I add on the high side, it has to expand if it's going to do anything. I think we're on the same page there. I could see doing something like adding large diameter tubing sections inside the cabin and plumbed into the suction line, running under the seats for instance, in order to add volume and also insure against slugging the compressor, and I think using the largest possible displacement on the compressor is going to be crucial.

Now regarding the cabin air, I've made one or two assumptions that might bite me. Is there a POA that is easily adapted to a cabin evap? I don't have any idea what that would look like and I need a better idea of what I'm dealing with there. The idea is to trigger the solenoid valves with boost, so normally the IC would be shut off from the high side and flow would be through the cabin evap. The idea is to charge the system with just enough freon (or perhaps propane) to fill the drier/accumulator, thereby throttling the A6 by reducing inlet pressure. After boost the A6 would have high inlet pressure until it refills the drier.

Back when I bought the M90, a Magnuson engineer wrote telling me he felt a 60-70* drop in discharge temp was achievable with an air/water IC of approx. 4-3/8" x 9" and max 2 row so this is something of a benchmark for me to work from. The BMW evap core is considerably larger and has the advantage of operating below ambient. Boost will "probably" be limited to less than 8 psi but static compression is 10.6 so cooling the charge is going to be relatively important, depending on final boost pressures. It's not expected to be an all out racing machine, and in fact will often serve as my daily driver. So we're not going all out on this challenge, just trying to get a workable system that actually provides a measurable amount of benefit.


NickD on Mon December 07, 2009 12:59 PM User is offline

Just a thought, a five second boost should get you from 0-60, a 10 second boost, either in jail or up someone's exhaust pipe. Maybe we should just concentrate on keeping you cool. As said before, a V7 does not require a POA, just an accumulator and a fixed orifice, very simple system. But it's your call.

Looks like you have a spare V belt pulley on your crankshaft, more than likely for a compressor. V7's that I am aware of are serpentine. Were you planning on tapping your compressor with your supercharger?

ice-n-tropics on Mon December 07, 2009 5:30 PM User is offline

Hot Rod MG:
We had a heck of a time routing the drivers side exhaust for as 215 ci in a MG. Finally reversed the exhaust manifold forward (180 deg flip) and then 180 deg underneath again to aim toward the rear of the car.
Maybe Ford copied Bohica with their patent on a intercooler chiller system. Ford hydronic charged air cooler chilled a reservoir of antifreeze type solution and pumped it into the intercooler with a closed system.
instead of putting ice on the exterior of the conventional intercooler at the starting line.
1 BTU changes 1 lb. H2O by 1 deg F
The A-6 at high extreme evaporator heat load maxs out at 30,000 BTU/hr (125 BTU/15 sec.) or roughly 10 lb of refrigerant/min (2.5 lbs./15 sec.).
That's about enough refrigeration to cool 125 matches , not 400+ cfm of superheated air.
Even a refrigerant storage tank at 200 psi feeding multiple TXVs can't feed enough BTUs
Got to have some thermal storage to drop the charged air temp to ambient or below and drag your beast.
Old IV Guy

Isentropic Efficiency=Ratio of Theoretical Compression Energy/Actual Energy. How To Air Condition Your Hot Rod

Edited: Mon December 07, 2009 at 6:10 PM by ice-n-tropics

Jim Blackwood on Mon December 07, 2009 6:14 PM User is offline

Yeah, I know what you're saying. Legal is good and I rarely exceed the speed limit these days (who can afford the "road tax"?) But, I'm already pretty much committed to the M112 at a 2.4 overdrive ratio so I'm pretty sure I'll have enough boost to have to have an intercooler of some sort. And to be honest I sort of enjoy the technological challenge.

Bohica, I haven't found any information on that Seltek compressor (HX21) and my knowledge in this area is very limited. I do know that I want to maximize displacement and finding displacement figures on compressors has not been that easy. I also need an 8 rib pulley which isn't all that common. So I'd be very pleased to take advantage of your recommendations if possible. I'd like to match the A6's displacement (about 10.6 cubic inches) or exceed it if possible. Your comment about the POA is right on the money I think. I could be wrong but it seems like it could be a detriment on the intercooler side of the system, but helpful on the cabin air side if it's feasible to set the system up that way. Will the POA allow backflow so I could utilize the volume of the cabin evap as accumulator volume on the suction side? (Do one of you guys have a link to a photo of a POA?)

Your cooler based system is pretty extreme! I don't expect to reach that kind of boost levels though so I think I'll pass on that one.

So far I've not been finding compressors with an 8 rib pulley and the only one I've seen that matches the A6's displacement is that new double Sanden unit and they don't sell it with a serpentine belt pulley. That's why I'm looking for the big A6 pulley. Turns out there were additional applications for it on commercial equipment and other brands like Caterpillar used it. Nick, the front drive is all new for the 340, starting with an 8 rib fluid damper to drive the blower and the A/C compressor. It may drive the alternator as well, but I can't start on that until I have the compressor in place. That sets the location of the belt, the damper, the blower, and the plenum that houses the IC evaporator. So I'm really at a standstill until I get this worked out.

>"Were you planning on tapping your compressor with your supercharger? "
I didn't really understand what you were asking. Sorry.


Jim Blackwood on Mon December 07, 2009 7:11 PM User is offline

Old IV Guy, thanks for that info, it's very helpful. If I understand you correctly then, the A6 can pump 2-1/2 lbs of refrigerant in 15 seconds, maxed out? How is this affected by suction side pressure? Would it increase as pressure increases or is that the absolute max? What I mean is, if the (hot) evaporator is being flooded to the extent that it's hiking the suction pressure way up will the A6 pump more than that?

I honestly do appreciate your superlatives (downright funny in fact) but I'm afraid they border on BS. No offense intended, but as a child I dumped a 1 lb propane bottle on a bunch of bumblebees (one at a time) and I can assure you that 2-1/2 lbs of refrigerant will do more than put out 125 matches. And frozen Bumblybees wake up mad! But maybe you could help me get a feel for just how much cooling capacity a 2-1/2 ton TXV running wide open could pass? I'd like to be optimistic about it and I have a feeling you'll hold my feet to the ground. Can you translate it into a common example for instance?

As of now, I'm thinking I have to deal with probably about 600 to 900 cfm of hot air and I have a 3 x 8 x 13" evaporator to cool it with. If I can drop the temperature 10-20 degrees there will be a noticeable benefit and more would naturally be even better. How far will 2-1/2 tons go? (150 to ~225 cubic ft in 15 seconds. How much can I cool it with a max refrigerant dump through the 2-1/2 ton TXV if the hot air is at 200-300 degrees F?)

By the way, I had the same thoughts regarding Ford's system. Good chance that patent could be invalidated if someone really wanted to push the issue, but at least they got it on the market. I've seen that reverse exhaust setup on the MG. Butt ugly, but cheap and effective. Gotta love that. It's much easier these days, if more expensive. However one method involved cutting and welding two manifolds to make a center dump and I though that was a pretty neat solution. I went with the hand built custom headers. Lots of work. Don't think I'll be doing that again any time soon.



bohica2xo on Tue December 08, 2009 2:38 AM User is offline


Sorry for the confusion, the Seltec model number is properly stated as TM 21HD. They are 214 cc/rev displacement units - Tim @ AMA can get them. Should have an 8 groove clutch available, since they are used on shuttle busses.


I don't know about the Ford unit, I dreamed up that rig recalling the old Typhoon system - and thinking cold water was a better plan. It gave me a place to put the steaks on the way home too... You know me, I just do what I want with my own junk. Dodge finally put a hydraulic fan drive in a car - I guess the one in my '79 E350 was a bit ahead of it's time too.

Was it "extreme"? Perhaps. But if you ever worked on a 900, you would understand. Torching a set of pistons is a big deal. They are FWD, with the crankshaft axis the same as a RWD car - but the engine is in backwards. The acessory drive belts are up against the firewall, and the clutch disc is right behind the radiator. To make it more interesting, the exterior top surface of the transaxle is the interior surface of the engine oil pan. This means that the entire power package comes out - and the "oil pan" weighs 200+ pounds. Once you get an engine build done, you want to get 200k out of it before you do it again. The good news is you can swap a clutch disc in 20 minutes without a jack. I sold that beast to a kid with big dreams, I hope he is still alive.

Regarding the system with "tubing under the seat", I believe that Boyle's law will bite you right in the ass. Like I said before, vessel size is important. You can meter off liquid refrigerant (orifice & check valve) and sequester it - all day long if you like. Store 20 pounds of it. Dump it through a huge orifice with a solenoid valve - shazam! -40f air.

As long as you have enough space to capture that much liquid as heated gas - and maintain a low side pressure of 30 psi or so. Take a quick look at any P/T chart, and decide how cold your intercooler needs to get. That is the maximum pressure you can have in your gas accumulator. I can see a couple of 1 inch bore tubes under the seats - leading to a 30 pound recovery tank in the trunk. With 300f blower discharge temps, the volume of the refrigerant is substantial (damn you Boyle!)

The final issue you MUST plan for is oil return. The refrigerant carries the lubrication for the compressor. Between 3% and 6% of the liquid (by weight) should be oil. Once it becomes gas, the oil drops out. It migrates back to the compressor because of the high velocity vapor headed back to the pump. If you use a huge accumulator & large lines, a sump & seperate oil return may be necessary.

I have solved this problem on motorhomes with 60 feet of plumbing by installing an oil seperator on the compressor discharge, and directing the oil back to the compressor intake via an orifice. Yet another piece of hardware, and this must be mounted close to the compressor - before the condensor. Oh yeah, the condensor. You will need one big enough to support the heat transfer you expect...


"Among the many misdeeds of the British rule in India, history will look upon the act of depriving a whole nation of arms, as the blackest."
~ Mahatma Gandhi, Gandhi, An Autobiography, M. K. Gandhi, page 446.

NickD on Tue December 08, 2009 5:20 AM User is offline

Forgive me for my ignorance, is this how your intercooler is installed?

Just seems mounted right above the engine is a rather hot spot to begin with, with minimum air flow. Reason for supercharging is to get a greater air/fuel mixture into the cylinder, faster the engine runs in a naturally aspirated engine, the only thing drawing in this mixture is the suction caused by the piston going down. And with less time, less mixture gets in that drastically reduce the power output of the engine. Even at lower engine speeds where torque is maximum, the cylinder may only fill to 60% of it's capacity.

Detroit went completely backwards on this in the muscle car years doing the cheapest thing possible, just increase the displacement of the engine. Hey I got 500 CID, but in reality, more like a 250 CID engine. One limiting factor is air restriction and shortening the air flow route by fitting each cylinder with it's own carburetor, for the output a separated tuned pipe or headers as they are called helped to increase the suction. A huge restriction occurs by only using two valves per cylinder, cure was to use a cam grind to keep the valves opened longer, problem with that, damn engine wouldn't idle, but hey that was cool. Carburetor also had the job of mixing the fuel with the air creating even more of a restriction. The common cures for these problems is using either 4-5 valves per cylinder, variable valve timing optimized over the speed range, and fuel injection permitting a very large TB to be incorporated.

Cure for valve float with a massive valve train was to put the cam on top of the valve rather than to stiffen the valve springs that is robbing the engine of it's power potential. MB even came out with such a cam that completely eliminated the valve springs with an action that would not only open the valve, but close it as well. But a typical well designed engine capable of an 8,000 rpm red line would use on 35 pound springs.

Yet another approached introduced by aviation is the supercharger, love the prefix super, needed, because the higher you go, the thinner the air, essentially, just an air compressor that was later adopted to automotive use. But the problem with compressing any gas is heat, and the hotter the gas, the less density it has, so the heat has to be dissipated first before entering the engine, an intercooler.

This seems to be the problem here, how to cool that hot air. The way my Supra dealt with this problem is to mount the IC in front of the condenser with very large tubes feeding and gathering back the cooler air. Augmented by an engine driven fan and two auxiliary electric fans. With some mods, capable of producing 675 HP with a 190 CID engine, all those other restrictions listed above where already dealt with.

Now for IC cooling, would think the way the IC is mounted right smack in front of the grille the car would perform far superior driving it on a -35*F as opposed to a 90*F summer day. Well a little better, but not nothing to write home about. Something to consider in your project. You are starting off with a heavy valve train, only two valves per cylinder engine that requires extremely tight power robbing valve springs and wondering about using another power robbing AC compressor to hopefully get some additional cooling. Seems like robbing Peter to pay Paul approach.

Just yet another problem with cars like this besides those sirened tax collectors, insurance rates. An added benefit of a well designed engine is far better efficiency to obtain better fuel economy, but don't try telling that to your insurance agent.

Jim Blackwood on Tue December 08, 2009 10:26 AM User is offline

Thanks guys, a lot of good points there. I'm not sure I'll get them all answered but will give it a try, in no particular order. About that old intercooler design... well to start with it was a compromise from the very beginning due to space constraints and that was caused by adapting an existing intake manifold, so this time I started with a clean design and built (or rather am building) the entire intake. That first design was a series of compromises, and to tell the truth, though the hardware was in place I never actually got around to datalogging the temps above and below the core so to this day I can't really say how well it worked. What I can say is that it allowed me to run 16+ psi of intake boost (on a Roots type blower) on an 8.5:1 engine with stock type aftermarket cast pistons for quite some time (est. 20K+ miles) with the only piston damage being some erosion of the top ring land on one piston. I was pretty happy with that. The IC itself was unconventional. It was based on "heat pipe" technology, being a sealed radiator, evacuated and charged with pure ethanol to 50% capacity so that the liquid line hit half way up the unit. Note the 15* installation angle, which, along with acceleration forces keep the liquid in the lower half and promote the return of condensate. Equilibrium being what it is, any heat introduced to the lower end immediately caused phase change and migration of evaporate to the cold end where it condenses and returns. The thermal density is established by the mass of the unit and the heat required to vaporize the ethanol. Excessive heat input would eventually evaporate off all of the ethanol of course and increase internal pressure accordingly but no rupture or bulging was ever noted. I think it was reasonably effective but should have been bigger, plus it just looks a bit odd. The original plan was to enclose the upper end and force cool air through it but I never built that part. The most extreme testing occurred one hot summer day at the local dragstrip (something I rarely indulge) where I made a series of about a half dozen back-to-back runs and eventually managed to blow a head gasket. (standard .040" composition type, then upgraded to copper) It was never run quite that hard again. Although a few unobstructed high speed runs of mountain roads similar to "The Dragon" could be considered a close second, acceleration was never as sustained and cool down was more frequent. I was very happy with the performance of the car, and satisfied with the intercooler for the intended purpose. But there was a great deal of room for improvement. As far as the theory goes, a heat pipe can conduct heat away somewhat faster than a solid copper bar of the same size. That would be the equivalent of the tubes in the radiator. In other respects it is more or less equivalent to an air/liquid heat exchanger. So a relatively effective device in a small package. I could still go that route with the new engine, but packaging is the biggest challenge as you can see by the configuration and the angle. I thought I'd try something different, and potentially as effective.

So getting back to all that hot gas for a moment, I certainly agree that is the challenge. According to Old Guy, if an A6 can pump off 2-1/2 lbs of hot refrigerant in 15 seconds maybe there's a chance this can work. The TM 21 at 13 would be even more effective, and lighter. Could be overkill but if it could be made to fit, the TM 31 at 19 would appear to be capable of drawing off nearly 5 lbs of refrigerant in that same 15 seconds. So the question for me is, how much will one pound of refrigerant cool 200 cu.ft. of air? With a rough figure for that I can determine if this approach is worth pursuing. Now I realize it isn't quite that simple and things like evap pressure must be accounted for in the calculations but I'm just going for a rough idea here, say an average value. So if that 1 lb of refrigerant can produce a 5 degree drop we're at least in the ball park, but if it can only produce 1/2 of a degree it probably can't be made to work with the current technology. In that case I'd have to agree with Old Guy and it'd be time to punt. Or take another look at the heat pipe exchanger.

So as the last point, yes I agree this is an inefficient way to make power, and is probably a poor choice in terms of optimizing the car. But, it's a toy and toys don't have to be efficient, just fun. We have another car that the denizens of BritishV8 (website and forum) are building as a group project where we are using the cubic inches approach and have transplanted a 455 Buick big block into an MGB-GT. Amazingly enough it fits well, is only about 100 lbs heavier than stock, and has very good weight balance. Others in the group have taken the high tech approach. Dale Knapke for instance has a TR6 with a turbo'd 4 and at something like 23 lbs of intercooled boost I think he's pumping out around 600 horses. We all applaud each other's efforts and enjoy all of it. It's a very diverse group.


bohica2xo on Tue December 08, 2009 1:59 PM User is offline

The TM31 is a nice compressor, but it carries a hefty price tag. Diesel pusher bus motorhome owners gasp when I quote them a replacment cost. It will not tolerate serious overspeed, but it does have an oil sump & sightglass...

The 900 did well at Pikes Peak, but I suspect a better (braver) driver would have found the limit of the cooling near the end of the run.

Now for the science

R134a has a latent heat of vaporization of 215 kJ / kg. That works out to about 92 btu's per pound of refrigerant.

Your heat pipe fluid (ethanol) was 855 kJ / kg. You know the ethanol worked, and you should know the mass. Obviously to match the ethanol performance with 134a, you will need to move 4 times the mass. What you don't know is the flow rate of the ethanol, or how it actually behaved.

It is probably easiest to keep the units all mass, especially since we usually do boost calculations in mass. So, calculate your boosted mass flow expectations and expected blower discharge temperature.

From there it is math. You know the mass of air. You know the delta T you want for that air. The latent heat of vaporization of the working fluid is known. The mass of the working fluid will be a proportion of the mass of the air to be cooled to the latent heat of the working fluid.

I suspect that you did not record the delta T across the old design. That data along with the actual operating pressures of the working fluid would make a better set of data points. If you have that data, then you have the actual heat needed.

I recall a pressure "drop" (increase in density) across the intercooler of substantial ammounts when it was very cold. I took my boost regulation from the cold side, so that things backed off when the temperature rise caught up with the cooling capacity. With no cooling in the intercooler, the actual pressure drop was very small. Two boost gauges always confused folks.

Of course you could change the working fluid. A higher latent heat means more cooling...

Ammonia 1369 kJ / kg

Water 2260 kJ / kg


"Among the many misdeeds of the British rule in India, history will look upon the act of depriving a whole nation of arms, as the blackest."
~ Mahatma Gandhi, Gandhi, An Autobiography, M. K. Gandhi, page 446.

Back to Automotive Air Conditioning Forum

We've updated our forums!
Click here to visit the new forum

Archive Home

Copyright © 2016 Arizona Mobile Air Inc.