Heat Pump Compressors Shutting Off Update: Mismatched Air Handler?

[jjnet]

Background of Previous Posts On This Subject:

All new equipment in first winter:

Trane XL19i 2TWZ9036A 3 ton two compressor heat pump.
Trane TWE037E13F variable speed air handler w/ strip heat backup (Trane 1410, one stage 9.6kw @ 240v).
Trane Zoning: 6 Zones
Trane Electrostatic Air Filter

Does anyone have an opinion on whether we have a mismatched air handler for our heat pump? Here’s the background.

Problem: 1st Stage 1 ½ ton compressor runs for 15 or 20 minutes, then appears (to the untrained eye) to shut off, and the suction line goes cold. HP stays off for a few minutes, then comes back on. This is not a defrost cycle. The stage’s contactor outputs remain hot with 240v even when the compressor stops working, and the outdoor fan continues to run.

(Defrost cycles are pretty clear because the outdoor fan stops, the zone controller opens all the zone dampers, and the strip heat kicks in, which causes the air handler fan to instantly speed up and make substantially more noise at all registers. Defrost cycles seem to happen at about ½ hour intervals and last from 1 ½ minutes to just under 3 minutes, depending on how cold it is outside.)

The second stage is even worse. It only runs a few minutes before going cold, and will not come back on for a long time. We’ve spoken to our contractor verbally and sent 3 detailed emails to 2 different addresses. (He’s occasionally told us that he didn’t get an email from us because we sent it to his wife’s address, even though we know he’s responded to other messages to that address.) He told us that it sounds like our discharge air temperature sensor is shutting the system down because it’s too close to the strip heating elements. We have not been able to communicate to him that the problem happens when strip heat is not active and that it has something to do with the compressors. He said he would check with Trane and get back to us. That was a full month ago and we’ve heard nothing.

We finally contacted another local Trane dealer a few weeks ago. Like every other HVAC company around here, they are incredibly busy. But after a few weeks they were finally able to work us into their schedule and sent two techs over yesterday to get pressure and temp readings to try to diagnose the problem. Unlike our original contractor, this dealer does not have extensive experience with Trane zoning. But like our original guy, they have not seen many 2 compressor systems and have only installed one other air/air heat pump (that’s one more than our original contractor, who is an expert in ground source heat pumps but not on air/air HP’s). They made it clear from the start that they would have a learning curve on our system, but could certainly pass the info on to Trane and get their recommendations.

They were here for about 3 hours watching what the system did in each stage of both heating and cooling and at many zone damper configurations. Here’s what they found:

The head pressures and temperatures went way too high (I think temp was up to 380 degrees) at about the time intervals we expected for each compressor. When it looked and sounded to us like the compressor was turning off, it was actually temporarily switching the compressors into a bypass mode to respond to the high pressures and temps. After it did this a few times it would finally have to shut down for an extended period, most likely due to internal thermal overload protection.

They thought the system might be somewhat overcharged, but felt that the air handler fan was running at too low a speed to get rid of enough refrigerant heat in the indoor coil. We experimented with the system for a long time, so they could see how the zone controller varied the indoor fan speed with different zones open and closed. We made all zones wide open by choosing set points of 90 across the board. We bumped up the fan from 400 cfm/ton to 450 and changed from Residential to Commercial air flow. We changed the strip heat air flow from 900 to 1100 In all cases, they felt that the fan should be running at a higher speed than it was. Or, if the indoor coil were larger, the system might work at the present air flows.

The possibility that the indoor coil might be too small brought up a conversation I had with Trane in Denver a few months ago, when one employee told us that we had the wrong air handler for our heat pump. He said that Trane does not recommend the TWE037 for this HP, but says to use the TWE040 instead. And all the Trane charging curve info on the heat pump Service Facts sheet is for the TWE040. These tables include adjustment factors for all the other Trane air handlers that should be used with this HP, but this list did not include our TWE037 air handler.

So at this point we are waiting for Trane’s interpretation of all this. If anyone has experience with similar components or symptoms, we’d really appreciate hearing their perspective on this.

In the meantime, to check the possibility that the discharge air sensor is involved, we recently added our own wireless temp sensor in the supply plenum. The hottest it’s ever gotten was 115, and that was with the strip heat on for 20 minutes or so. With all the heat pump testing yesterday it never got higher than that 115 point, even when the refrigerant was over 300. I don’t know what the standard discharge air threshold is, but they were pretty sure that it was well over 115.

In addition, to see if our ducting is at fault, we’ve made up a table listing each zone, its square footage, heat loss, heat gain and recommended CFM from HVAC-Calc. This table summarizes the duct sizes and lengths to each supply outlet and each return. We’ve set up this table as an HTM file with the following link:

http://www.frontier.net/~jj/HVACZoneDucts.htm

[BaldLoonie]

I found 2 things:

ARI listing for that unit shows only the TWE040

A-S spec sheet for their version of it does the same

So I'd say it shouldn't be on the 037 model

I know nothing about the 18/19 model nor Trane's zone panel and how it stages and handles airflow with 1 or 2 zones calling. If you have the shop or bedroom zone calling, even on low that thing should want to move 600 CFM through a 50 or 75 CFM duct. Those 2 little zones shouldn't be allowed to call for equipment.

Trane suppliers are required to keep a factory trained tech rep on staff, maybe one of these dealers can get him out to see you.

[jjnet]

Thanks a lot, Baldloonie. If the air handler is the problem, is there any point in checking into whether a larger coil in the same box is available?

[BaldLoonie]

Probably just have to swap air handlers. But that shouldn't cause all this.

So is your panel set up that if the 2 tiny zones call, the equipment will come on? I sure don't like that!

[davesrha]

May need a by-pass damper installed so when the smallest zone is calling and others are closed the bypassed air (extra air)may be rerouted to the return air plenum.

[jrbenny]

They have a bypass that is controlled by the zone control board.

Equipment mismatch is the problem.

http://www.aireng.com/clientuploads/training/APPROVEDCOMBNATIONS.doc

Outdoor coil is too large for the indoor coil. No room for the refrigerant when it wants to run indoors.

[jjnet]

Thanks again, Baldloonie and daversha. Yes, the two small zones are set up to be able to call for the compressors. We have a bypass system with three options: (1) Trane Relief, where the controller cracks slightly open the closed dampers for zone’s which are not calling, (2) 8” Bypass back into the return plenum, or (3) Dumping the 8” bypass into one 8” and one 4” duct.

Bypassing into the return plenum can cause the discharge air temp sensor to shut the system down, so we’ve avoided that so far, though we can easily switch to it. We tried the Trane relief setting and found that it was not ideal. It ended up dumping hot air into zones that were already too hot from our passive solar setup. We’re currently using the dump option, because this way we can choose where to send bypassed air instead of randomly dumping it everywhere in the house. We send it to two locations which we deliberately undercondition, our greenhouse and basement hallways.

I should emphasize that most of the time the testing was going on yesterday, we had the system running flat out with all zones calling for 90 degree set points and presumably the air handler maxed out with dip switch settings and maximum zone calls. The high pressure/high temp problem still occurs in this case where a bypass is not relevant.

[dash]

jjnet ,Did you miss the post just above yours?

[jjnet]

Yes I did miss it, Dash. While I was composing the message JRBenny jumped in and I missed it till just now. Thanks much to all. If we have a mismatch problem, we'll hope for a straightforward resolution. Right now, it would be far easier to swap to a 2.5 ton heat pump rather than replace the air handler with the TWE040. We'll see what transpires with Trane and our new contractor. Even though we've never worked with them until yesterday, they've been much more responsive than our original contractor ever was.

[jrbenny]

Unfortunately, that 037 won't work a 2.5 ton XL19i. Take a look at that link I posted. You'll need to change that air handler regardless of the ODU.

[jjnet]

Thanks again, JRBenny. That significantly clarifies our choices. We were strongly considering the 2.5 ton heat pump possibility, because our entire utility room, plenums, etc. were designed and built around the 037 air handler's dimensions.

By a stroke of luck the 040 will fit our supply plenum because the plenum had to be 26" wide to accommodate the required takeoffs. The bigger problem is that the 040 is 7" higher than the 037. If we throw out our electrostatic air cleaner, we can get those 7" back. The only other alternative is to cut down the return plenum's height by 7" and keep the electrostatic cleaner. Trane recommends 14" of plenum height below the air handler, so that's what we've got below the air cleaner. Unless the cleaner’s 7 vertical inches can be considered part of that 14", we're going to be out of spec if we try to go that way.

To switch to the 040 we're probably also going to have to move our water heater out of the utility room, because the slot we saved for it next to the air handler will simply be too narrow for the 040's full height 26" width. Though not ideal, that move can be accomplished, if necessary, with a bit of plumbing, and wall deconstruction.

I must absolutely thank you for being the single most knowledgeable and helpful person on the Trane component issues you’ve helped us with over the last couple years. We have spent literally hours on the phone with Trane personnel in Denver trying to get clarification on issues that you know cold. There appear to be only a few people in our state that have anywhere near as much knowledge as you do about Trane’s products. And you have done all of this when you’re not only not employed and paid by us, but as I understand it, not currently on Trane’s payroll either. We salute not only the power of concentrating knowledge and experience which the internet makes possible, but also you personally for so generously sharing with us the benefit of your years of experience. While I’m at it, thanks also to Dash and all the others on this forum who have helped to point us in the right direction as we’ve struggled along through this process.

[jrbenny]

When did you start the equipment? Did it work okay in cooling?

Thanks for the compliments. Good luck with your project. Keep us posted.


ps. I take paypal.http://www.smileypad.com/v10/Cache/Holidays/Dancing-Chilli.gif

Okay, not really.

[jjnet]

Though we hoped to get the system going while still in cooling season, it didn't happen. Ductwork and wiring were completed in August. Then we lost three months of waiting for our contractor to make up the plenums from our plans, put the heat pump on its pad, braze and vacuum check the lines and add to the default factory charge. So it worked out that we got the system up only this past November, almost two years to the day since we took delivery.

We ran it in cooling mode briefly to verify that it was switching correctly. The techs that were here last week checked out cooling behavior in both compressors. I'm not certain but I think they saw some variant of the same problems that showed up in heating. Their best guess at the time was not enough airflow through the indoor coil. I did tell them of the conversation I had with Trane where I was told we had the wrong air handler, and I printed out and gave them a copy of the table with Trane’s list of invalid combinations from the link you provided. That info was critical in letting us know our search for the problem was finally over.

The good news is that, using the strip heat, we've verified that the ductwork and zoning components seem to be working perfectly. We can set temps and count on each zone remaining right where we've set it with minimal noise and very brief use of the strip heat. We leave the fan on continuously because whole-house humidification was one of the primary goals of the system. Other than the cost of the electricity and replacement cylinders, the Carel steam humidifier is accomplishing exactly what we wanted.

In the meantime, we’ll have to approach our contractor and Trane about the mismatch. Everybody makes mistakes, and we’ve made our share. At this point we just want to be done with the project, regardless of the complications involved.

[Swampfox]

How do you run the fan continuously with a zoned system? any zone with the fan set to on will get heated/cooled when any of the other zones calls.

[jjnet]

Swampfox, the Trane zoning controller and master thermostat allow us to open only zones we want to when the fan is set to continuous on. If one of the dip switches on the zone controller is set to allow individual zone fan control, each zone has a fan on/off status that we can set. If no zones are calling, only those zones with fan set to on will have their dampers open. If one or more zones call, then their dampers are opened and the non-calling ones are closed. This has helped us send hot greenhouse air to our basement on sunny days, elminating the need to heat that floor entirely.

[mark beiser]

As others have confirmed, that air handler is definatly a mismatch for the outdoor unit.

What I havn't seen anyone else pick up on is that the system is likely over zoned, depending on how it is configured.
The smallest zone capable of independantly triggering the system to come on should be at least 20% of the load. That is straight out of Trane's zoning class and literature.

If configured correctly, the zone system will modulate dampers to attempt to satisfy every zone around the same time. The smaller zones, that are less than 20% of the total load, should be configured as "no vote" zones so that are not capable of calling for the system to come on unless one of the "vote" zones is calling.

With correctly matched equipment and the zone system configured correctly, no bypass is required, or desired. I know the control has the built in ability to run a bypass damper, but when i took Trane's zoning class last year, they were rather specific about not using it, especially on heat pumps.

There are a lot of dip switches involved with configuring the zone control, but there is a handy computer program available to the contractor where they can punch in the requirements for each zone, assuming they did the detailed load calculations required to properly size everything. The program then generates a printable form showing what every dip switch should be set at on the main control and all the zone cards.
A laptop or handheld PC can also be connected to the zone control to do diagnostic checks and view everything the zone system has done for the last 16 days.

Later this year Trane should have a new system out that is more plug-n-play.

[BaldLoonie]

I did express concern that 2 zones have capability of under 100 CFM and I thought should not be able to call for equipment to turn on. The poster said they have 2 zones being used as dumps so the excess air has a place to go. Assuming 1 compressor, that's 600 CFM that should be moved, I didn't like trying to force that into a 5" run!

[mark beiser]

Guess I missed that when I was skimming through the replys at 4am =)

[jjnet]

Thanks, Baldloonie and Mark. My duct table, to fit on one page, is a bit unclear. The dump can dump simultaneously to an 8" greenhouse floor diffuser, and to the basement hallways with a 5" ceiling register. We deliberately undercondition these two zones, so it's never going to cause a problem if we send excess conditioned air to them. The 8" and 5" ducts each have their own manual balancing damper from the bypass duct, so we can control how much goes to each location. So far we just leave both of them open.

We have set the dip switches on the zone controller and zone cards according to the percentage of the total cfm each zone can take, based on duct size. We've found experimentally that with this set up, as far as we can tell, the bypass opens only when one of those two small zones is calling by itself. As soon as any other zone starts calling, the bypass starts to close. I haven't checked the exposed damper shaft to see exactly where the bypass damper stops in each situation. But I know for certain that it opens all the way when only one of those two small zones is calling.

We have control over how close to fully open the bypass damper goes by setting its zone dip switches. Its zone damper is a modulating type and may be fine tuned in real time by the zone controller. It's possible that it cracks open in some instances that we haven't spotted yet. I would have to watch the open/close leds for that damper to see if it's actually doing that. But we felt it was critical that it opens fully when each of the small zones is calling by itself, and we experimentally verified which dip switch setting for the bypass damper makes certain that occurs. We felt that we should, if anything, err on the side of being more open rather than less.

As I understand it, the variable speed air handler fan can drop to as low as 1/3 of its max cfm speed. So only 400 cfm (minus the small zone’s cfm) needs to be pushed through the dump ducts. One 8” plus one 5” seemed like plenty for that.