GSHP short cycle

tabnodak

Anyone have an idea on what could be causing this problem? 4 zone GSHP hydronic floor heat controlled by Tekmar 519 stats with aux floor sensor activated. Room air temperature used to sense demand. Stat calls for heat ("heat on" indicated), zone pump and geo start, stat senses that demand is satisfied ("heat on" indicator goes off), zone pump and geo turn off - this cycle lasts approximately 40 - 60 seconds. Does not appear to be isolated to one zone or stat.

Rich_49

Is there a buffer tank ? What model heat pump ?

tabnodak

HSS buffer tank, Geo Comfort GWT060B. Floor temp sensor shows temp room air temp. Stats often call for heat at or a degree or so above set point, runs for 40-60 seconds, stats turn off the "heat on" indicator and geo shuts down for the 5 minute bleed down period.

tabnodak

Above should read, "Floor temp sensor shows temp above room air temp. Room air is the set point.

Rich_49

Sounds as if the buffer tank was sized incorrectly . Proper sizing calculation would be taken between the system and HP . Your heat pump should not short cycle in the way it is .

tabnodak

I was thinking a stat problem. Can't figure why does it calls for heat, then satisfy in 40 seconds? Doesn't seem like the zone pumps could circulate enough water in 40 seconds to raise the temp, and don't think the geo can provide heat that fast.

Wondering if anyone has experienced a problem with the Tekmar 519 stats?

Harvey Ramer

Is your floor sensor laying right up against the tubing in the floor? That would do it.

Ironman

Is your floor wood or slab?

Rich_49

The only call for heat the heat pump should see is from the buffer tank . The sensor in the tank should turn the heat pump on and off based on the temp in the tank . There should be enough heated fluid in the tank and it should be set up and sized properly so the call from the tank will require the heat pump to not short cycle . Let's not get off on a tangent here . However Ironman and Harvey do bring up valid points it still does not change the fact that a call from the stat or floor sensor should not communicate with the pump at all .

tabnodak

Appreciate the dialog, I'll respond to questions:
Ramer: tubing is stapled to 2" XPS, covered with 2" sand, covered by unfinished concrete. Believe floor sensor was installed in slab.
Ironman: concrete slabs, no floor covering.
Rich: understand, think geo looks for temp less than 105F in buffer, then starts if there is a call. This appears to by working. This is set up as a cold system, so geo rarely shuts down because of buffer reaching high temp(110F). Geo normally shuts down at end of zone calls, regardless of buffer temp.

These are all good questions. I keep coming back to the stats, why should it be that with the set point 68 and the stat calls at 68(as indicated on stat), then runs for 40 seconds, but stat still reads 68. Have also see a situation where stat in different zone (set point 55F) calls for heat when room temp is 59F, geo runs for 40 seconds, stat "heat on" extinguished, geo stops? Curious!

Wish I could get more info on Tekmar 519, does it have " smarts" that some how try to balance floor and air temp? Or a timer that "exercises" a zone on some set of conditions by calling for heat for a 40-60 second period regardless of the room temp, or maybe on a timed interval? Could this be a feature of the stat when the room air temp is used in conjunction with the aux floor sensor?

Rich_49

Unfortunate thing about all GSHP is that power used by the Source side pumps is not required to be factored into the COP numbers . has been this way for a few years now ever since affected parties lobbied to have the numbers skewed that way to only highlight the HP efficiency . Nobody tells you about those power hogs on the other side , if many even realize this is the case .
Now to the radiant tubing part of this . Were room by room calcs done ? If so , was tubing spacing altered to different centers in the different rooms ? Sounds like many parts of this job went awry . Probably not zoned taking into account , Flux , solar gain , use groups .
Again , the buffer tank - heat pump communication is not properly set up . If it were this condition would not exist . Buffers properly sized take into account the high temp , differential and system conditions to eliminate short cycling . Someone did not do something the right way . Everyone wants to blame controls as opposed to somebody . Simple fact is that these t stats should never communicate with the HP just the system side pumps , controls and the like and communication with the HP should be left to the Tank .
What kind of temps are running in the loops , is there ODR set up , what is the pumping and control strategy on the system side . How many and how deep are the VBHs (vert. bore hole) or horizontal loops , whichever the case may be ?
Again , I think we will witness a poorly designed or installed system by one or more parties that though having a good reputation may not really be up to the task .

tabnodak

Slow down Rich, I appreciate that you may have a great deal of experience with properly and poorly designed systems, but I'm not quiet ready to throw my HVAC professionals under the bus.

Seems like I may not be doing a very effective job of describing the issue as I am experiencing it. Hope I'm using the correct terms and am not causing any confusion. I'll try again.

As I understand the system, the stat communicates with the buffer control board, which turns the zone pump on. Then if the fluid in the buffer is below 105F, the buffer controller tells the HP to start. When the stat tells the buffer that the demand has been met, the buffer controller turns off the zone pump, and if there is no other zone calling, the buffer controller tells the HP to turn off.

This is the sequence, the question remains, how can a demand at the stat be satisfied in 40 seconds?

Floor loops are running at buffer temp. (90 - 110F) out and come back about 15 degrees lower. No outdoor reset. 6 - 200+ ft vertical wells.

If you are still convinced that the system is improperly designed and installed, there doesn't seem to be anything more you can help me with. I'll say thanks for your comments, I appreciate your time and your willingness to discuss my questions.

Ironman

Try disconnecting the slab sensor and see if it changes any thing. If so, then it's probably too close to the tubing as Harvey suggested.

I ve used several 518/519 stats and never had this issue.

tabnodak

Thanks for your comments.

Disabled the aux floor sensors on all 4 stats, we'll see if things change.

Harvey Ramer

Let us know. We all have a great deal of curiosity

tabnodak

May be getting closer, don't think I've had the problem since turning off the floor senosrs. Have discovered Tekmar 519 uses "pulse width modulation", but don't understand it other than it seems to compare floor and room temp and adjusts call time to limit over/under shoots- is that why there is no anticipator?

Basement Floor temp and room temp stay within a few degrees of each other - 68-69F.

Garage and workshop differential is somewhat larger, but only by a few degrees.

Am I on to something?

SWEI

Pulse Width Modulation (PWM) varies the duty cycle (on versus off time) to effect a proportional response using an on/off zone valve. It works well with high mass emitters.

tabnodak

That being said, if you have read through this discussion thread, maybe you can help me with these questions.

Is the high mass emitter my floor slabs?

The duty cycle being controlled is my circulator pumps based on tstat settings?

If the room temp is controlling the call for heat on the Tekmar 519, What influence does the floor sensor have on the PWM and ultimately the duty cycle??

tabnodak

SWEI,
I have been reading some of your responses regarding radiant stats, anticipators, PWM, etc. and believe you may understand how this Tekmar 519 functions.

It seems to me that when the aux floor sensor is active, it functions in the PWM mode, and when the aux floor sensor is deactivated, it functions in differential mode.

So, the "40 second" call for heat by the tstat I was experiencing with the aux floor sensor may be related to a short PWM??

I'm not an HVAC expert, only trying to understand my system. I understand that mine is a cold system and I know it can't be good to short cycle the HP, because when the HP turns off, it stops at whatever the buffer temp is at the time and it goes into it's 5 minute pressure bleed down mode. Basically, if the buffer is below 105F (start temp for HP), I have expended the energy to start and run the HP for 40 seconds, somehow satisfied a call for heat, and not raised the buffer temp 1 degree. If another zone calls, the circulator may run, but it will be circulating water at the buffer temp until the 5 minutes is completed. I suppose some heat transfer may occur during that time. Maybe it's good to take more heat out of the buffer.

Seems like the PWM would work well in a hot system where the buffer tank would be supplying max temp water to respond to a call for heat and the PWM tstat might be satisfied without starting the HP.

If the solution is to leave the aux floor sensor off, I lose the capability to monitor/control min and max slab temps. Does this sound like a design issue with the system or an operational issue with the tstat?

Would appreciate your thoughts on all of this.

SWEI

tabnodak said:

Is the high mass emitter my floor slabs?

Yes. In engineering-speak, the mass of the slab becomes a low-pass averaging filter that changes the on/off flow signals to a very slowly changing temperature. Once we understand this, we have the ability to control the speed of the flywheel (slab temp) by slapping it every so often as it spins by. If we want it to go faster, we slap it a bit longer each time it comes around.

the "40 second" call for heat by the tstat I was experiencing with the aux floor sensor may be related to a short PWM?

Sounds exactly like that, especially if the 40 seconds is repetitive.

If the solution is to leave the aux floor sensor off, I lose the capability to monitor/control min and max slab temps. Does this sound like a design issue with the system or an operational issue with the tstat?

Could well be - I don't have more than basic knowledge of Tekmar's PWM strategy. I'm working now to develop my own (in an attempt to replicate the precision control we get when using proportional zone valves.)

tabnodak

Thanks for your comments.

Question, is there any validity to the logic of modifying a cold system such that when the buffer temp calls for the HP, and the hp satisfies a zone, the HP continues to run until all zones are satisfied AND the buffer temp reaches the high limit? Then when the next zone called, and if the buffer temp was above the HP start point, the zone pump would start, circulate higher buffer temp water, and in the case of the possible PWM "short slap" of 40 seconds, the zone might be satisfied without starting the HP?

If using the PWM is a good thing, seems like this might be an efficient solution.

Seems like the fewer times the HP starts would be better use of electricity and eliminating the short run times would be better use of the system?

Or am I not considering something, such as changes in HP efficiency as temps increase? Your thoughts please?

SWEI

PWM would be a very bad match for a heatpump. It should be controlling either a zone valve or a zone pump on the distribution side of the buffer and not actually issuing a call for heat to the heatpump itself. That should be done purely on the buffer tank temp and "ignore" what is going on on the secondary side.

tabnodak

Appreciate your comments, just want to make sure my explanation is not causing some confusion.

As I understand it; tstat makes the call to the buffer, buffer starts zone pump and makes the call for the the HP to start if buffer temp is below 105F. When tstat is satisfied, it talks to buffer, buffer shuts down zone pump and HP, regardless of what the buffer temp might be (could be anywhere between 85F-110F).

I'm just suggesting that it might be a better use of electrical energy if the zone pump would shut down after the call was satisfied, BUT, the HP would run long enough to raise the buffer to 110F. Then the next time a zone called or PWM called for a 40 second "slap", the buffer temp might be high enough to be able to satisfy the zone, or run out the time for the 40 seconds and the buffer might not have to short cycle the heat pump.

Sorry if I have confused the issue.