HyChill in 2.7 or 3.0

[Graeme]

Is anyone using HyChill in a 2.7L or 3L D4?

I've gone the complete circle chasing why the a/c system only cools while idling, having replaced the compressor control valve (possibly because I powered it on full 12V) and replaced the ATC ecu (likely damaged trying to discover why the compressor control valve wasn't being powered). It seems that the compressor control valve is set to minimum flow once revs are increased a little and flow isn't restored by idling the engine again, but flow is restored by restarting the engine.

From the WSM...
The ATC module automatically adjusts the displacement of the A/C
compressor between the minimum and maximum values, to match the thermal load of the evaporator. By matching refrigerant flow and the thermal load of the evaporator, the ATC module maintains cabin comfort whilst also considering fuel economy.

I'm wondering if the ATC ecu detects a discrepancy in the performance of the system once engine revs exceed around 1000 rpm due to the refrigerant being HyChill rather than the specified R134A so idles the system.

[shack]

Is anyone using HyChill in a 2.7L or 3L D4?

I've gone the complete circle chasing why the a/c system only cools while idling, having replaced the compressor control valve (possibly because I powered it on full 12V) and replaced the ATC ecu (likely damaged trying to discover why the compressor control valve wasn't being powered). It seems that the compressor control valve is set to minimum flow once revs are increased a little and flow isn't restored by idling the engine again, but flow is restored by restarting the engine.

From the WSM...
The ATC module automatically adjusts the displacement of the A/C
compressor between the minimum and maximum values, to match the thermal load of the evaporator. By matching refrigerant flow and the thermal load of the evaporator, the ATC module maintains cabin comfort whilst also considering fuel economy.

I'm wondering if the ATC ecu detects a discrepancy in the performance of the system once engine revs exceed around 1000 rpm due to the refrigerant being HyChill rather than the specified R134A so idles the system.Yes I have.

And the system died, but I've been blaming a partially blocked condenser.

I have also been wondering if the variable displacement compressors don't like the Hychill, but there doesn't seem to be any logical reason for this.

Anyway the compressor destroyed itself and I'm still not sure why. (Actually it might be 2).

I really hope you get to the bottom of this, it's so frustrating, and so difficult to swap parts on these systems. I know I said earlier this thread was giving me PTSD, and I wasn't joking!

[Graeme]

Thanks for your comments. I really think that the ATC module is shutting down the system rather than a blockage because the a/c system never restarts after extended idling but starts cooling as soon as the engine is restarted.

[Graeme]

I should add that the first HyChill fill that showed high low-side pressures and low high-side pressures that prompted the replacement of a suspected blocked dryer was done by 1 workshop. The 2nd HyChill fill was done by another workshop who only had an on/off switch for the compressor control valve fitted by me rather than a working ATC system controlling the compressor. He filled the HyChill recommended quantity but added a little more to get the system as cold as he thought it should be, noting that the compressor was running at maximim capacity. Also, this mechanic stated that he usually has to add a little more, possibly indicating that his measuring equipment might be optimistic.

I'll put the gauges on tomorrow morning. The IIDtool shows static pressure of around 100 psi and around 160-170 psi at ambient 30 deg, and the evaporator temperature around 10 deg with the engine idling, yet we know that the earlier low-side high pressure was around 350 psi. Something seems amiss.

[BradC]

The pressure discrepancy shouldn't be an issue. If I get a chance, somewhere I have an overlaid pt chart. It might be an issue on mechanical displacement control (based on suction pressure) but I can't see how it'd make a difference on the D3/4 system. The outside temperature and airflow would cause more pressure variation than the difference in PT curve for the different refrigerants.

[BradC]

Have you watched the displacement valve on the Gap live values to see if it's actually trying to open?

[BradC]

He filled the HyChill recommended quantity but added a little more to get the system as cold as he thought it should be, noting that the compressor was running at maximim capacity. Also, this mechanic stated that he usually has to add a little more, possibly indicating that his measuring equipment might be optimistic.

Sorry, third post on this one but this is common. HyChill deliberately specify an undercharge compared to conventional refrigerant. I have no idea why they'd do something so dumb, but they do and have done as long as I've been looking at it. Originally I believe the theory was as HCs transfer heat much better in the condenser you don't need as much liquid back up to get the required subcooling, but a more cynical take would be "a smaller charge costs less so we look good".

The real problem with this is there's less liquid available to account for leakage and off-chart conditions, and as HCs do transfer heat so much better by the time you've noticed impairment in the cooling due to leakage the evap is full of oil and the compressor has ground to dust due to lack of lubrication. I have a detailed spreadsheet somewhere where I compared the liquid volume of R12 and R134a against the roughly 60/40 blend of propane and iso-butane that is Hychill on a weight/volume basis. From rough memory it worked out at about 40%, so a 512g r134a charge would be replaced by a 204g HyChill charge on a volume per volume basis (and it's liquid between the condenser and the top of the evaporator, so volume matters. Weight is just a convenient way of measuring liquefied gas).

One of my pet peeves with HyChill.

[Graeme]

My suspicion is that the ATC ecu is shutting down the system because the pressure doesn't rise to anywhere near where it should be (for R134A) for the increased engine revs. I need to check my phone's IIDtool setting that it is reporting all faults for this G2 device which isn't my normal G4 device in case I'm not seeing a low pressure fault.
Late last night my son found the R134A high and low side pressure range chart for various ambient temperatures but only HyChill 30 high and low side pressure ranges without ambient temperatures. I thought that the IIDtool-reported mostly around 10 deg evaporator temperature wasn't low enough for maximum cooling at 30 deg ambient, so another indication that the system was underfilled. Maybe its time to get it filled with the proper volume of R134A, for which there was a 3-week wait prior to the last HyChill fill.

[scarry]

Maybe its time to get it filled with the proper volume of R134A, for which there was a 3-week wait prior to the last HyChill fill.

At least if you have the correct refrigerant,and the correct amount,that eliminates a refrigerant problem,and you can look elsewhere,if there is still a problem.

Hychill does operate slightly higher pressure than 134a,but not much difference at all.

Standing pressures should be similar at similar ambients.

In your earlier post was the Suction or was it the Discharge pressure at 350PSI?

https://hychill.com.au/media/pages/info/3910c8ea3d-1662605920/hychill-pressure-temperature-chart-automotive.pdf

[Graeme]

Evaporator discharge was high. The low side was high suggesting no path through the condensor then dryer to the compressor.
Yes, using R134A would remove refrigerant from the equation.

My main tractor's R134a a/c isn't cooling much at all, discovered a couple of days ago after it hasn't really been used for 3-4 months. It was topped-up 5 years ago. Its pressures are ball-park with the D4 this morning at 20 deg ambient.

D4 static 80, idling low-side 60, high-side 94, evap temp 12.5 deg
tractor static 96, idling low-side 58, high-side 109

It seems to me that both are low on refrigerant even though the D4 produces cooler air while idling.
The tractor's high side valve must leak very slightly as there was a small pop as the cap was loosened, but presumably not an issue being 5 years since the last top-up.

The person who comes to do the farm a/c declined to do the D4 after the engine swap because he didn't know how Land Rover a/c works. I wonder if I can convince him to do the D4 with R134a when I get him to do the tractor.

[Graeme]

The a/c is finally working properly, with evap temp hovering between -0.5 and 0 deg on temp setting Lo, fan on max and engine up to 3000 rpm and the pressure reaching 180 psi. I had earlier wriggled the pressure sensor connector and wires where there must have been a contact issue. With temp set to 21 deg and middle fan speed, the evap temp rose to 7-8 deg, but began dropping again when the temp was set back to Lo.

[scarry]

Low side pressure of 350PSI means something is drastically wrong.

If you can con the AC guy to give it a good vac and put in the exact amount of R134a all should be good.
As long as the dryer is the same size as the original,sometimes if it isn't the measured weight can be not quite correct.

[Graeme]

The condenser with attached dryer that is a copy of the original was fitted and since then the low side pressure has always been lower than static when the system is cooling, so that 350 psi situation is now history.

Unfortunately the good working system was short-lived. I think there is a wiring issue, possibly with the connector next to the battery in which the a/c wires from the engine harness are located. Alternatively there could be an issue with the engine fan for which I haven't yet found the actual speed in the Gap tool, only the commanded duty cycle although I haven't moitored it yet as I keep loosing my live value selections when I try to save the selection. I also have yet to connect my little scope to the compressor valve feed to confirm that the ATC is no longer sending a high duty cycle signal trying to open the valve once the system stops cooling.

Unfortunately or otherwise its too windy for me to get back to my urgently-needed weed spraying, so I'll keep at it for another hour or so.

[Graeme]

Unfortunately the ATC module is supplying full battery voltage to the compressor control valve, excepting for regular short intervals of approimately 5% duration of 0V. This, with the observed high-side and low-side gauge pressures, means that the control valve isn't increasing the compressor capacity as required by the ATC module. The valve must be increasing capacity slightly as some pressure increase is observed along with slight cooling at the evaporator, moreso as engine revs are increased.

The ATC power to the control valve ceases whilst ever the pressure sensor connector is unplugged, but resumes immediately the connector is refitted. The ATC power to the valve is unaffected if the engine fan's connector is unplugged.

Hopefully we can arrange to have the system thoroughly flushed to clean it out without removing any components.

Edit: I'll try switching the a/c on and off at higher engine revs to see if the bursts of higher pump volume can dislodge any contamination of the valve.

[BradC]

Edit: I'll try switching the a/c on and off at higher engine revs to see if the bursts of higher pump volume can dislodge any contamination of the valve.

My experience with VW control valves is once they're done, they're done. The valves have fine screens on the inlet side. If stuff has gone in there to jam the valve, then it's done.

[Graeme]

These valves have a screen around the outside but the symptoms don't indicate to me that the screen is permanently almost fully blocked.

Switching the a/c on then off with revs eventually caused pressure to build, up to 210 psi around 3000 rpm but easily 180 psi 1500+ rpm. However going up and down in revs would eventually cause pressure to drop right back and evaporator temperature rise indicating that the valve was no longer being opened, but it was quite easy to get pressure again. I'll have another stint later to see if I can clear whatever is causing the pressure fault.

[DoubleChevron]

Can you just remove the valve on the compressor and make it run flat out. I purchased an old jaguar a few weeks back, its A/C wouldn't work (it was quite weird, if I grabbed the high side it was burning hot .... and the suction side was hot rather than cold.... bizare... I figured the compressor must be cactus. Anyway I put a message on the jaguar forum before I started spending $$$$ and was rewarded with the longest bloody diagnostic post I've ever seen on the cars ventilation. At the bottom of it was this

http://www.jagrepair.com/images/TSB/S-Type/SType%20Does%20Not%20Cool%20Sufficiently.pdf

Its a scroll type compressor and jaguar themselves must have realised the efficiency valve/crap on it.... was well crap. So I dumped the gas and asssembled the "efficiency" valve backwards as suggested, dumped some hychill in it and was rewarded with freezing cold A/C. So i went from thinking thousands for a repair to pocket change for a squirt of hychill :)

The more modern a car, the more crap it is when it comes to "efficiency"/pollution crap that doesn't work and breaks :(

seeya
Shane L.

[Graeme]

Thanks for the suggestion and link. However the solenoid valve LR06164 is a sealed unit that is retained by a circlip.

[BradC]

http://www.jagrepair.com/images/TSB/S-Type/SType%20Does%20Not%20Cool%20Sufficiently.pdf

Its a scroll type compressor and jaguar themselves must have realised the efficiency valve/crap on it.... was well crap

No, it's a suction pressure activated valve and perhaps Jaguar found their implementation was crap so they put out a TSB to bypass it. In all honesty they probably ripped it off from GM who had been doing it badly for years. It's almost essential to prevent evaporator freezing under low load.

The original concept was the compressor displacement would operate to maintain a constant low side (suction) pressure. This would provide a constant evaporator temperature at whatever was required to keep it cold enough and not freeze. The world finally figured out this method was ****, so manufacturers moved to a more complex (and expensive) evap temperature sensor that went through a pile of electronics to then drive a displacement valve (solenoid) to modulate the compressor displacement.

They would not have moved to an electronic control if the in-compressor method was workable. So your car got caught in the bit "in between". It's not relevant to any electrically modulated vehicle. Full stop.

[Graeme]

The mobile R134a man is coming the day after tomorrow to do the tractor. I'll be quizzing him regarding using Hychill in a Sanden R134A compressor that has the displacement control valve in case he knows of any issues with the valve sticking, which is what I think is happening.

[DiscoJeffster]

The mobile R134a man is coming the day after tomorrow to do the tractor. I'll be quizzing him regarding using Hychill in a Sanden R134A compressor that has the displacement control valve in case he knows of any issues with the valve sticking, which is what I think is happening.

Why not just get him to change it to 134 while he’s there and lose the guesswork?

[BradC]

I've used HyChill in VAG Sanden variable displacement (and conventional fixed displacement of many different makes) compressors for years. The only possible thing I can think of is the hydrocarbon thins out the PAG oil to a point it loses lubrication, generates some Aluminium grunge that makes its way around to jam the valve. Actually, in most cases it's even mobilised the lubricant enough to staunch compressor shaft seal leaks that were prevalent with R134a.

Having said that, every system I've put HC in had a stuffed valved from r134a already. In most cases the HC extended the system life a few years before the valve needed replacing. Maybe it did that by affecting the lubricity of the oil, but I'd be surprised.

The recommendation is to remove/flush the PAG oil and replace it with a POA or Mineral oil if you are moving to HC's, but probably only because PAG is a **** lubricant that is horrifically hygroscopic. The only reason they use that crap is because it's one of the only lubricants that r134 will actually move.

[Graeme]

Why not just get him to change it to 134 while he’s there and lose the guesswork?
He declined a request to charge the D4 in October because he doesn't know Land Rovers, so I will be attempting to get him to agree this time while on site. He may not come with the required 600g of R134a to spare or he may be too busy on the day to stay the extra time, but I will do my best to convince him that I need the Hychill replaced by R134a. I can show him live values with the IIDtool - I've read the manual (if all else fails then RTFM) and now have a favourites defined to get to the display quickly.

[BradC]

He declined a request to charge the D4 in October because he doesn't know Land Rovers

It is just a vehicle A/C system, same as any other modern vehicle. There is _nothing_ Land Rover specific in there. It's a Compressor, Condenser, Dryer, TXV & Evaporator (in that order). No different to anything else. Heck, swap the TXV for a capillary tube and it's just like your fridge. Just vac it down and drop the measured charge of gas in there.

The high side pressure sensor is there for protection only. Outside of protective extremes the displacement valve is controlled to maintain evaporator temperature, the high side pressure isn't really involved. No different to any Volkswagen from the last 20 years either.

[Graeme]

I suspect that he hasn't worked on an a/c system in a modern vehicle and if not then I hope to educate him about the displacement valve. I consider that I now know enough about the D4's a/c system to get him to trust me. He only has to vacuum out the system then fill with the stipulated weight of R134a and the appropriate quantity of oil.

[BradC]

Be careful with the "oil" thing. Unless you've pulled out the compressor, condenser and evaporator and drained them into a measuring cylinder then you have very little idea how much actual oil is in the system. Most manuals will say, put oil in the compressor, install and run for x minutes, then pull the compressor and drain it. There will be x oil in the other components.

That was easy to do on old cars where getting the compressor out, or getting a dipstick into the top fill hole was easy/possible. Probably not so much on the D3/4. Provided you didn't completely flush all the old oil from the condenser (which you put in new) and evaporator, there should be enough oil in the new compressor. They come "oiled up". Usually the book says "drain the oil from the new compressor and only put back as much as you drained from the old compressor". Don't go adding a significant quantity unless you know you lost a pile.

Again, too little is always bad and too much is inefficient. To a point. Don't add oil unless you really know you need to.

[Graeme]

I didn't drain the new condenser/dryer. The fitted plugs were removed once the condenser was installed and the pipes which had been plugged with clean rag were then quickly attached.

Considering what you wrote earlier about oils, I suspect that the oil added when the Hychill was installed might not be compatible with R134. Does this mean that the system has to be flushed before vacuuming?

Edit: When I released the Hychill prior to changing the valve, I released it slowly into a tall plastic jug so as to catch the oil, which wouldn't have been more than about 10cc.

[BradC]

Considering what you wrote earlier about oils, I suspect that the oil added when the Hychill was installed might not be compatible with R134. Does this mean that the system has to be flushed before vacuuming?

Might be worth asking the question because you never know what they might have added, but to be honest the POA oil HyChill recommend / supply (SRO500) is workable with pretty much anything. +/- 10ml is not even worth worrying about.

I use mineral oil because I have it and if it was good enough for the Dinosaurs it's good enough for me and i don't tend to sell cars, I just run them into the ground. The few cars I have sold have had comprehensive briefings with the new owners.

[DiscoJeffster]

If I recall, the condenser is worth 30ml of oil and the drier is much the same - rough guide on how much oil to add when replacing components. The condenser does not come with any oil in it. When you changed the condenser, you should have replaced that much oil back into the system.

You may find you are now running lower oil quantity than required.

Edit: found it’s approx 30ml for each of these component or an ounce, as a rough guide

[scarry]

If I recall, the condenser is worth 30ml of oil and the drier is much the same - rough guide on how much oil to add when replacing components. The condenser does not come with any oil in it. When you changed the condenser, you should have replaced that much oil back into the system.

You may find you are now running lower oil quantity than required.

Edit: found it’s approx 30ml for each of these component or an ounce, as a rough guide

The best thing to do is drain the oil out of the old component,if possible and that will give you a good idea what is going on.

Without doing that it is all guesswork.

We do this on larger systems,although most have a compressor oil level glass as well.

[Graeme]

The system is now charged with 600g of R134a and 30ml of oil added but the compressor is not producing any pressure. Yesterday afternoon I had a last try with the Hychil with the same result, so it appears that the new valve has gone the way of the original and that the R134a shouldn't have been installed. Furthermore we couldn't hear the=is valve clicking when 12V was applied while the engine wasn't running (the original was easily heard), yet a small spark was produced when touching the power wire on and off and the coil still has its original 12 ohms resistance. Hence the valve's internals are considered stuck. I'll have a few more attempts to get it to work then get the person back to recover the R134a so that the valve can again be replaced.

The original filter/dryer indeed had a very blocked gauze with small speckles indicating that the original compressor was failing, but not surprising after 300K kms. There was a slight trace of similar grey material on the inlet gauze on the original valve.

The system held good vacuum prior to installing the R134a.

It was suggested that a genuine Sanden valve be fitted next time, but from LR is nearly $300 compared to $155 for the Jayair from a local a/c parts outlet. However is the Jayair version likely to be any better than a $30 + $10 frt one from AliEpress, noting that the AliExpress one has R134a on its label?

I'm inclined to buy the ALiExpress one then remove the compressor when the valve arrives, even though the current one's failure could be covered by warranty.

[DiscoJeffster]

This has turned into quite a nightmare, as I say from my privileged perch.

Can you just simply send back the entire compressor as faulty and get it replaced under warranty? Might be a tall ask, and I’m sure you’ve considered it.

[Graeme]

No, I can't send it back at least because HyChill rather than the stipulated R134a was used.
I'm hoping that the seals in the valves can't tolerate Hychill and is the reason why the compressor has R134a on its label. Seeing the Aliepress valve also with R134a on ite label got me thinking about how brake fluid swells normal seals which is what I now suspect has hppened to both valves rather than clogged with debris.

I managed to get the system cooling partially 3 times with R134a so I'll keep trying until the 3rd valve gets close and not have the R134 recoveed in case the seals come good.

[scarry]

No, I can't send it back at least because HyChill rather than the stipulated R134a was used.
I'm hoping that the seals in the valves can't tolerate Hychill and is the reason why the compressor has R134a on its label. Seeing the Aliepress valve also with R134a on ite label got me thinking about how brake fluid swells normal seals which is what I now suspect has hppened to both valves rather than clogged with debris.

I managed to get the system cooling partially 3 times with R134a so I'll keep trying until the 3rd valve gets close and not have the R134 recoveed in case the seals come good.

It is usually some oils that swell the seals,not the refrigerant.
Although these days,and for decades,all seals have been fine for all oils used in Refr/AC,uncluding Auto air.

Good luck with it,it sure has been a headache[bigsad]

[Graeme]

Can a compressor be damaged by very high low-side pressure? A blurry picture of the gauges when the dryer was blocked shows the low side pressure needle more than 3/4 the way round the dial.

I recall seeing a picture or drawing of reed valves inside a compressor and wondered if they are susceptible to damage or if anything else can be damaged such that the compressor can only occasionally be made to work? However such damage wouldn't seem to explain why a new valve works whilst the engine is idling but not when revs are increased.

The a/c mechanic said today that he would replace the complete compressor rather than just replace the valve, but there's a big difference in cost and his work is almost entirely with fixed capacity compressors as fitted to heavy equipment.

Edit: It's time for a new compressor.

[Graeme]

Fingers are crossed for a good result with another new compressor which has been penciled-in for R134 Monday a/n.

When turning the latest compressor by hand prior to it being installed so the plugs were still fitted, minor pulsing / compression could be felt, but none from the removed compressor with the plugs having been transferred after the new compressor was installed. I had turned both compressors by the front of the pulley but tried the old one again this time with the clutch engaged to be able to turn the pulley more quickly, but still zero compression. Is it normal to be able to feel compression when turning by hand?

[scarry]

Is it normal to be able to feel compression when turning by hand?

Yes,with any belt drive or coupling style compressor,compression should be felt when turning by hand.

But i dont know how the valve operates,could it be stuck in the unloaded position,then you may not feel compression?

The reason your suction guage was reading a high pressure is because the compressor is not pumping,ie,broken internally,or maybe unloaded?
That is as long as the clutch was engaged properly.

Anyway,good luck with the new one,hopefully it is all now sorted.[smilebigeye]

[Graeme]

The pistons have nil stroke when the valve isn't energised, but perhaps assembly-time oil in the control section was providing a little and me looking for something positive.

[BradC]

Generally (and I say that with only the knowledge gleaned from manufacturers manuals and data sheets) a variable displacement compressor will stop at about 5% displacement. The purpose is to continue the movement of refrigerant through the system which carries oil. If it was to drop to 0% then there would be nothing to lubricate the moving bits as they'd be still rotating but the oil wouldn't be moving around. These compressors don't have a lubrication system as such. The refrigerant picks up oil in the sump and moves it about. No refrigerant movement, no lubrication.

I'm fairly convinced that's why LR decided to add a clutch to the compressor on the D4. No rotation, no requirement for lubrication. Every other vehicle I've seen with a variable displacement compressor ensures it moves oil around while it's rotating.

Hydrocarbons pick up and move lubricant orders of magnitude better than r134a (which is frankly a **** refrigerant anyway and terrible with oil, which is why they need crap like PAG because it's the only oil that crap will move around). Your theory on seals is plausible. I have a recovery machine that I bought from China many moons ago. I do a lot of work with hydrocarbons and they didn't do any favors to the o-rings in the valve system on my machine. I replaced those with Viton o-rings and never had an issue after. My point is it is potentially possible that the seals in the compressor displacement valve were of a material that was/is fine with r134a/PAG but not so much with HC's. Having said that, if the valve seals went bad I'd also expect the shaft seal to fail in the same way. These compressors seem to use lip seals vs the old carbon face seal. They all leak, but lip seals with the right lubricant seem to leak less (at least during the warranty period).

On the other hand, I've used HC's for years in VAG variable displacement Sanden compressors and had no issues, so I'd be loathe to blame the refrigerant.

I'm watching with interest. I still use r134a in my D3, but I had been considering switching next re-gas.

[Graeme]

The A/C system is now working as it should. It was working with 400g of R134A albeit indicating that it needed more refrigerant then topped-up to 600g.

The mechanic dismantles compressors to diagnose faults, including Sanden compressors of which he has quite a high opinion. He will dismantle the removed one, possibly tomorrow, then let me know what he finds.

[scarry]

Generally (and I say that with only the knowledge gleaned from manufacturers manuals and data sheets) a variable displacement compressor will stop at about 5% displacement.

Hydrocarbons pick up and move lubricant orders of magnitude better than r134a (which is frankly a **** refrigerant anyway and terrible with oil, which is why they need crap like PAG because it's the only oil that crap will move around).

In unloaded situations,there is also less refrigerant velocity,so oil return can be an issue.

In our game,we never have issues with oil return using R134a,there are much worse refrigerants around,such as R22,particularly at lower SSTs.
Even with 50/50 POE/Mineral oil,134a is fine.

[BradC]

I always thought R22 was pretty good. It'll transport any oil, it just doesn't cope with high superheats so is lousy in a LBP scenario. For A/C it's a ripper. Not as good as R290, but then different strokes.

My leak detector is much more sensitive to HCFC than HFC so I keep a bit of R22 around to pop into a system and then back up with a pile of nitrogen for those bizarre and really hard to locate leaks (where nitrogen and Big-Blu just doesn't cut it).

[scarry]

I always thought R22 was pretty good. It'll transport any oil, it just doesn't cope with high superheats so is lousy in a LBP scenario. For A/C it's a ripper. Not as good as R290, but then different strokes.

My leak detector is much more sensitive to HCFC than HFC so I keep a bit of R22 around to pop into a system and then back up with a pile of nitrogen for those bizarre and really hard to locate leaks (where nitrogen and Big-Blu just doesn't cut it).

We haven't really noticed the different senitivity with ours,we use the Fieldpiece D82,find them excellent,but get worn out quickly when they are used almost every day.

Just picked up one for HCs,at huge cost,it is a DTEK STRATUS.

Something like $2.5K,including the rediculously expensive sensors.

[BradC]

Pah, easiest leak detector for HC's is a lighter! I joke.... mostly.

I'm still using an old heated diode unit. If I were relying on it for a crust I'd certainly upgrade but the sensitivity difference between HCFC and HFC is at least an order of magnitude on my old clunker. I've kinda looked at both the IR and ultrasonic units over the years, but as I have to slip it past the war office it's a hard sell for the couple of times a year I break it out.

I did have one leak that was a complete bastard to find. I ended up wrapping the whole pipe segment in gladwrap and then just poking holes in it to stuff the leak detector into. Turned out to be a smaller than pinhole in a brazed joint. Not even big-blu bubbled up, but after half an hour in gladwrap it showed itself.

[scarry]

Pah, easiest leak detector for HC's is a lighter! I joke.... mostly.

I'm still using an old heated diode unit. If I were relying on it for a crust I'd certainly upgrade but the sensitivity difference between HCFC and HFC is at least an order of magnitude on my old clunker. I've kinda looked at both the IR and ultrasonic units over the years, but as I have to slip it past the war office it's a hard sell for the couple of times a year I break it out.

I did have one leak that was a complete bastard to find. I ended up wrapping the whole pipe segment in gladwrap and then just poking holes in it to stuff the leak detector into. Turned out to be a smaller than pinhole in a brazed joint. Not even big-blu bubbled up, but after half an hour in gladwrap it showed itself.

I could write a book about where and how hard some leaks were to find and where they were.....

One i can remember was a little refrig unit,i endeded up bringing it home,taking all the electrics and fan motor off it,and putting it in the pool,it was a pin hole on the shell of a Danfoss drier,corroded from the inside.
No sign of any oil,which is often the case on the liquid line anyway.

Sometimes,with say evap coils,even pumped up to 300PSI,and put in the pool.there is only the tiniest of bubbles every 15 to 30 seconds or so,and that is enough to cause a problem.

[Graeme]

Notwithanding that the original 320K km compressor was subsequently found to be on the way out when it was replaced and therefore the condenser/dryer should also have been replaced while the body was off which would have saved 1 recharge, the total cost of the exercise was well over $3K in parts and professional labour.

[BradC]

Ouch! That's a heck of a bill. I don't even think I know a duck that big.

I'd be looking at that compressor teardown and if it looks like it has been affected by the hydrocarbons I'd be having a bit of a chat to HyChill. I'd be surprised, but I can also imagine that someone, somewhere at some-point made a "cost down" decision and changed a seal material.

Yes, when a compressor is failing it usually drops debris into the system, and the modern condensers are so intricate they usually act as a bit of a filter and catch it, so if you are doing a compressor a condenser/dryer should be standard because there's no practical way to properly flush them. In the old days with cast iron internals it was usually just a broken ring, valve or leaky carbon seal but these newer alloy compressors love to grind themselves to paste.

Looking on with morbid curiosity.

[shack]

And I still want to know if hychill is ok in the D3 compressor!

[Bohica]

And I still want to know if hychill is ok in the D3 compressor!

from Grok

No, HyChill (a hydrocarbon refrigerant blend like Minus 30, mainly propane/isobutane) is not recommended or reliably "OK" for the Land Rover Discovery 3 (LR3, 2004–2009) AC compressor. While some people use it as a cheaper, lower-pressure drop-in alternative to the factory R134a, real-world experiences in Discovery 3/4 models show frequent issues, and it carries safety, compatibility, and warranty risks.

The reasons why!
And it references AULRO and Shack's post!!


Factory Specs for Discovery 3 AC System



The system uses R134a refrigerant (not R1234yf, which came later on newer Land Rovers).
Charge amount is typically 550–600g (without rear AC) or 810–900g (with rear evaporator), depending on engine (e.g., 2.7 TDV6 or 4.4 V8) and exact configuration. Oil is usually PAG46/ISO 46, ~130–180 ml.
Compressors are often variable-displacement types (e.g., Denso or similar), electronically controlled by the ATC (Automatic Temperature Control) module for efficiency and comfort.

HyChill claims compatibility with R134a systems/compressors, lower operating pressures (less load on the compressor), better efficiency, and only ~30–40% of the normal charge weight. It works with existing oils (or their SRO 500) and requires proper evacuation/flushing.
Reported Issues in Discovery 3/4 (and Similar Land Rovers)

Discovery owners (especially on AULRO forums) have shared mixed but mostly problematic experiences:


Poor cooling and system shutdowns — The ATC/ECU or compressor control valve often detects mismatched pressures/temperatures (HyChill has a different pressure-temperature curve than R134a) and reduces or shuts off compressor displacement, especially above idle. Cooling may work OK at idle but fails at higher RPMs. Restarting the engine sometimes resets it temporarily.
Compressor damage/failure — At least one user reported the compressor "destroying itself" (possibly twice) after switching to HyChill. Others note risks of inadequate oil return due to the much lower charge volume, leading to wear, contamination, or seizure. Variable-displacement compressors seem particularly sensitive.
Pressure anomalies — Examples include abnormally high low-side or low high-side pressures, blocked dryer suspicions, or evaporator temps not dropping enough for strong cooling in hot conditions.
Other concerns — Leaks may be faster through standard hoses/seals (HyChill recommends full-barrier hoses for hydrocarbons). Some mechanics refuse to service systems with it, and warranties on compressors/parts are often voided.

In one recent thread, a user directly asked: "And I still want to know if hychill is ok in the D3 compressor!" after expensive repairs — highlighting ongoing uncertainty and frustration.
Safety and Other Risks



Flammability — HyChill is highly flammable (A3 class, like LPG). The EPA strongly advises against hydrocarbon refrigerants in vehicle AC systems due to fire/explosion risk in a crash or leak (especially near hot components or in an accident). While the total charge is small (~150–300g), some view it as an unnecessary added hazard compared to R134a. Real incidents (e.g., fires in other HC systems) have been cited in discussions.
Legality/Regulations — In some regions (e.g., parts of Australia like QLD historically), there are restrictions. Many shops won't touch it, and it may violate vehicle regulations or insurance terms. It's not EPA-approved for MVAC (motor vehicle air conditioning) in places like the US.
Long-term reliability — Lower charge can affect oil circulation; improper installation (no full flush, wrong oil, no new dryer/TX valve) often leads to failures blamed on the gas.

HyChill promotes it heavily for reduced compressor load and eco/ cost benefits, and some users in older R12/R134a cars report good results with no issues. However, in electronically controlled modern systems like the Discovery 3's, the consensus from owner forums leans negative.
Recommendation

Stick with proper R134a for reliability, proper compressor/ATC behavior, safety, and serviceability. If your system needs work:


Have a qualified AC specialist recover, pressure-test/vacuum, replace the receiver-dryer and any faulty parts (e.g., TX valve), flush if contaminated, and recharge to spec with the correct PAG oil.
If you're set on trying HyChill (e.g., for cost in a high-mileage vehicle), use a shop experienced with it, follow their exact procedure (flush, correct reduced charge by weight, barrier hoses if advised, new labels), and monitor pressures/temps closely — but be prepared for potential compressor issues or cooling inconsistencies.

If your compressor has already failed or you're seeing specific symptoms, it might be unrelated (common D3 AC problems include leaks, clogged dryers, or control valve issues), but switching refrigerants without addressing root causes often makes things worse.
For your Melbourne location, local Land Rover specialists or reputable AC shops can confirm with your exact VIN and test the system. Always prioritize safety over short-term savings with flammable alternatives in a vehicle AC. If you share more details (e.g., symptoms, current refrigerant, compressor type), I can help narrow it down further!

[BradC]

from Grok

So a complete pile of google accumulated bull**** with absolutely zero evidence and 100% speculation.

Perfect example is "The ATC/ECU or compressor control valve often detects mismatched pressures/temperatures (HyChill has a different pressure-temperature curve than R134a) and reduces or shuts off compressor displacement, especially above idle. Cooling may work OK at idle but fails at higher RPMs. Restarting the engine sometimes resets it temporarily."

Which is total and utter horse hockey, because the "ATC/ECU" has no ****ing idea what the P/T curve is, no idea what the low side pressure is other than the evap temperature, and no idea of the system capacity. In fact all the ATC/ECU knows is "it's either getting cold, or the high side pressure is too high and I need to shut down".

The whole thing is a steaming pile, with zero evidence and a crapload of speculation. Let's wait until Graham has a definitive answer from the compressor teardown. I've used HCs in variable displacement compressors for nearly 15 years. Same compressor manufacturer, different vehicle manufacturer. Same compressor, different end cap and pulley. In fact I've used them in Sanden, Denso, Tecumseh & Diesel Kiki with no issues.

This bit intrigues me, because as far as I know I'm the only person to ever mention it on either LandRover or Volvo forums because I've calculated the fluid volumes vs charge weight recommendation and warn about the potential for oil return when leakage inevitable happens : "Others note risks of inadequate oil return due to the much lower charge volume, leading to wear, contamination, or seizure. Variable-displacement compressors seem particularly sensitive.".

My comment on that is that HyChill recommends a smaller charge on a volumetric basis than the original HFC charge which leaves little room for leakage. I'm not saying if you use the recommended HyChill weights you'll lunch your system, just that there's less room for error. It's not "much lower charge volume", it's low 10's of percent.

Thus demonstrating AI is crap. Stop posting that **** please.

[Graeme]

Pressure anomalies — Examples include abnormally high low-side or low high-side pressures, blocked dryer suspicions, or evaporator temps not dropping enough for strong cooling in hot conditions.

HyChill was never blamed for nor was the cause of the blocked dryer, which was in fact caused by the end-if-life original compressor. This misquoting is what disgusts me about so-called AI.

[RANDLOVER]

IIRC the guy who topped up my a/c suspected a really tiny leak, and tried to find it by charging it with nitrogen with a hydrogen trace element and the associated sniffer to no avail. The last top up seems to have lasted 5 years.

My mechanic asked if he replaced the drier but it's unclear from his invoice.

[Graeme]

The fault with the 1st new compressor was that the pistons and bores were badly scored and the piston rings were loose in the cylinders. The swash plate and valves were in good condition.

Presumably debris from the failing original compressor found its way into the new compressor from the mostly blocked dryer, which would have been subjected to the higher pressure at which HyChill runs than the pressure at which R134a runs. Had R134a been used then the lower pressure might not have pushed the debris through the dryer. Had the condenser and dryer been replaced with the compressor then this would not have happened - a lesson learnt.

[scarry]

The other issue is often a fair bit of debris and dirty oil from a failing compressor can often get into the evaporator.

The evaporator should be flushed out before a new compressor is fitted, as well.

Suction line dryers can also be used,but would be difficult to fit in most Auto Air systems.

[BradC]

The fault with the 1st new compressor was that the pistons and bores were badly scored and the piston rings were loose in the cylinders. The swash plate and valves were in good condition.

Presumably debris from the failing original compressor found its way into the new compressor from the mostly blocked dryer, which would have been subjected to the higher pressure at which HyChill runs than the pressure at which R134a runs.

Sorry, but nah. The difference in pressure would be similar to the difference between running the A/C on a 25C day and 30C day. Seriously, that's the sort of difference we're talking about. So that's not Hychills fault, that sounds like a system full of destroyed compressor debris that found its way back into the new compressor. The thing to do is take the old compressor and drop the oil into a beaker. *any* sign of sludge and/or "sparklies" and you know the system is toast. If you put a new compressor in without doing a condenser replacement & flush the rest then you're setting yourself up for failure regardless of the regrigerant.

If I get a chance I'll knock up a comparison PT chart. We're talking 10's of PSI at most.

One thing HyChill would do is help scrub the system clean. Hydrocarbons are brilliant at moving any oil and debris about, so if the evap was full of crap it would have washed it into the compressor sump. r134a would have done the same thing, it just would have taken a lot longer.

[BradC]

Ok, Blue is effectively HyChill (60/40 Propane/Iso-Butane). Orange is r134a.
I've kinda mixed units. Y Axis is PSI and X axis is Bubble point in Centigrade (HyChill has a different Bubble & Dew point as it's a blend, r134a is consistent). Nothing like mixing Metric and "Freedom units".
These were derived from NIST RefProp.

So the pressure difference is there, but not enough to damage a compressor. Now if you'd put pure propane in, then we'd be having a different conversation.

Edit : Dunno where my image went, let's try again
196000

Edit : Just for ****s and giggles, I re-created the graph with Propane as the Yellow line. Don't do that.

[Graeme]

The suggestion regarding the slightly higher HyChill pressure was that it may have pushed more debris through the filter in the dryer, not that the extra pressure would have any effect on the compressor.

If the initial recharge had been done by a proper a/c mechanic/workshop instead of a local general mechanic who didn't have much a/c knowledge then hopefully the advice would have been to replace the condenser/dryer and the evporator would have been flushed prior to recharging. The local mechanic was used because he could do the recharge within a few days rather than 3 weeks quoted by the a/c workshop which was quite important at the time as a possible buyer was waiting for the a/c to be fixed. The a/c workshop was the only R134a option in the area once the mobile a/c mechanic declined to work on a Land Rover.

[BradC]

Yes, my point is the "slightly high pressure" is so slight as to be insignificant. More likely is the better solvency of the hydrocarbons helped move the debris. Regardless, if that was the cause of the compressor failure it would have failed the same way with r134a, just maybe not quite as quickly.

[Graeme]

I suspect that the new compressor was able to force debris though the dryer's filter whereas the old compressor couldn't due to being worn-out. Unfortunately the low-side 400+psi wasn't immediately recognised as a blocked filter/dryer indicating that the compressor should not be run until the filter/dryer was replaced that perhaps a proper a/c mechanic would have recognised.

The mobile a/c mechanic on both instances of charging with R134a only partly filled the system (400g) to then check whether or not the system was functioning properly, allowing that it was low on refrigerant. I feel that had this approach been taken with the 1st recharge then the compressor may have survived.