My project is completed! However...

Hello everyone! I've finally completed my R290 cold trap project, and it's working pretty ok-ish, meaning that it can indeed reach temperatures below -35°C, which is what I needed. However, having designed the circuit myself and being somewhat of a newbie, I definitely think something could have been done better. First of all, I must say that I didn't seal off the two access ports for now, mainly because I might need to add or remove refrigerant, and also because brazing with the gas inside scares me a bit, but I know professionals do it this way. Anyway, I've tested extensively with an Inficon for HC and have detected no leaks.

When charging the gas, I evacuated the system down to 250 microns, but unfortunately, the stiffness of the charging hose affected the accuracy of the gas gram measurement. Therefore, I charged it considering the superheat temperature, so in the end, I have no idea how many grams I put in.

After about an hour of operation, here are my observations:

Liquid line: 27°C, 13.5 bar
Suction line: -31°C, 0.3 bar

Attachment 16331

It's worth noting that this cold trap could operate under very low load conditions for the compressor, so I had planned for a liquid accumulator, and it seems to be doing its job because I'm not detecting any anomalies in the compressor, such as an increase in current consumption which could indicate the compressor is operating with insufficient lubrication. I've also added more thermocouples to the system, such as:

Ambient: 15°C
Compressor: 50°C
Just out of the compressor: 65°C
Evaporator inlet: -41°C
Evaporator outlet after hours with no load: -38°C

I used 3.5 meters of 0.9mm ID capillary tube as indicated by the charts to achieve -35/-40°C evaporator temperatures. However, either it's too long, or I made a mistake by coiling it around the suction line including the liquid accumulator, because the capillary freezes very visibly.

Attachment 16332

The suction line is also very cold, and the liquid accumulator freezes as shown in the photo.

Attachment 16333

However, I imagine that with a higher ambient temperature and a higher load inside the cold trap, the situation might improve, but I still need to verify this.

Sounds like youve got it running reasonably well. Be interesting to hear how things are at full load.

The capillary freezing might just be from the subcooling. Measuring the temperature of the capillary tube might tell you whats happening?

Ice on low temp pipework seems normal. Its the superheat and subcooling values that are the important thing.

Kay86, this sounds like an interesting project!

I am sad I haven't seen your earlier posts on this :/



I am an industrial Refrigeration Technician and deal mostly with larger NH3 systems now.

But when I were an apprentice and in training in 1996, when there wasn't any jobs or service, I played around a lot with discarded hermetic fridge compressors that still worked and made "Ice signs" of the company logo using scrap copper pipes that I bent to shape and a discarded TX valve.

so reading about your project really brought me back :)


I see you call it a "cold trap" I've googled it and have seen multiple uses for a cold trap, are you using it for anything in particular, or is it just an experiment?

Quote:

---

Quote:

---

and also because brazing with the gas inside scares me a bit, but I know professionals do it this way.

---

No we don't... especially not with R-290. Never ever ever ever try to braze copper pipe with pressure inside the pipes.

There are "saddle taps" that you can braze onto a a charged system, like in this video:
https://www.youtube.com/watch?v=JWuanarXixw

This guy seems comfortable with it and he's using a rod with no flux so he probably knows what he's doing and has laser control of the temperature.


Me, I wouldn't have the nerve to do that.

I've replaced DN80 Suction valves on plate freezers on ammonia systems with 4 tons of liquid ammonia with the plant running, just have to change the set point on the compressors from -0,3 bar to 0 or 0,1 bar, close the quick closing valves (there is a bypass) that keeps the pressure down.
unscrew all but 2 bolts, check if there is vacuum, then adjust the suction pressure setpoint on the compressor until it is equal to ambient pressure... or slightly above, then unscrew the last two bolts and yank out the old valve with your right hand, and smack the new valve in with your left hand and have a bolt ready in your right hand again.

But I don't have the nerve to put a torch to a copper pipe with refrigerant inside.

For this I have a pair of refco piercing pliers
Attachment 16334

Snap them on and you have a service connection and can empty the system... then when it's empty you can remove the pliers and solder on a service valve.

---

Sorry, a slight digression on my part there :/

Also the charging procedure, stiff hose and it affected the weight.
I don't work much on smaller applications, but when I do, I use a hose with a shut of valve
similar to this:
Attachment 16335
This way I can vacuum the hose along with the rest of the system all the way to the refrigerant bottle, then when ready to fill the system I close this valve, open the liquid connection on the bottle and that fills up the hose all the way to the compressor with liquid, so then I can set the scale to 0 grams so I only measure the amount that goes into the system and not the weight that goes into the hose.

Or you could fill gas direct on the suction side. it takes longer, but it is more accurate :)

You said it's a cold trap, I've been googling it and I know what it is, just not sure what you are going to use it for... is it a science experiment, or like I did when I were an apprentice "I just wanted to see if I could do it" kind of thing?

What is your goal with this project? I find it very interesting and brings me back to my days of experimenting :)

I'm not familiar with R-290, so I have to go by rule of thumb and experience, with a properly sized air cooled condenser, the condenser pressure should be approximately 10-15K/C above ambient, oh and by the way, in refrigeration we measure pressure in Celsius meaning your 13.5 bar is 42.5°C in pressure, you should have 20 - 25°C (7-8 bar, or 10-ish) on your condenser if it was sized for your application, is there any heat load on the cold side that would load the condenser?

Or once you reach -36°C (0.3 bar) in suction pressure is it purely maintenance?

Could also be air in the system if you were unlucky during the vacuuming and switching from vacuum to charging, guess on a system like this with 250 grams-ish, it there was air in the charging hose, that could be enough to block the condenser.

I have to say I love your scientific approach to this, and it reminds me of my days of playing around :)

I see you are using capillary tube, which is very good if you get it just right with the length and if it is cut properly so there is no restriction in the inlet and outlet... I've seen people cut them to length with pliers and then try to open the hole again with needles and it made me cry a little inside.

if it's with a fixed load, I would recommend trying with a needle valve like this
Attachment 16336

with a temperature probe on the evaporator outlet you could take the suction pressure down to -40°C (0.1 bar)

I can't recommend taking R-290 or any other flammable refrigerants to 0 bar or down to a vacuum, as with leakages of introduction of air into the system over time, they may spontaneously explode just from the compression of the compressor... It's called the "Diesel effect" as diesel engines doesn't rely on spark plugs to ignite the fuel, it relies on compression.

So can happen with flammable refrigerants if there is air in the system or if it leaks over time and reaches the perfect mixture:
A mixture of refrigeration oil, refrigerant, and air in the compressor may cause non-ignition combustion (diesel combustion). The risk of such accidents must also be assessed when using weakly or strongly flammable refrigerants in split air conditioners.


I would like to see more about this project, but please be careful :)