hi,mike
i don't agree with you
i think the tank in the system very concernment.
your system is small,but if your do some bigger ,i think you need it .
hi,mike
i don't agree with you
i think the tank in the system very concernment.
your system is small,but if your do some bigger ,i think you need it .
Kang -- Well to be clear, this discussion is about the CryoBUG. And for this unit an expansion tank is not needed. And even in a larger AutoC system this is not always a requirement (i.e.; Telemark Water Vapor Cryotraps do not use expansion tanks). Of course a lot of this depends upon what kind of static pressure and head pressure that the compressors can easily tolerate.Originally Posted by kang-China
So for instance on a Telemark TVP-3500 they typically have a room temperature static balance pressure equal to 235-240 psig, and a start-up head pressure of nearly 425 psig. For the scroll compressors that they normally use, this is well within their specifications. Of course in their situation they use a staggered start-up of the two compressors, and what is called an unloader valve to keep things somewhat reasonable. But I assure you there is no expansion tank.
Bottom line, it's all dependent upon system design as to whether an expansion tank is needed or not.
Thank you for your question
Last edited by mytekcontrols; 01-29-2012 at 02:05 PM.
Michael St. Pierre
- Worked 15 years for Polycold Systems
- Now Self-Employed
- Manufacture Heat Load Controllers
- Also do contract service work on Polycold units
Side note: I usually don't respond to PM's or emails regarding the projects that I post in the forums. I feel it's much more fair to all, to answer questions within the forum topics themselves.
Thank you for the explanation.
I guess I've always had an uncanny feeling above the 300psig mark. Maybe it's the few years of building pumpkin cannons that launch at 300psig. And knowing that a pumpkin can go through cinder blocks at that pressure.
But, that's what pressure testing is for isn't it?
It's interesting to see that you can find window air conditioners with r410a standard now that r22 is gone. What a change. What a loveeeeellyyy change
If you have a cooling question or concern feel free to contact me.
Yes I totally agree! It opens up a lot more possibilities in small system design.It's interesting to see that you can find window air conditioners with r410a standard now that r22 is gone. What a change. What a loveeeeellyyy change
Michael St. Pierre
- Worked 15 years for Polycold Systems
- Now Self-Employed
- Manufacture Heat Load Controllers
- Also do contract service work on Polycold units
Side note: I usually don't respond to PM's or emails regarding the projects that I post in the forums. I feel it's much more fair to all, to answer questions within the forum topics themselves.
-155 C and dropping
At 1 hour 30 minutes after warm start...
CryoBUG 4-3-12 Test TC Meter.jpg
New even smaller HX Stack design showing Bullet Strainer Phase Separator...
CryoBUG V5.jpg
Piping Diagram of new configuration (oil management not implemented yet)...
CryoBUG V5 Diagram.png
Test chart for yesterdays test (4/3/2012)...
CryoBUG V5 Chart 4-3-12.png
Hydrocarbon Charge:
R-600
R-170
R-14
Argon
Total = 90 grams (3.17 ounces), static balance pressure = 235 psig
Still seeing some freeze-out issues, but it didn't occur until I passed below -155 C. Not sure if it is a refrigerant (possibly R-600) or the oil. Obviously not doing much in the way of oil separation yet (Temprite 340 not being used, only proposed), so it certainly could be the oil that is the problem. But step by step the "BUG" is getting colder and more stable (running pressures were very steady down to -155C).
Next up on the agenda: swap out R-290 for the R-600, since it has a much colder freezing point.
Michael St. Pierre
- Worked 15 years for Polycold Systems
- Now Self-Employed
- Manufacture Heat Load Controllers
- Also do contract service work on Polycold units
Side note: I usually don't respond to PM's or emails regarding the projects that I post in the forums. I feel it's much more fair to all, to answer questions within the forum topics themselves.
Awesome result for such a compact unit!
I'm kind of surprised to see it working so well with only one HX and an Aux HX...
Q: Will there actually be much R-600 left in the final stage (which would freeze up)?
TC3 (Phase sep) goes really low and should condense most of the R-600, right?
If it were a normal AutoC with multiple Phase Separators I would say you're probably right in your assumption. But this system is so simple, and only having a single separation point makes me wonder if just by chance a small amount made it's way beyond and into the evaporation circuit. It wouldn't take much to cause a problem.
Anyway in order to be sure of what's happening, and to see what it'll do, swapping out the butane for propane will let me rule out a refrigerant as being a freeze-out problem. Thus leaving only the oil.
Edit: There is also the possibility that I have too much R-600 in the charge and/or R-170, and when things go into full condensing mode these liquids overflow out the top of the small phase separator. Because everything looked just fine up to the point where it was beginning to flat line, and presumably everything that was going to condense, had condensed.
Last edited by mytekcontrols; 04-08-2012 at 06:33 PM.
Michael St. Pierre
- Worked 15 years for Polycold Systems
- Now Self-Employed
- Manufacture Heat Load Controllers
- Also do contract service work on Polycold units
Side note: I usually don't respond to PM's or emails regarding the projects that I post in the forums. I feel it's much more fair to all, to answer questions within the forum topics themselves.
Here are a couple of pictures showing a break down of the individual coils and final assembly.
This is the Cascade Condenser/SubCooler HX...
CryoBUG_CC2-SC_V5.jpg
Completed assembly with the Auxiliary Condenser wrapped around the Cascade (w/insulation space)...
CryoBUG_Stack_V5_Assy.jpg
That bullet strainer is the one and only Phase Separator.
Dimensions: 4" diameter, 8" high
Getting pretty tiny (I kinda got in the habit of squashing BUGs)
Last edited by mytekcontrols; 04-04-2012 at 04:15 PM.
Michael St. Pierre
- Worked 15 years for Polycold Systems
- Now Self-Employed
- Manufacture Heat Load Controllers
- Also do contract service work on Polycold units
Side note: I usually don't respond to PM's or emails regarding the projects that I post in the forums. I feel it's much more fair to all, to answer questions within the forum topics themselves.
Hi Michael,
Those tube-in-tube coils are very versatile. However they are bad in one way. The heat-transfer area is essentially the outer surface area of the inner tube, while the heat-loses area is that of the outer tube. So very good insulation is needed for good performance. I am guessing you ran the CRYO-Bug deep within a box of insulation. Oh with that thought in mind, have you ever used perlite for insulation. Where I used to work we made cryogenic pumps and blowers and would sometime enclose the unit within a box filled with perlite. To work on the unit ment getting out the shop vac and sucking out the perlite (which could be reused).
Plate-HX have a much lower ratio of heat-loss area to heat-transfer area, but other issues. I am currently building an Autocascade with one water-cooled condenser and 2 cascade condensers and 2 phase-separators. I am using small plate HX's, and I realized that the cascade condensers (by design) only partially condense (two-phase flow at inlet and outlets). So I need to mount them horizontally with the inlet and outlet ports facing down. Otherwise they would pool a certain amount within them acting like phase-separators-before-the-phase-separator and play havic with the refrigerant mixture distribution. This is another area where tube-in-tube HX's excell (no trapping of two-phase condensate within them).
Michael (and others) I truly enjoy reading at learning from all the posters.
Thanks,
Kevin
Hi Kevin
Yeah I stick the HX stack into a plastic pot and stuff it full of fiberglass insulation. Then I cover the top with plastic sheeting, and use duct tape to seal it to the pot. However I do like the idea of using perlite, it would be far less itchy.
The problem with plate HX's is the limitations in orientation as you mentioned, and for what I'm trying to do with CryoBUG, not really small enough. Also it wouldn't be feasible to have a cap tube feed about mid way into the suction circuit as I was able to do on my Cascade/SubCooler HX. So it would take 3 plates to do what I am doing with essentially 2 tube-in-tube HX's. I would also imagine that the weight would be greater with 3 plates vs my stack.
Good luck on your AutoC. Be interesting to see how close your simulation compares to reality.
Michael St. Pierre
- Worked 15 years for Polycold Systems
- Now Self-Employed
- Manufacture Heat Load Controllers
- Also do contract service work on Polycold units
Side note: I usually don't respond to PM's or emails regarding the projects that I post in the forums. I feel it's much more fair to all, to answer questions within the forum topics themselves.
What're the HX dimensions at this point?
Also, where is the freeze out occuring? In the evaporator? At the captube before? Identifying the block could help you decide what's causing the block. What's the actual freezing temperature of butane?
If you have a cooling question or concern feel free to contact me.
Hi Adam
It appears that the freeze-out occurs just past where the 1st cap tube is feeding. If you look at the chart I provided earlier, you will see that CC Suct starts getting much colder just before the evaporator starts to warm up. This temperature point is just past the #1 cap tube feed. Now as I said earlier it could be caused by the phase separator overflowing, because it's volume is rather small (proportional with the stack). We're looking at a volume of 1/2 an ounce. My total charge was about 3 ounces, so it is conceivable that once everything condensed, the phase separator flooded.
As for HX sizing, I can't give out that information or the sizing of the cap tubes, since for the time being I would like to limit the competition. After all I am shooting to make this into a viable product.
At this point I have recharged it with Propane, Ethane, R14, and Argon. And I also reduced the amounts. So this gives me a refrigerant charge that can not freeze until reaching -185 C (not likely I'll get that cold). So when I run my next test, at least one variable has been eliminated.
As to your question about the freezing point of Butane, it is -138 C.
Last edited by mytekcontrols; 04-04-2012 at 10:41 PM.
Michael St. Pierre
- Worked 15 years for Polycold Systems
- Now Self-Employed
- Manufacture Heat Load Controllers
- Also do contract service work on Polycold units
Side note: I usually don't respond to PM's or emails regarding the projects that I post in the forums. I feel it's much more fair to all, to answer questions within the forum topics themselves.
Ah must have missed that info, my apologies.
Looks good though, glad to see you playing with a HC mix.
If you have a cooling question or concern feel free to contact me.
Yeah I resisted using HC's for as long as I could, but when it became obvious that a lot of the liquid HFC's just don't have cold enough freezing points, I had to make the switch. The only HFC that might still be a possibility, is R227ea. It has a freezing point nearly the same as R-600, and a boiling point close to IsoButane (R-600a). Problem is its very expensive, and other than buying an FM-200 fire extinguisher, hard to get in small quantities.
I really want to get this single-stage AutoC design to work reliably at -150 C or slightly colder. The advantage is one of size and current draw. The fewer the stages the more efficient it becomes, thus only needing a small compressor (I'm only using a 4.8cc/rev Rotary).
Last edited by mytekcontrols; 04-04-2012 at 10:40 PM.
Michael St. Pierre
- Worked 15 years for Polycold Systems
- Now Self-Employed
- Manufacture Heat Load Controllers
- Also do contract service work on Polycold units
Side note: I usually don't respond to PM's or emails regarding the projects that I post in the forums. I feel it's much more fair to all, to answer questions within the forum topics themselves.
I'm not sure that you are getting good phase separation, but I don't think (in this case) that it really matters. The R134a has extremely low vapor pressure at -150C (NIST refprop lowest valid temp. was -103C and there V.P was 403 Pa (0.058 psia)) and my thought is it never evaporates in your final cold evaporator -- just flows through as thermal mass. In warmer system ( evap. -50C) single phase-separator designs result in large temperature "glide" (like, -50C in -25C out, and Hysys simulation show poor phase-separation) so more of a problem if the design goal is near constant temperature evaporator at -50C. But for your COLD system, liquid R134a circulating through final evaporator, though not hurting tempatures, will lower thermodynamic efficiency (its mass must be cooled).
I also checked n-butane at -138C on refprop (0.00010134 psia) so its not evaporating in a 24-psia evaporator.
Kevin
Last edited by Kevin Hotton; 04-05-2012 at 08:32 PM.
After playing around with this system for a while, I no longer feel the same way as when I originally posted that statement. Later I discovered that the Temprite 340's hold up an ounce of refrigerant before it reaches the liquid outlet port (not good on such a small system). So I had to over charge the liquids. And I really wasn't seeing cold enough temps at the first stages, unless I pumped up the liquid amounts even more, which I suspect just allowed it to slop over.
So in my latest design the use of 340's for Phase Separators is out. I also downsized the heat exchangers, while improving the pressure drop on the suction side by keeping the cap tubes external. I put a gauge on the top of my aux condenser and saw a pressure differential of between 1-3 psi through-out the entire cool down period. When switching to this new design, I also went to HC refrigerants, only retaining the R14 and Argon. And I cut my charge weight by half. So at this point I'm feeling pretty good about the results I'm seeing.
Almost forgot... I also ran a series of cap tube tests with the new stack, starting out with extremely long and small cap tubes, and then retested with ever decreasing length until I zeroed in on what I wanted to see for running pressures.
I do a lot of guessing based on intuition when I develop AutoC's, because unfortunately I do not have access to the software that you do Kevin (although I wish I did). However I'll be most interested to see the results from a system built and charged based upon a simulation, to see how far that technology has come. And please don't take this as defensive on my part, but really more based on my curiosity.
Last edited by mytekcontrols; 04-05-2012 at 08:44 AM.
Michael St. Pierre
- Worked 15 years for Polycold Systems
- Now Self-Employed
- Manufacture Heat Load Controllers
- Also do contract service work on Polycold units
Side note: I usually don't respond to PM's or emails regarding the projects that I post in the forums. I feel it's much more fair to all, to answer questions within the forum topics themselves.
Thought this chart might give insight into autocascade evaporator performance.
Here is a chart of the final evaporator in my current AutoCascade cycle (2 Phase-separators). The compressor inlet composition is (mole fractions: iso-butane 0.38; ethylene 0.68). At the evaporator inlet (after 2 PS’s) the composition is: iso-butane 0.0474; and ethylene 0.9526 yet the small amount of iso-butane present causes a large temperature glide effect – if a evaporator outlet condition of saturated vapor is specified. All saturated vapor out gives the highest cooling duty, but the most glide. With my current design it shows that if glide is to be reduced to less than 5 degrees C, that outlet quality will be 0.81 and about 2/3 of peak cooling duty remains. For my application the glide is not a problem.
Kevin
Last edited by Kevin Hotton; 04-05-2012 at 08:45 AM.
Kevin I appreciate you taking the time to post this chart, and the great contributions you have been making to better define the AutoC. But this old brain of mine is having a hard time wrapping itself around that chart
I'm used to dealing with an X/Y chart having only 2 axises defined (for instance: temperature vs time). But it appears that on your chart there are 3 definitions, temperature (Y-left), phase quality (X), and something else??? (Y-right). Sorry but I'm more of a hands-on type guy, never was particularly good at math or physics, and certainly take a different approach to understanding AutoC's.
Can you elaborate on what this chart means in lay mans terms? sorry to be a pain
Michael St. Pierre
- Worked 15 years for Polycold Systems
- Now Self-Employed
- Manufacture Heat Load Controllers
- Also do contract service work on Polycold units
Side note: I usually don't respond to PM's or emails regarding the projects that I post in the forums. I feel it's much more fair to all, to answer questions within the forum topics themselves.
Hi Michael,
The second y-axis is to show (for a given evaporator exit quality) the relative amount of cooling duty compared to maximun cooling duty. The saturated exit condition is what I am using for maximum cooling duty (the greatest number of Watts of cooling but the most temperature glide). The x-axis if refrigerant quality (a thermodynamic term that indicates the fraction of the mass flow that is in the vapor state). So from the chart if you pick a quality of 0.81 and go vertically up to the pink line and horizontically over to the right-hand y-axis you read off about 0.66 This mean that the cooling duty at this evaporator exit quality is 66% of the peak cooling duty available with a evaporator exit quality of 1.
I hope this answers your question.
Kevin
Yep that sure does
Makes perfect sense now. Thank you ever so much for clearing that up, my brain thanks you
Michael St. Pierre
- Worked 15 years for Polycold Systems
- Now Self-Employed
- Manufacture Heat Load Controllers
- Also do contract service work on Polycold units
Side note: I usually don't respond to PM's or emails regarding the projects that I post in the forums. I feel it's much more fair to all, to answer questions within the forum topics themselves.
I thought this may help show that it is possible in an autocascade evaporator for a high-boiling point refrigerant to flow through an evaporator unnoticed. That is good refrigerant separation may not have occurred. This is my theory of the 1-phase-separator cryoBUG.
I just guessed at a refrigerant composition for the evaporator inlet of: mole fractions 0.10 iso-butune, 0.85 R-14, and 0.05 argon. The evaporator was modeled at a constant 2-bar absolute pressure. These are the results shown in two graphs. During cool-down of the cryoBUG the evaporator exit quality will shift from saturated vapor outlet conditions towards lower and lower quality – ultimately reaching an exit quality that provide just enough cooling duty to overcome thermal losses due to less than perfect insulation. As this shift occurs the evaporator exit temperature drops toward its inlet temperature.
In the first graph it can be seen that the first vapor to boil off is predominately argon (vapor composition 62% argon, 38% R-14, 0% iso-butane). As the vapor creation continues the composition becomes less argon and more R-14. Finally not till a quality of 0.88 does the first iso-butane boil. When this happens nearly all of the argon has already evaporated as well as a good portion of the R-14. The evaporating iso-butane quickly raises the evaporator’s exit temperature.
The second graph is meant to show how cooling duty falls off as evaporator exit quality is reduced. Hence, I would suspect that the cryoBUG’s evaporator exit temperature will rise considerably if cooling duty is increased and evaporator exit quality approaches 1.00.
Michael, I in no way am faulting your design or system – to the contrary, it is very nice. All I am doing is trying to understand it.
Kevin
You and me both
It has been an interesting journey thus far, and quite different then working with a 3-stage AutoC as I'm used to.
Over the last couple of days I've run two test charges with R290 (propane), and two more with R600 (N-butane). In all cases I'm still using R170, R14, and Argon as the lower boiling refrigerants. And in all 4 cases I have been using smaller amounts of the propane and N-Butane, and not experiencing any freeze-out issues. Although I am also not achieving as cold of temps as I saw with my first N-Butane test. What I am noticing, is that the suction stays relatively high up into the mid 30's. This I feel is detrimental to lower temperature operation. My theory is that with a greater amount of N-butane, I was seeing a lot of it slop over the phase separator about midway into the run, and then as the evaporator passed -140 C, it began to freeze either in the cap tube or soon thereafter. As it froze, the cap tube effectively fed less and less, which resulted in lower pressures and lower evaporator temperatures. This was great until the flow reduced so low, that there wasn't enough refrigerant gas expanding to carry the heat away.
So I will be looking at reducing cap tube flow intentionally, trying to find that sweet spot where I get the -150 C I am shooting for, but also having enough flow to deal with extra load if needed.
BTW I don't know if it was intentional, but your chart is based on refrigerants I didn't even use, and is missing some that I did (see below).
The two mixtures I am working with
Mix A:
R600 (N-butane)
R170 (ethane)
R14
Argon
Mix B:
R290 (propane)
R170
R14
Argon
However I would like to eventually switch over to isobutane (R600a) as you have shown. Just got to remember where I can buy small quantities from (Ultralo1 gave me a source once, but I can't find it).
P.S. Thanks for all the great charts, but I sure hope there isn't a test at the end of the week
Edit: Too bad there wasn't a reliable variable orifice device that got smaller as the temperature passing through it dropped. Sorta like a TXV, but not so mechanical in nature. Basically more like a sieve material that changed pore size based on temperature.
Last edited by mytekcontrols; 04-07-2012 at 08:23 PM.
Michael St. Pierre
- Worked 15 years for Polycold Systems
- Now Self-Employed
- Manufacture Heat Load Controllers
- Also do contract service work on Polycold units
Side note: I usually don't respond to PM's or emails regarding the projects that I post in the forums. I feel it's much more fair to all, to answer questions within the forum topics themselves.
Hello Mytek,
Such a masterpiece, I'm still impressed about your home-made condenser/sub-cooler.
I've just a question : do you have any advice if I want to make the last stage of my three-stages an auto-cascade pattern ? I'm thinking of using R14 with methane to achieve something around -150°C within a cooling chamber. I wanna try using a CPEV for R14, and captube for methane, what can you tell me ?
Keep going, I'm following that thread almost every day !
I've always thought that making the last stage of a cascade as an AutoC would have it's advantages. For instance, not having to run the suction at a vacuum for one. And the use of a CPEV is an interesting one, but not sure if that is necessarily the best approach. However with that said, the closest thing in my experience was the use of an CPEV between the AutoC's stack and compressor, with the purpose of keeping the unit from running too cold and freezing-out a particular refrigerant in the blend.
If it's automatic flow regulation versus heat load you are looking for, perhaps it would be better to use a TXV instead. Of course it would have to accommodate your unique application. Probably being a custom charged TXV.
And of course for ultimate reliability, cap tubes for everything would be the best way to go, since any regulation valve can fail over time.
Glad you are enjoying the discussion
EDIT: Just found this information that may be applicable...
Electronic Expansion Valve (EEV)
With an electronic expansion valve (EEV), you can tell the system what superheat you
want and it will set it up. The primary characteristic of EEV is its ability to rotate a
prescribed small angle (step) in response to each control pulse applied to its windings.
EEV consists of a synchronous electronic motor that can divide a full rotation into a large
number of steps, 500 steps/rev. With such a wide range, an EEV valve can go from full
open to totally closed and closes down when system is satisfied.
Last edited by mytekcontrols; 04-08-2012 at 08:45 AM.
Michael St. Pierre
- Worked 15 years for Polycold Systems
- Now Self-Employed
- Manufacture Heat Load Controllers
- Also do contract service work on Polycold units
Side note: I usually don't respond to PM's or emails regarding the projects that I post in the forums. I feel it's much more fair to all, to answer questions within the forum topics themselves.
Hey Mytek,
Thanks for fast answering ( and sorry if it's not " english correct ", I'm french ).
My initial goal was to use only R14 ( no AutoC so... ) with CPEV and a simple ventilated evaporator. In that purpose ( cooling chamber ), there's almost no load, except insulation losses, so I think I might be able to bring down the tiny volume of the chamber ( 50*40*40 cm ) closer to evaporation T°, and then playing with the CPEV ( at ambiant room of course ! ).
AutoC's are barely unknown to me, but I really want to give a try if it's worth it
Is anyone tried ethylene and R14 in AutoC ? Maybe it's a key to get low pressure ratio isn't it ?
Electronic expansion devices are clearly not in my budget ( maybe in another version ! ), so what do you recommend ? Cap tube, CPEV, custom TXV ?
Side note : in my actual two-stages cascade, R23 TXV makes clicky sound when closing down, like " tic...tic...tic " ( T° around -90°C ), don't know if it could gets any worse @ lower T°, since you're kind of expert, maybe you have an advice ?
Many thanks, keep going, this thread goes in my personal bookmark !
Vincent.
Posting Permissions
Reply With Quote
Bookmarks