Evap freezing at low fan speeds

[thp]

Year: 1993
Make: VW
Model: Eurovan
Engine Size: 2.5, front and rear AC
Refrigerant Type: R-134A
Ambient Temp: 80F

System cools great. However the compressor cycles every ~10 seconds on front evaporator anti-ice switch when blower is set below full blast. Basically the low side drops too low when the blower is reduced. Problem doesn't occur when rear AC is also running. I tried changing the TXV with another, but no improvement. I'm wondering if the compressor control valve could be at fault.

Here's the performance I'm seeing:

1500 RPM, max blower, rear AC off:
Vent: 38 F
Low: 27 psi
High: 245 psi

1500 RPM, min blower, rear AC off:
Vent: 36F
Low: 19, drops further as evaporator begins to ice up (Needed to bypass anti-ice switch to keep compressor running)
High: 220

1500 RPM, max blower, rear AC running:
Vent: 50F (front) 54F (rear)
Low: 35
High: 250

1500 RPM, min blower, rear AC running:
Vent: 47F (front) 49F (rear)
Low: 26
High: 245

Here's the history: I fixed leaks in the system and replaced some components:

- New front and rear TXVs
- new receiver/drier
- replaced original Sanden compressor with new Denso 471703
- I made sure only as much oil went in with the new compressor as came out with the old
- weighed in correct 1350g charge
- Replaced evaporator temperature switch and verified that new one is opening at about 32F.
- Verified that condenser fans are running correctly

Old compressor had not burned out. I flushed the evaporator and the effluent looked clean. When system is running I don't see bubbles in the sight glass and the liquid refrigerant looks clear. The evaporator fins are clean.

One issue: Right after installation I slugged the new compressor for a few seconds when I accidentally opened a valve on the manifold gage, short-circuiting high side to low. I'm wondering if this could have damaged the control valve? (How would I check this?)

Thanks for any suggestions!

[JohnHere]

Your varying pressures and vent temps are certainly wacky. Are you sure that the charge is still correct and that you don't have another leak someplace? I suggest recovering the charge to verify that first.

[thp]

The charge is dead on. I verified a couple times, including when I changed the TXV.

What part looks wacky? I would think the changes from turning on or off the rear AC are expected. When it's on a solenoid opens and you get the pressure drop across the rear TXV and evap added in parallel to the front.

The high side moves maybe 20 psi when the radiator fans cycle between medium and max speed as the engine coolant heats up. I would think that's also expected. Though I tried my best to capture the measurements under similar conditions (radiator fully warm, only checked when the fans were running max, etc).

Seems like the problem is that the low side is dropping too low when the fan's turned down. Isn't it (mostly) the job of the compressor control valve to regulate the low-side pressure when that happens?

Thanks for the help.

[Tim]

Maybe a thermostat issue. Is the coil freezing up? Not 100% sure on the cycling of these, electronic or pressure. The Sanden compressor on these does not have a control valve.

[thp]

This system is not meant to cycle. The service manual claims the compressor has an internal pressure regulator, which I assumed was an internal control valve? I haven't looked closely at the compressor recently, but I seem to remember a circular nubbin that looks like one. I'm not running a Sanden, I'm running a replacement Denso 471703, which the parts house said was compatible. (But maybe it really isn't???)

The only pressure switch is an over/under safety switch on the high side line next to the sight glass. That switch is not opening.

The cycling I'm seeing is from the anti-ice switch on the front evaporator. It declutches the compressor when it detects freezing. But it's just meant as a safety. It's not supposed to kick in normally, and causes short cycling when it does because it has very little hysteresis.

Yes the front evaporator is freezing. I verified with a thermocouple that it's dropping to about 30F when the temperature switch opens.

Apart from the front anti-ice switch, the only thermostat (and really the only electronic control) in the system is for the rear evaporator--it opens and closes the rear evap solenoid to regulate rear evap temperature. There's no problem with that part of the system.

Thanks for the response.

[Tim]

Is this a US vehicle? As all my listings show this to have a Sanden. Which do not have a control valve for this model compressor? I don't have a system break down on your vehicle anymore. I would think there is some sort of cycling device. Especially since it uses Expansion blocks to regulate refrigerant flow. Are you using R12 expansion blocks with R134a?

SD7H15 - 6 POLY CLUTCH

https://www.ackits.com/index.php?route= ... 14-2826NEW

What help has your parts house been? You have a ton of data have they provided any feedback on the products they sold you?

[thp]

Yes, US vehicle. And yes the system was originally fit with a Sanden SD7H15 as you show. A number of online parts houses show the Denso as being compatible, but who knows. I didn't buy it from a specialty AC parts shop (I didn't know about shops like this at the time...) and this is a subtle issue so I haven't leaned on them for support.

Both TXVs are the same. They're for R134a. They're the correct part according to the VW parts catalog. Pic attached. The only questionable thing is that they're a German brand, but Chinese manufacture (the OEM parts are no longer available). Swapping in a different 3rd party valve (also Chinese) results in basically the same behavior. If anyone has a lead on one that's more reputable I'd gladly give it a shot...

The goofy thing is that the book claims that the compressor is variable displacement and will regulate pressure. But maybe the book is wrong? It wouldn't be the first time I found a mistake. See the note at the bottom of the pressure table attached.

The system diagram is also attached. There's no cycling switch on the low side, just a pressure switch on the high. According to the book it turns off the clutch below 29 psi or above 232 psi.

Given that there's no cycling switch, and if the compressor really doesn't have any kind of pressure regulation built-in (and after staring at the SD7H15 service manual, it surely looks like there's nothing in there), it seems inevitable that the evaporator will drop to freezing under some combinations of fan speed and ambient temp...? Or is a TXV alone really able to regulate the temperature of an evaporator above freezing?

I appreciate all your help!

[bohica2xo]

If the vehicle was equipped with a freeze switch on the evaporator, then it never had a variable displacement compressor. The two systems are mutually exclusive.

[JohnHere]

The reason I originally suggested that your vent temps and pressures are wacky is because they are, at least to me. With the rear A/C off, you have vent temps of 36-38 degrees, which are pretty cold. With the rear A/C on, the temps range from 47-54 degrees, which are on the warm-ish side. Such a wide swing is a little unusual, I think.

Although I've never worked on a vehicle like yours, those that I have worked on (with front and rear A/C) haven't varied so much. Typically, at approximately your same ambient, I see vent temps of ~42 degrees with the rear A/C off to ~45 degrees with the rear A/C on. The vent temps do rise a bit with both front and rear A/C running due to the additional heat load on the refrigerant, which is why the evaporators don't normally freeze in that scenario. But on a properly operating system, I can't recall seeing vent temps climb so drastically from the 30's (rear A/C off) into the mid-50's (rear A/C on) with an 80-degree ambient and 1,500 rpm engine speed.

As for the pressures (at the same ambient and engine rpm) I would expect to see a high side of ~200 and a bit lower with the fans churning away. So your high-side readings seem elevated to me. On the low side, I would expect a reading of ~30. Once again, pressures of from ~19-35 psi show a rather wide swing. I can see why the front evaporator is freezing, especially at the low blower speed and pressures of 19 psi and lower because the evap temps at those pressures would fall well below 32 degrees.

Your pressures and vent temps led me to consider whether the charge might be off. However, you verified that the charge is correct, ruling that out. We also verified that the compressor doesn't have a control valve and isn't a variable displacement type.

We know that the thermistor in the evaporator senses temperature and regulates the compressor (turns it off and on) to keep the evaporator above freezing. You mentioned that you verified that it opens at 32 degrees. But I wonder if it's working correctly when installed in the vehicle. It could be in the wrong spot, for instance.

[thp]

The evaporator switch is a capillary type. It gets inserted into a hole in the side of the evap core, which I assume lines up with a tube in the side of the evap itself. I don't think there's any mistake possible there, and the temp is definitely getting down to freezing.

One idea is to retrofit a temp switch with more hysteresis so that the compressor isn't cycling so frequently (which I assume is a bad thing).

Another thought: Is it possible to retrofit a variable displacement compressor? Or would that cause other problems?

Going down the list of components:

Compressor: No control valve, so nothing to go wrong there. It's able to get the high side up to high pressure, so seems like it's working OK?
TXV: Tried a couple chinese aftermarket units and got the same behavior. I could try to find a NOS OEM unit, but I fear it'll be a lot of expense for maybe no benefit. Also the original unit had porous brazing that caused a leak. I'm afraid that a NOS unit might have similar issues...
Charge: verified charge a couple times.
Oil: Could too much oil cause this issue? I haven't flushed the entire system (it's a big job with the dual evaporators + rear cutout solenoid...) so I can't be sure how much is in there. Not too much oil came out when I flushed the front evap, and the liquid refrigerant in the sight glass is crystal clear.
Fans: Seem to be working normally.
Condenser: Unit is original. I "flushed" it, but it's parallel flow. None of the symptoms seem to point in this direction, but would it make sense to change it with a new one?

Thanks again.