This is the point where we have moved beyond my area of expertise. Left to my own devices I would be resorting to trial and error.
Printable View
No problem. I'll work on a tentative specification next week, on my travels & post it up for comment.Quote:
GaryQuote:
Originally Posted by desA
At this stage, we could look at the design requirements for the VIC, though.
Fluid 1 : Refrigerant - R-134A - liquid
Entry condition : Liquid sub-cooled xx.xx'C off Tc,sat=75'C condition
Exit condition : T,exit = T,inlet - y.y'C
Allowable pressure drop = ?
@ mass flowrate of refrigerant.
Fluid 2 : Refrigerant - R-134A - vapour
Entry condition : ?
Exit condition : SH = 7K
Allowable pressure drop = ?
@ mass flowrate of refrigerant.
Minimum/maximum nozzle velocities for oil transport.
Minimum vertical height between nozzle centres.
What else?
This is the point where we have moved beyond my area of expertise. Left to my own devices I would be resorting to trial and error.
My initial thoughts are to design the VIC to the work case scenario & then rate its performance at start-up & mid-range. Would you think the worst case scenario would be at start-up, or hot-condition?
How would you know that you'd got every last gram of refrigerant out of the machine? What about the oil charge?Quote:
GaryQuote:
Originally Posted by desA
How will we know the mass charge in the system - in case of future servicing?
I would pump all of the refrigerant out and weigh it, then double check by weighing it all back in and seeing if the operating parameters are the same.
A further technical question.
Would you ever consider using a low back-pressure check-valve in the discharge line?
^ What feedback is there, in the field, regarding scroll-type compressors with low mass, disc-type check valves installed in the compressor discharge tube?
Are these discs always reliable - especially at continual high operating temperatures?
I have heard of failures in this aspect, regarding disc damage, & failing to seal off the compressor against reverse expansion. This has always been a particular concern.
Would addition of such a check valve fall into the compressor manufacturer's responsibility, or the refrigerant loop designer?
What performance degradation, if any, could be expected from the installation of a properly-sized, check-valve in the discharge line?
Why I ask, is that the lab test machine has a 1/2" check valve installed. This can be seen in the condenser/compressor gallery picture from yesterday.
In general any restriction in the discharge line puts stress on the compressor. I would not do it unless the compressor manufacturer requires it... and then I would want a rock solid reason for it. It's a pretty safe bet that if they require it, they have a rock solid reason.
I'm guessing it has something to do with pushing the scroll in reverse.
Now that we have managed to obtain, after partially insulating the condenser, at Tc,sat=75'C, a SC = 5.8K and approach = 7.75K, where do we go from here?Quote:
Gary:
For our purposes:
It can be assumed that the liquid will gain considerable subcooling between the condenser outlet and the TXV inlet. Thus far, our ideal SC seems to be 7K at the condenser outlet at 75C Tc,sat. I assume that this will rise well above the 8.5K minimum long before it reaches the TXV inlet.
Would we be looking to increase SC slightly more, back to 7K, by adding refrigerant, or is 5.8K sufficient for now?
What I'm seeing here is a trade-off for COP,hp between the additional heat-transfer to the water via sub-cooling, versus the lower compressor input power.
From a heat-exchanger design perspective:
Assuming an acceptable temperature-cross for the condenser of (-Tcross)=Tcond,exit - Twater,out = 0.9K, then for the last run, with Twater,out=67.25'C, the lowest acceptable outlet temp from the condenser should be:
Tcond,exit = 67.25 + 0.9 = 68.15K
Sub-cooling : SC = 75 - 68.15 = 6.85K
So, I estimate that we should try to increase SC from 5.8K to 6.85K at some point.
Partially insulating the condenser gave us a drop in SC, so I assume that fully insulating it would also drop the SC.
But at this point, since we don't have the proper insulation materials or the water regulating valve, and we are not ready to make the partition changes, nor remount the drier vertically... then what's left is to raise the SC to 6.85K or thereabouts and see what that does.
Agreed. We had a fairly large change in SC from 8.5K to 5.8K represents -31.8% change. That's significant.
Additional insulation will further affect SC, to be sure.
Very fair comments. Leave this with me, I'll get all the materials to hand during my trip next week. I suggest the following plan of action:Quote:
But at this point, since we don't have the proper insulation materials or the water regulating valve, and we are not ready to make the partition changes, nor remount the drier vertically... then what's left is to raise the SC to 6.85K or thereabouts and see what that does.
1. Modify partition separating condenser from compressor.
2. Properly insulate condenser & partition walls.
3. Test system response to these changes.
4. Re-position filter drier to vertical orientation.
5. Test system response to this change.
6. Install water regulating valve.
7. Test system response to this change.
8. Adjust PRV & iterate adjustment/testing.
^ :)
The water flow regulating valve, using pressure as reference input. What acronym would that be?
One of our initial assumptions was that we would not be able to exceed Te,sat=15C, but in fact we have gone considerably beyond this. I'm wondering if there is some max that we need to stop at. Possibly a max amperage?... or a max V*I? Any idea what point the internal overload trips at?
Revised plan of action:
1. Modify partition separating condenser from compressor.
2. Properly insulate condenser & partition walls.
3. Test system response to these changes.
4. Re-position filter drier to vertical orientation.
5. Test system response to this change.
6. Install water regulating valve.
7. Test system response to this change.
8. Adjust WRV & iterate adjustment/testing.
While I'm on the shopping trail, is there anything else I should get? Over here, due to distance from major metropolitan areas, I need to plan a little in advance in order to get parts & components.
Eventually we will need a temp control (preferably two-stage) for the fan, the center of its range being in the neighborhood of 100C.
And a multi-speed fan motor for it to control. Its physical size, shaft size, mounting and electrical characteristics should be identical to your present fan motor. Its top speed amperage should be at least equal to the present motor.
Ok. I'll chat to my suppliers in BKK & source a suitable temp controller. I do have two spare temp controllers - although not two-stage. I'll go through the manuals to see if these could be used, alternatively look for a better option.
Have you used phase-control devices for conventional single-phase motor control? These can be used to alter the fan motor speed & seem to be pretty evenly priced. I'd have to order in from the US for these.Quote:
And a multi-speed fan motor for it to control. Its physical size, shaft size, mounting and electrical characteristics should be identical to your present fan motor. Its top speed amperage should be at least equal to the present motor.
Very good point. This lower boundary of the compressor has been worrying me. I has expected that, Te,sat would have reduced a little, with the fine-tuning & do expect this, with a fan speed control strategy.
You're right that we need to check this. Give me a few hours to check my information & report back.
We have yet to see a load that the compressor cannot handle, but I assume this will eventually happen. Ideally we could hit the limit with Ta,in=25C, then find a much lower fan speed that would hit that limit at Ta,in=35C. We will need to do a lot of experimenting to find the ideal control points and fan speeds. A phase control device providing variable fan speed might make this process easier... and we could probably switch it through a single stage controller.
The finer we tune this and the higher the Ta,in goes, the higher that Te,sat is going to go... and that's a very good thing... until we overload the compressor... which is a very bad thing.
We want the compressor to work its butt off... but we don't want to give it a hernia.
Just short of overload is where we need to bring in the fan speed control... but we need to define "overload".
Good. That would suit me just fine.
Let me research it for a few days & chat to the suppliers. If it makes sense, on the back of the single stage temp controller, as the switch, then I'll get the parts ordered in. We can discuss more along the way as the info comes in.
How will we know when the critical point has been reached for the compressor? Are there clear warning signs - other than black smoke? :eek:Quote:
Gary:Quote:
Originally Posted by desA
Very good point. This lower boundary of the compressor has been worrying me. I has expected that, Te,sat would have reduced a little, with the fine-tuning & do expect this, with a fan speed control strategy.
The finer we tune this and the higher the Ta,in goes, the higher that Te,sat is going to go... and that's a very good thing... until we overload the compressor... which is a very bad thing.
We want the compressor to work its butt off... but we don't want to give it a hernia.
Just short of overload is where we need to bring in the fan speed control.
I've been researching the compressor manufacturer's technical documentation & application guidelines with regard to the Te,sat>15'C limit.
Nothing is mentioned at all regarding V*I limits in regards to the heat-pump application.
The most applicable document covers considerations for high ambient conditions, & then it concerns itself more with high condensing temps (a given for an AWHP), & issues for high Tc,sat coupled with low Te,sat, leading to high compression ratios.
The two main criteria they focus on are:
1. Compressor discharge temp - measured 150mm from discharge port:
1.1 Extreme temp condition : Tcomp,disch' = 135'C (275'F);
1.2 Danger level : Tcomp,disch' = 121.11'C (250'F);
1.3 Reasonable life expectancy : Tcomp,disch' < 107.2'C (225'F) <=== fair
2. Oil sump temp:
2.1 Conservative : Tcomp,sump < 93.3'C (200'F); <=== reasonable
2.2 Acceptable (welded comp) : Tcomp,sump < 115.6'C (240'F).
Remember, this type of scroll compressor uses the incoming suction vapour to cool the motor windings.
I would suggest that as long as we control the evap SH to be reasonable, & control the discharge temp & sump oil temp, then we should be safe to exceed the envelope right boundary a small margin.
How small is safe? The compressor manufacturer's idea is for the OEM to test the arrangement under the more strenuous conditions & satisfy themselves & the supplier, if needs be, that the duty is safe for the compressor.
We have kept the compressor well within these boundaries so far, and by insulating the condenser and partitioning the compressor into the evap compartment, we should keep the compressor even cooler.
All of the criteria mentioned (compression ratio, discharge temp, oil sump temp) plus the V*I will show itself as high discharge temp, as the heat from the motor transfers to the refrigerant being pumped through.
So... discharge temp seems to be the best control criteria for our fan speed control and we need to keep it below the 107C limit.
I'm thinking that as long as we do not exceed 107C discharge, we can continue to load up the compressor. On the other hand, if the compressor kicks out on internal overload, then we have gone too far and need to adjust our fan speed control to reduce the load and drop the discharge temp.
So far this compressor has shown no signs of weakness. If compressors could speak I think it would be taunting us, saying things like,
"Is that all you got, Sissy?"
"You hit like a girl."
"Where's that 35C air?... Bring it on."
We need to teach this compressor a lesson; Make an example of it. Kick its T,comp,b. It doesn't know who it's messin' with... lol
True. I began stripping the compartment wall this afternoon - will cut out a section & re-work upon my return in approx 1 week's time.
This is certainly what the literature seems to be showing. It makes good sense.Quote:
All of the criteria mentioned (compression ratio, discharge temp, oil sump temp) plus the V*I will show itself as high discharge temp, as the heat from the motor transfers to the refrigerant being pumped through.
Agreed. This is an elegant way to control the system - simple & direct.Quote:
So... discharge temp seems to be the best control criteria for our fan speed control and we need to keep it below the 107C limit.
This scroll does have an internal temperature cut-out/cut-in switch which is set to nominal 290 'F (143.3'C) / 140'F (60'C).Quote:
I'm thinking that as long as we do not exceed 107C discharge, we can continue to load up the compressor. On the other hand, if the compressor kicks out on internal overload, then we have gone too far and need to adjust our fan speed control to reduce the load and drop the discharge temp.
The system also has a LP/HP cut-out switch, to limit under & over pressures.
Water heat-up trial (brief feedback):
Performed a water heat-up trial from 28.7'C - 65.4'C. The duration was timed, & T,tank , Tw,out , current values taken at each time step.
The full analysis will follow (hopefully tomorrow morning).
General points of interest:
1. System began heating straight off the mark - no slouching, or temp delays.
2. It felt 'smooth' all the way through the range.
3. The heat-up performance improved by 10.2% over that for the machine condition before tuning & partial insulation began. !!!
We are now making progress & I'm extremely happy with the result. Pretty incredible... Thanks, Gary, so much for your kind mentoring along this path - it is very much appreciated.
Agreed.Quote:
Gary:Quote:
Originally Posted by desA
This scroll does have an internal temperature cut-out/cut-in switch which is set to nominal 290 'F (143.3'C) / 140'F (60'C).
Internal temperature runs about 50F/28K higher than discharge temp, so this would translate to about 240F/115C on the discharge line, making 107C an entirely reasonable target temp.
107'C seems to be a safe upper working limit, for long-term operation & a reasonable compressor lifespan. We can fix that as our target.
Note on run#4 the drop in voltage caused a rise in discharge temp (Tc,sup). The fan control strategy will reduce the load to help protect the compressor from the region's voltage drop problems as well.
Well spotted. I'd originally attributed the Tcomp,disch rise to the hotter incoming air onto the evap coil, but I see that you are correct.
With voltage drop, comes an associated current climb, until a threshold voltage of around 185V is reached. Below this, the amperage seems to run away & climb rapidly. After this, the compressor seems to cut out, & re-start This could actually be caused by the reduced pump flow forcing up the internal temp & pressure - resulting in a HP trip on the LP/HP trip.
With the associated current climb, on low voltage, comes increased base temp Tcomp,b.
I'd imagine then, that this increased power must be feeding through to the compressor discharge temperature, & so causing it to rise.
Well spotted.
I like this strategy more & more as it develops. This is an elegant way to smoothly manage the compressor. :)Quote:
The fan control strategy will reduce the load to help protect the compressor from the region's voltage drop problems as well.
That makes sense. I'll be meeting with my controller supplier tomorrow & will be training on the new controllers I'm now using. I'll have a chat to him about using multiple temp inputs & fan control. I'll post something one this once it all makes sense.
The current settings (I'll firm up when I return to my lab) as follows - in accordance with compressor manufacturer's guidelines (slightly conservative settings, below their limit recommendations):Quote:
What saturation temp/pressure do you have the HP control set at?
HP = 400 psi(g) = 2.757 MPa(g)
LP ~ 30 psi(g) = 0.207 MPa(g)