Need help tuning my cascade low stage

[LardArse]

I can't seem to find a sweet spot for the low stage of my Cascade using R410A + 1/2HP compressor, mainly in terms of refrigerant charge. I've tried some cap tube sizes, 3.9m of 0.031, 2.5m of 0.026, 2.5m of 0.031, etc, just to see if cap tube size is my problem. I always seem to have too much or too little refrigerant. With seemingly insufficient charge, my high side is only 5 PSI or so with the high stage cooling it rather well. Adding in refrigerant bit by bit, It would go up to around 40+ to 90 PSI. Then low side at 15-18HG or so, the evap would start to get cold and then temps drop to a spot and then rise again and keep on raising. Then low side's at 20HG or so. Its starved of refrigerant. I put in a bit more refrigerant then temps would start to lower again. From 10+ HG it will slowly climb up and then liquid floods my compressor and low side goes to ~ 5 PSI. I get a very fast pulldown and hit -50+C in a very short time with this charge when first starting up but after a very short while, my evap/suction gets flooded and temps go bad. Removing just abit of refrigerant, it doesn't even get cold. I played a bit with pinching the cap tube and it helps a bit, but only temporarily and just delays the flooding. So my evap is either very starved or very flooded. What can I do?

Yes, without heat load, but I'm just testing here.

[berkut]

Put some load on it, maybe it will help Check if the HX has a proper flow (downwards)

I had similar problems with my "semicascade" when i testes it with R22. Aldough i had a injector to controll the flow and when i had a compressor flood i simply closed the injector.. And in my case- putting some load on it with a torch helped

[LardArse]

Ok, I might heatload it, but I see other people testing without heatload and not having this problem, granted incorrect charge at that. The HX does have proper flow its tilted downwards:

http://bravo.ausgamers.com/lardarse/lard/newhx1.JPG

http://bravo.ausgamers.com/lardarse/lard/newhx2.JPG

[Gary Lloyd]

Nobody ever listens, but I'll say it anyway: Turn off the low stage compressor and fine tune the high stage first.

What size cap tube is in the high stage? What are the pressures? What is the suction line temperature? Is the bottom of the compressor cold?

[LardArse]

Actually Gary, I did "finetune" hi stage first. About 3.2M of 0.031 cap tube, 1 HP R22, Suction line 6" from compressor = -24C, HX @ ~ -48C, low side= 10 HG, compressor is hot, the bottom is definitely not cold.

[Gary Lloyd]

I would guestimate the cap tube is good for about 200watts load.

Is the drier tilted downward into the cap tube?

Does the cap tube enter the bottom of the interstage heat exchanger, and suction line exit the top?

[Gary Lloyd]

You say that the low stage compressor floods when you add refrigerant. Let's consider what is happening here.

The liquid leaves the interstage heat exchanger at about -50C. It goes through the cap tube, which is inside the suction line, to the evaporator. Let's, for the sake of illustration, say that it then evaporates at -100C inside the evaporator.

The vapor leaves the evaporator at let's say -95C, where it then exchanges heat with the cap tube and is warmed up to, at best, -50C (the temperature of the liquid inside the cap tube).

What temperature should that vapor be when it reaches the compressor inlet? At most about -40C, and even that is probably a little too warm. Does this mean the compressor is flooding? No, it is in fact receiving relatively warm vapor, not liquid.

If you are trying to eliminate frost at the compressor inlet, you are starving the evaporator. Contrary to popular belief, frost does NOT indicate liquid. It indicates 32F/0C and nothing more. It is the lazy man's way of measuring the suction line temperature without having to actually strap on a temperature sensor.

[Gary Lloyd]

So... Let's say you charge the low stage until the suction line temperature near the compressor inlet is -40C, but this causes the high side pressure to skyrocket. This would indicate that the cap tube is too restrictive. You are having to fill the interstage with liquid, and raise its pressure substantially, in order to push enough refrigerant through to feed the evaporator.

But wait, there's more: This is only true if the high stage system is working properly. So, maybe we should focus on the high stage before trouble shooting the low stage?

[LardArse]

Well, the interstage is set up for counterflow, hi stage enters from the bottom and exits the top. Both driers are tilted downwards into the cap tubes.
Its not just frost on the compressor inlet that I say its flooded. Low side pressure suddenly shoot up, and you can hear the compressor make that "ball" sound you get when liquid is going back to it.

[Gary Lloyd]

What pressures and temperatures? Both high and low stages?

[Gary Lloyd]

I think the high stage cap tube is borderline. A 200watt refrigerant flow may not be enough. You may need to shorten it to about 2M.

The low stage cap tube is anybody's guess at this point.

[Gary Lloyd]

It occurs to me that this system should have a low stage discharge/suction heat exchanger in order to add heat to the suction line.

This could be just a matter of soft soldering the lines together for a foot or so.

[LardArse]

Ok, Gary, I've just done a run to -77C

http://bravo.ausgamers.com/lardarse/lard/minus77.JPG

Cap tube like this, 1.7m 0.026 spliced onto 1.8m of 0.031. A little insulation done but not thorough. No heatload yet, will put on TEC in a bit. But I still face the problem of starved/flooded. It gets to -77C and would have continued to drop but at this point, evap gets starved as I see the cap tube frost disaappear. So temps increase again and refrigerant doesn't seem to be moving back in. Then I add in just a bit more, its flooded all the way to the compressor again

http://bravo.ausgamers.com/lardarse/lard/gpucastroubleshoot1.JPG

As the temps drop to -77C, low side is 15+ HG, high side around 30+ PSI.
But when I add in a bit more refrigerant to get freon back into evap again, liquid slugs and low side is 5+ PSI, high side ~200 PSI.
What should I do now, reduce cap tube or pinch my cap tube to curb flooding?

[Gary Lloyd]

What you should never do is pinch the cap tube. What you should always do is put a temperature sensor on the suction line near the compressor. I need to know the suction line temperature.

[Gary Lloyd]

Here's what I need to know:

High stage low side pressure or SST
Low stage high side pressure or SCT
Low stage liquid line temperature
Low stage low side pressure or SST
Evaporator temperature
Low stage suction line temperature near compressor

This is the bare minimum list.

[LardArse]

I've switched over to 1.7m 0.026 + 0.8m 0.031 of cap tube. With a seemingly overcharge:

High stage low side pressure or SST: 10 HG
Low stage high side pressure or SCT : 180 PSI
Low stage liquid line temperature : Have to tear down insulation to monitor this
Low stage low side pressure or SST: ~2 HG
Evaporator temperature: -52C
Low stage suction line temperature near compressor: -54C

[Gary Lloyd]

Remove refrigerant until your suction line temp is about -40C, then take all new readings.

I know what's wrong, but let's confirm it.

[LardArse]

Well I've found a way to overcome the refrigerant not traveling into evap problem by not letting high stage get too cold before turning on the low stage. It requires about 100 PSI or so to push the refrigerant in. I've just adjusted charge a bit, every little bit seems to make a difference.
I've also put on a heat load of ~ 100w or so with a pelt. Now, evap stabilize at -50C, suction line at -51c, compressor not flooded, bottom feels warm, low side @ 0 HG, high side @ 160 PSI.

Edit: After 20 minutes or so, low side moves to 10 PSI, High side nearing 300 PSI, evap heats up to -42C and continues to warm suction line at -43C. Discharge line gets very hot.
Does this mean that the high stage is getting overwhelmed?

(Probably 3.2m of 0.031 too restrictive for heat load?)

[LardArse]

I should post some pics to see if there are any mistakes in my setup.

Compressor discharge line to oil seperator, which is 30cm tall pipe 25mm wide.

http://bravo.ausgamers.com/lardarse/lard/gpuoilsep.JPG

Low stage Liquid line + insulated drier and cap tube

http://bravo.ausgamers.com/lardarse/lard/gpucasliquidline1.JPG

Evaporator position + suction line

http://bravo.ausgamers.com/lardarse/lard/gpucasevap1.JPG

Suction line with loop to help fight liquid slugging and oil return cap tube thru manual valve.

http://bravo.ausgamers.com/lardarse/lard/gpucassuction1.JPG

[Gary Lloyd]

Let me know when you're done screwing it up, and we'll try to fix it.

[LardArse]

What did I do?:confused:
I tried to get suction line to -40C like you told me to but if I go lower on refrigerant charge, freon doesn't even come in and suction wont even get cold. As it is, I have to not let my interstage HX get too cold so that I have enough pressure to move the freon in.
I can't get suction line to -40C, attempting to do so by removing refrigerant bit by bit ressults in the freon not moving in at all.

[Gary Lloyd]

What did I do?


You screwed with the high stage system, which as far as I could tell from here, was working very well. Now you need to shut down the low stage system and get the high stage working properly again.

The problem with your low stage system is that the cap tube is WAAAAAAAY too restrictive. I would get rid of the .026 cap tubing altogether and go with just the .031.

[Gary Lloyd]

Okay, here is what's happening:

Inside the interstage, heat is being transferred from the low stage refrigerant to the high stage refrigerant. The temperatures of the two refrigerants should be very close to each other, within 5-10C, since they are in almost direct contact.

The high stage refrigerant temperature is about -50C, so the low stage refrigerant temperature should be roughly -45C to -40C. For R410A, this translates to about 7-12 psi on your low stage high side guage.

But this doesn't give you enough pressure to push the refrigerant through to feed the evaporator. Why? Because the cap tube is MUCH too restrictive. You are having to build up enormous pressures in the high side in order to push the refrigerant through. In doing so, you are completely filling the interstage with liquid refrigerant, to the point where it cannot do its job.

What is the right cap tube size for the low stage? We don't know, because this is all experimental. We need to find out.

We don't know the right size for the cap tube, but we know what it looks like when we see it. Assuming the high stage is working properly, the low stage high side pressure should correspond to a temperature (SCT) that is 5-10C above the high stage low side SST. In other words, if the high stage refrigerant is evaporating at -50C, then the low stage refrigerant should be condensing at -45C to -40C.

At the same time, the cap tube must allow enough refrigerant flow to flood the evaporator under heavy load, but not enough refrigerant flow to flood the compressor under light load.

That's the ideal cap tube size.

[Gary Lloyd]

I want everyone to know that I don't ask for a long list of temperatures and pressures just for grins and giggles. These numbers mean things. They show me how well heat is being transferred from one point to the next (or where it is failing to transfer). I can "see" how much refrigerant is in the system, where it is located, and what it is doing.

Although I have a pretty fair grasp of design principles, I claim no particular expertise in designing systems. But once the system is built, up and running, we are on MY turf. It's what I do, and who I am. :D

[LardArse]

Ok Gary, you da man so I wouldn't argue ;)
I just turned off the high stage for testing, that was all, no refrigerant or cap tube change at all.
It will be tricky to replace that 0.026 cap tube though. My evap is like this

http://bravo.ausgamers.com/lardarse/lard/gpuevap3.JPG

http://bravo.ausgamers.com/lardarse/lard/gpuevap6.JPG

The cap tube is not brazed like in the picture so I should be able to pull it out. But putting in a 0.031 cap tube back to the center will be quite impossible without unbrazing the cover or drilling a hole. Do you think it will make a big difference if I put the cap tube in near the outlet pipe and not the center like it originally is?

[Gary Lloyd]

Your best bet would be to leave a few inches of .026 in the evap, and attach the .031 to the .026 near the evap. A few inches of .026 won't restrict much.

[LardArse]

What method of cap tube connection would you suggest that would leave the cap tube connection as thin as possible to slip through my 1/4" flex hose? Presently I use a crimped out + brazed 1/4" pipe to connect my cap tubes but this leaves me with a big connection which would not get thru/block the refrigerant flow back the suction line.

[Gary Lloyd]

It would help to have a pic of the suction line connection at the evap, stripped of its insulation, to see what we are talking about. Best would be to connect it outside of the suction line.

[Gary Lloyd]

I'm confused. Is this an actual pic of the evap or not?:

http://bravo.ausgamers.com/lardarse/lard/gpuevap6.JPG

[DaBit]

Originally posted by LardArse
What method of cap tube connection would you suggest that would leave the cap tube connection as thin as possible to slip through my 1/4" flex hose?

Stick an 1mm (~ .030") drill bit in the .031" captube, and swivel it round to enlarge the hole. Eventually you can taper down the .026" (but make sure that no debris enters the tube) to make the fit easier.



--
======/
/=========
\=========
======\
--


Make sure the .026" fits _very_ tightly into the .031", otherwise the captubes *will* fill up with brazing filler. The .026" -> .031" interface must block the brazing filler material.

[LardArse]

Gary, yes, that is the evap, both pics are, just that i removed the brazed cap tube for fear of it melting.
Dabit, thx for the tip, I will try that. I tried that with a big cap tube to a 0.026 and it worked fine, just wasn't sure it was possible to enlarge the 0.031 to fit in a 0.026. I find that brazing the "cup" upside down or sideways to be safer in this case.

[Gary Lloyd]

That sounds like a winner, DaBit. :D

The alternative would be to drill a hole in the side of the tube between the evap and the connector, and drill a hole in the center of the evap top directly over the center of the maze. Then bring the .031 through the hole in the tube and into the hole in the top.

[LardArse]

I dont really like a cap tube out the centre due to the fact I like to apply direct pressure over the center down onto the core.

[Gary Lloyd]

Then DaBit's method would seem to be the best way for you. :D

[DaBit]

Pictue of my method (.028" in .047" this time) :

http://www.icecoldcomputing.com/generic/imagehandler.php?hval=6a92b65bac0ddd562da6f2b22a56 4592&quality=high

In the 'brazing a baker block' is more about how I brazed up my evap today.

[Gary Lloyd]

Another alternative would be a 1/8 tubing sleeve.

[LardArse]

2M of 0.031 seems too restrictive as well. My high side pressure is only around 10 PSI or so with HX in the region of -40+C to -50C, it is still not enough to push the refrigerant through unless I overcharge or turn off the high stage to increase pressure. I know Dabit found 2m of 0.031 not restrictive enough but this does not seem to be the case for me. Not just due to lower condensing temps but perhaps also due to the fact that my cap tube is moving upwards as this is for a GPU above. By monitoring the cap tube frost movement, I would think that 50cm might be the length to use. However, when I turn on heat load, the HX might get hotter raising pressure and 50cm might end up being too unrestrictive.
I think I need to switch to a bigger cap tube like 0.047 or so. I also need about 2.2m of cap tube length to run up to evap above. So 50 cm of 0.031 = 2+m of what size cap tube? With a bigger cap tube, I might have to run cap tube outside of my 1/4" flex hose as the big cap tube might block refrigerant flow?

[Gary Lloyd]

Does there seem to be any improvement at all? It is possible that you have a restriction in the short piece of .026 inside the block. You might try getting the suction line temp to about -40C or something just short of flooding the compressor, and report all of the temps and pressures so we can see if there is improvement.

[DaBit]

Originally posted by LardArse
[B]2M of 0.031 seems too restrictive as well. My high side pressure is only around 10 PSI or so with HX in the region of -40+C to -50C, it is still not enough to push the refrigerant through unless I overcharge or turn off the high stage to increase pressure. I know Dabit found 2m of 0.031 not restrictive enough but this does not seem to be the case for me.

This is weird. I was also running -35C .. -40C condensing, with low stage high side pressures around 10-20psi.

I had to pinch the captube really bad to correct the flow.

What I noticed when using R410A, is that the flow through the captube is *very* sensitive to fluctuating high stage temperatures. Not only SCT, but also subcool. I expect that the regular captube flow pattern screws up due to the very low pressure drop. With 15 psi pressure drop, a few degrees increase in SCT or subcool has a way too large effect on flow rate.

You might consider increasing pressure drop over the captube to stabilize it's operation.

[LardArse]

I have totally removed the 0.026 and replaced it with 0.031. A new drier to cancel out the blocked drier problem, and also vaccuumed it overnight to make sure its almost moisture free. Perhaps the difference with mine is that my cap tube is running upwards as seen in the pics, although the liquidline and drier is facing down. Will this running uphill of the cap tube make this difference? What would you suggest Dabit, run a big cap tube from the drier connected to a small one? Perhaps for the steep climb I'll use a big cap tube and then this will connect to a small one for the gradual climb portion? Sounds good? Or perhaps the other way around.

[DaBit]

Originally posted by LardArse
Perhaps the difference with mine is that my cap tube is running upwards as seen in the pics, although the liquidline and drier is facing down. Will this running uphill of the cap tube make this difference?

I even don't see how running the captube up for about 1.5 meter or so could make any difference at all.

So I would seek the origin of the problems elsewhere.

[LardArse]

It seems to be so though, I notice from frost movement that the steep upwards climb is very slow and would retreat often, but once it reaches the gentle slope part it would move through very fast. Then its flood time.
I'm thinking of putting on a big cap tube after drier then onto a 0.026 and back to a big cap tube then to the 0.031 to ease out the climb.

[Gary Lloyd]

The climb makes no difference whatsoever, and the frost tells you nothing useful.

[LardArse]

Doesn't it show where the liquid is? I know when my evap is starved, the frost doesn't even run up.

[Gary Lloyd]

Frost does not necessarily indicate liquid.

[LardArse]

Yes I do know that I mean the frost on my cap tube though, it should be an indicator since -40C liquid should frost up the cap tube when it travels up and no frot there mean the liquid is not cold enough or not even there.

[Gary Lloyd]

Frost indicates 0C or below. This could mean cold liquid or it could mean cold vapor. Being that the liquid leaving the interstage is very cold, lack of frost probably means not even there.

[LardArse]

Running it now, and High side is 5 PSI only. HX @ -50+C, frost on cap tube stops halfway, no cold at all on evap. I'm thinking the oil ep might be problematic. I might remove it test a run w/o it. Since its connected pretty close to the HX, it gets a bit cold from it and the refrigerant might be condensing in there? There should be not much of blockage here as I'm using layers and layers of wire mesh.

[Gary Lloyd]

What are the rest of the temperatures and pressures?

[LardArse]

Hi stage low side: ~13 HG
Low stage hi side: 5 PSI
Low stage low side: 25HG
low stage suction/evap: ambient 28C
Liquid line not monitored yet but should be ~ -50C like HX.
And this is the same charge that would get me flooded all the way to compressor without a cold enough HX/higher high side.

I'm thinking the oil ep might be problematic. I might remove it test a run w/o it. Since its connected pretty close to the HX, it gets a bit cold from it and the refrigerant might be condensing in there? There should be not much of blockage here as I'm using layers and layers of wire mesh.

[Gary Lloyd]

Add refrigerant until the low stage high side pressure is about 15 psi.

[Gary Lloyd]

Just to make sure, the low stage refrigerant is entering the top of the interstage HX and exiting the bottom, right? We want gravity feeding the liquid to the entrance of the cap tube.

[LardArse]

Yes, its not totally pointing straight down, but its tilted downwards, all the way downwards to the drier. But I'll try to increase the incline more to see if it helps.

[Gary Lloyd]

Did you add refrigerant yet?

If so, what are the temperatures and pressures?

[LardArse]

Yes, until 15 PSI, it would keep going down and I just kept adding to maintain. Still 25 HG low side, evap at ambient, cap tube frost at midway only.
Just to be sure my HX is not the problem here, the chanel for the gas line is something like a 40cm pipe 25mm in diameter, lots of copper fins inside. The sealed up pipe in the middle is just to increase surface area and to hold fin in place, not connected to the liquid line.
http://bravo.ausgamers.com/lardarse/lard/interheat1.JPG
Gas enters into this from a 1/4" pipe from oil seperator. and exits with 1/4" pipe connected to drier as well.
Due to Dabit using a much smaller HX, perhaps this is why he needs a small cap tube than me?
Also, hot gas enters near bottom of oil seperator at the same level as the discharge line and exits thru top. I see some rigs having the discharge line move down to ground level and entering the oil sep at this level, in other words having their oil seperator at ground level unlike me, would this make a difference too?

[Gary Lloyd]

At 5 psi you were probably feeding only vapor to the cap tube, but at 15 psi it should be feeding solid liquid.

Again, assuming solid liquid at the cap tube entrance, the numbers say that the cap tube is far too restrictive. Being that DaBit's experience shows that this cap tube size should be not restrictive enough, the conclusion would be that it is partially plugged/pinched/kinked. I would replace the cap tube.

[LardArse]

Gary, it is a new piece of cap tube. I couldn't have pinched it.

[Gary Lloyd]

Either the cap tube has vapor at its entrance or the liquid isn't getting through.

[LardArse]

The last time I had a sight glas on it there was a good liquid seal but that was with many different variables so it doen't mean anything I guess. But I'll just try another cap tube and put on a coil of copper pipe as a makeshift evap for testing, again. If that doesn't work, then I'm gonna remove the oil seperator to test.

[DaBit]

Originally posted by LardArse
[B]cap tube frost at midway only.

This is so weird. Nothing warmer than -40C/F flows through the captube so it should frost all the way. In my system everything after the interstage froze up. Drier, captube, sight glass.


Just to be sure my HX is not the problem here, the chanel for the gas line is something like a 40cm pipe 25mm in diameter, lots of copper fins inside.

Looks good. Where flows the high stage refrigerant?


I see some rigs having the discharge line move down to ground level and entering the oil sep at this level, in other words having their oil seperator at ground level unlike me, would this make a difference too?

Nope, it shouldn't.

I still think part of your problem comes from the very sensitive cap tube. I had problems stabilizing the R410A circuit too, and I also had 'oscillations'; starving evap followed by flooding evap, repeating every 30 secs or so.

Apply a load, also during testing. This works much better.

[LardArse]

Dabit, the frost at midway is when the evap is starved and refrigerant not pushing thru. Once high side is high enough, frost moves all the way thru but flooding as said before.
You're right when you say heatloading helps as I've put on a TEC outputting ~ 100w before and was maintaining -51C at evap and -53C at suction for some time, but that was with an overcharge I think cos the compressor gets flooded eventually. So far the smoothest running I had was by starting the low stage when hi side at static is around 50 PSI and nearing 100 PSI when running before the HX hit rock bottom. Cos when Hi stage hits rock bottom, static pressure for low stage is like 0 PSI. As this runs and evap drops, I then turn on heat load.
For the high stage refrigerant I think it is thinned out around the whole outside of the gas pipe and the mass of brass or whatever that silver metal bulk is acts as a cold sponge, thats my guess, It not possible to see the flow chanels of the high stage. That metal bulk really acts like a cold sponge, it takes signifcantly longer to get the HX down compared to my previous pipe in pipe HX but it also changes temps very slowly.

[DaBit]

Originally posted by LardArse
Dabit, the frost at midway is when the evap is starved and refrigerant not pushing thru.

When the compressor is running, there is a flow. And thus cold vapour flows through the captube, which means frost on the captube.


That metal bulk really acts like a cold sponge, it takes signifcantly longer to get the HX down compared to my previous pipe in pipe HX but it also changes temps very slowly.

You want this. It stabilizes operation.

[LardArse]

Ok, I've found 0.8m x 0.031 to be good for flow at about 10-15 PSI. Perhaps 0.9m to 1m when HX gets hotter with a heat load. I do not need to overcharge and flow is not floodingly fast but seems to be metered nicely with this cap tube. Thats tested with a long coil of pipe as evap temporarily without insulation. Ovbviously thats very different from your experience but this is happening in my case but probably there is a problem elsewhere as well. I have a small drier, smallest I could get. And I've used cap tubes from 2 different rolls of 0.031 so there should be no misstaking here, unless the 0.031 they sell me are different.
Well I'll continue to troubleshoot, but if I need to go with this kinda cap tube arrangement, what diameter of cap tube should I go for? I want around 2.5m of cap tube to play with so 0.8m x 0.031 = 2.5m x ?
I know it sounds like a restriction near the cap tube entrance so I'll recheck this area as well.

[DaBit]

Originally posted by LardArse
Well I'll continue to troubleshoot, but if I need to go with this kinda cap tube arrangement, what diameter of cap tube should I go for? I want around 2.5m of cap tube to play with so 0.8m x 0.031 = 2.5m x ?

There is a .PDF document somewhere on the Net which shows a table with length conversions to go from one tube to another. I don't have it here at the moment, but I thought it was from Supco or Electrolux.

[Learn]

Originally posted by DaBit
There is a .PDF document somewhere on the Net which shows a table with length conversions to go from one tube to another. I don't have it here at the moment, but I thought it was from Supco or Electrolux.

this?
captube_sizing.pdf (http://learn.chez.tiscali.fr/alternative_captube_sizing.pdf)

or this?
supco_captube.pdf (http://learn.chez.tiscali.fr/supco_captube.pdf)

[Gary Lloyd]

Ok, I've found 0.8m x 0.031 to be good for flow at about 10-15 PSI. Perhaps 0.9m to 1m when HX gets hotter with a heat load. I do not need to overcharge and flow is not floodingly fast but seems to be metered nicely with this cap tube.

Temperatures and pressures?

[isp]

to achieve the maximum in cooling, please trim your nails :D

LOL I'm just kidding sir :D

[LardArse]

Hi guys, been doing some testing with both R410 and R22 because I don't have a tank to reclaim the expensive R410 in so only using R22 in test drives but I do know that circumstances/pressures/cap tube lengths would be very different. Anyways, I think it is my cap tube being too unrestrictive resulting in either too much vapour or all vapour. Tried out 1.2M of 0.026 from drier connected to 2.3m of 0.031. With R410, it was good for 1 run and after that it seems that only vapour is going thru to the evap until I pinch the cap tubes a few times and then frost starts appearing after the pinch on the cap tube showing at least some amount of liquid.

So anyways, I went on to test 2m of 0.026 + 2.3m of 0.031 with R22. This seems quite ok with R22 at low stage maintaining evap temp of -56C with uninsulated liquid line and suction is about -40C. The frost on cap tube breaks into 2 with frost on cap tube after drier and no frost at mid section and then frost to evap. Low side is 13 HG, high side 100 PSI, HX @ ~ -40C. Now I'm thinking that there is still quite some vapour going into evap as usually when I get a nice liquid flood at evap, the pulldown is much faster than this. So it seems to me that cap tube is still not restrictive enough. A question: if I find a certain length of cap tube to work good with R22, for the same HX temperature, can I assume that for R410 it would be longer?

Also, I dunno if my observations are correct but it seems to me that there is much more liquid than gas in the system if the HX is hotter/higher pressure. This is normal right?

Finally, there are quite a few guys who pointed out that I might be facing oil logging problems but from opening up my suction pipe/hose and evap outlet, there is little to no oil found indicating that the oil seperator seems to be doing a pretty good job.

Thx for the help guys, really appreciate it.

[DaBit]

Originally posted by LardArse
A question: if I find a certain length of cap tube to work good with R22, for the same HX temperature, can I assume that for R410 it would be longer?

Yes, but only slightly. Vapour pressures really don't differ that much at those low temperatures.

I would consider condensing the R410A at a higher temperature to stabilize captube operation. I have been trying some R410A again, and stabilizing the captube is almost impossible.

An R410A SCT of about -10C would be enough (4.8 barg / 70 psig). How? Add a temperature controller such as an Eliwell 901, Carel IR32, or even a thermostat from a refrigerator, and have it switch a hot gas bypass solenoid. An initial setting of -15C would be fine. Or use a valve specifically designed for hot gas bypass.

Then, use R134a captube sizing for the low stage.

This setup has another advantage. If you are flooding the evap, lower the interstage temp. If you are starving the evap, increase interstage temp.

[LardArse]

I'm glad I'm not the only one facing problems with R410 at low stage. I've never expected the HX to be so low a temp as to cause metering problems. If need be I may shorten the cap tube on the High stage by quite a lot only that would make it very inefficient. I'm also working on my CPU cascade so I might also think about sharing the high stage in this case.
But I'd rather have my HX as low as can be to acheive best temps. Would switching to another refrigerant help? I've been thinking of R508, seems to be much more affordable than R23 here. Do I have a chance with this? In any case, I'll continue to work on stabilizing the R410 cap tube, I may need to have slight overcharge + accumulator to help.

[DaBit]

Slight overcharge plus accumulator would work, but you are certainly feeding a llot of vapour down the captube, essentially throwing away capacity.

R23, R508B, ethane or ethylene would be the solution. R23 is the least preferred refrigerant. It needs a high compression ratio, has bad thermophysical properties, and features very high discharge temperatures which ultimately lead to oil decomposition. R508B or ethane should be on your shopping list. Use ethylene only if you know you can handle the pressures involved.

[LardArse]

Thx for the info, R23 costs too much anyways. If the quote from my friendly A/C supplier holds then I should be able to get hold of some R508 for ~ US$200.

[LardArse]

Ok, now I need some advice here. 3m of 0.026 + 2.2m of 0.031 cap tube with R410A, HX hovering around -40C.
Starting up the low stage at the right time seems crucial in this testing. I have given up on testing for unloaded temps as this would only keep cycling. Had a run with a TEC heatload ~ 80-100w perhaps. Got the evap to stabilize at 1 of the runs with the heatload, Started the low stage when HX at ~ -35C.

Evap at -51C
Suction line at -45C
Lowside at 0 PSI
High side at 200 PSI.

It will maintain at this position for 20 minutes but then finally, low side and high side moves up bit by bit and temps start rising.
Turning the low stage off and waiting for HX to get colder and then turning low stage back on, the evap doesn't get cold then. Added in a bit more refrigerant/ removing some doesn't seem to yield back the same results.
Shortern or lengthen cap tube then? I would suppose shorten?
The high side pressure seems so sensitive.

[Gary Lloyd]

All of the numbers I have seen thus far indicate that the low stage cap tube is too restrictive. And stabilization at higher interstage temp would tend to confirm this. This says to me that it takes more pressure to force the refrigerant through, because the cap tube is too restrictive.

[LardArse]

Roger that but here lies the problem, if I open up the cap tube, too much vapour seems to be going in and not liquid.

[Gary Lloyd]

How do you know that?

I may have mentioned this before, but it bears repeating: I can't see the system from here. Temperatures and pressures tell me everything. Conclusions and speculations tell me nothing.

[LardArse]

Well the evap doesn't seem to be getting cold and then only cold at times and later starved and keep adding refrigerant bit by bit, the low side will then slowly move from 25 HG to 15 HG but evap doesn't seem to get cold much indicating that it is gas coming into the evap and back to compressor.

Just restarted the whole rig again, and
evap reaches -66C then slowly drops off,
suc tion at -10C and climbing,
low side at 18 HG
Hi side at 75 PSI
This without load

[Gary Lloyd]

Conclusions and speculations.

[LardArse]

Ok, I will keep cutting down on cap tube and then test. But I should fix some rules to help control the variables.

How much of the 0.026 should I cut down in steps?
At what static pressure/liquid line temp should I start the low stage?
At what temp/pressure should I start my heat load?

[Gary Lloyd]

It makes no sense to me at all that you are using two different cap tube sizes. Why not get rid of the .026, and just use the .031?

[LardArse]

Well I've found that having a more restrictive cap tube after the drier creates a better liquid seal.

[Gary Lloyd]

The problem isn't that you don't know. It is that you don't know that you don't know.

[DaBit]

Get the R410A SCT up, tune the captube, and then decrease the SCT in steps, optimizing the captube for the reduced SCT. These optimization steps consist of cutting little pieces off the captube since we need less restriction when the pressure drop lowers.

We need to troubleshoot this in a structural way, otherwise we won't find the sweet spot.

When optimizing the captube, charge refrigerant until low stage subcooling is 3-5K, then measure superheat. And test with load.

[LardArse]

Ok, its my bad, the high stage is the problem. I have shortened the cap tube of the high stage from 3.3m to 2m to 1.5m of 0.031 and this is where it is able to keep stable temps when the low stage is running with a heat load. The temps on the surface of the HX is very unreliable as gauge but monitoring temp of the low stage disharge inlet into the HX gives me a good idea of how well the high stage is cooling the low stage. Just the temp here going up 1C and the low stage evap immediately follows up 1C.

So now with a pelt outputting ~ 100w onto the evap, I get load temperature of -63C.

High stage R22
Low side: 0 PSI
High side: 225 PSI
Suction line when low stage turned on and loaded: -2C
Condensor air in: 29C
Condensor air out 41C

The condensor is a little small it seems but still holding up, and the fan blowing at it is not really powerful, I might think about changing it out in the future, but it holds temps steady enough so I'll leave it for now.

Low stage R410A
Low side: 15 HG
High side: 25 PSI
Evap loaded with ~ 100w: -63C
Suction line temp: -49C
Cap tube: 0.5m 0.026 + 1.8m of 0.031

This is stabilized temps and I'm happy finally, after adjusting low stage then high stage then low stage and on and on. This looks ok?

[DaBit]

Is the temperature directly after the interstage also -2C? If so, I would say it is starved.