HVAC Heat Recovery Unit

It could possibly be 'water hammer' if it was plumbed by an AC technician instead of a plumber.

Thanks Guest. I'm familiar with water hammer. Any idea why water hammer would be happening now when it wasn't a problem for 6 years? Could it be that a valve is not opening when the system starts?

If water is boiling in the hot water side of the HRU, you could get water hammer effects. Adequate flow (valves open, possible pump running?) would help prevent this.

Did you read the question- its a HRU on a residential heat pump- not likely to be boiling water is it!

Of course I read the question! If the discharge temperature of the refrigerant > 212°F, which is at least plausible; and if water flow is low, then boiling is entirely possible.

If you wish to learn more about "liquid hammer," check out the IIAR (International Institute of Ammonia Refrigeration) materials on "Vapor Propelled Liquid." There are some amazing transient phenomena that can occur. Expanding vapor can push high-velocity slugs of liquid ahead of it. Sudden introduction of cold liquid into a gas line can induce almost instant condensation, "sucking" a sizable liquid column in a reverse direction. These things have been known to blow elbows or caps off the ends of straight pipe runs.

If it is mounted on the wall you may have air trapped in an inverted loop.

A simple drawing or pic would help.

What you most likely have is a coaxial heat exchanger and in the trade we would call this arrangement a desuperheater. The water passing through when the heatpump is off makes it the coldest part of the system so the refrigerant migrates there and you accumulate liquid refrigerant in the coils. When the compressor starts it builds up the head pressure to a normal range around 240 psi with r-22 which puts the vapor temperature up to 105-100 after the system stabilizes. The hot refrigerant causes the liquid refrigerant to boil off causing "flash gas" which bubbles off through the liquid which is what causes your hammering. This is not uncommon. Your service tech/installer can call the coax manufacturer for tech assistance and find the right positioning to place the coax so the refrigerant will not accumulate in the coils. It's a quick modification not a major task. The excess vibration will cause metal fatigue in the copper tubing and could create a leak. It is a minor mistake in installation and it sounds like your installer is one of the 1% who actually takes pride in giving his customers a very efficient system rather than jam and slam.

Though I agree that this is probably the most likely cause of the vibration, whats changed? The OP said it has been working fine for 6 years and now has developed this vibration. So looking at Tornados response and combining yours as well I think there in lyes the answer.

There is probably a fractional horse power pump that is in the loop maintaining flow through the HRU and standard water heater so the tank does not get cold to allow refrigerant migration. If that Pump has failed then the HRU would have a tendency to "cool" allowing the refrigerant migration to occur resulting in the violent boil off when the system cycles on resulting in the vibration.

Of course with out having the piping diagrams this is my assumption of the situation.

In the normal course of events, the minerals in the water will form scale on the heat exchanger which reduces water flow and would raise the delta to the point of steam flashing off. A normal discharge temperature from the compressor could be in the neighborhood of 270'f. I assume that the owner is aware of this and would descale the coax occasionally and if it is the problem, why would the problem cease after a short period of run time. My first guess would be lower than normal temps this winter causing the effect I first described. A competent hands on trouble shooter can diagnose in under 5 minutes no matter the cause.

Actually, as in large tonnage chillers, scale build up reduces (insulates the tube bundles) refrigerant heat transfer ability, lowering the water temperatures and reducing efficient heat transfer usually resulting in surging or head pressure trips on the refrigerant side of the system. Which of course would result in the systems shut down on head pressure failures. If there were a constant flow through the HRU that should keep the HRU warm not allowing the migration of the refrigerant and good heat transfer which I think (as you originally posted) is the probable cause of the vibration.

Once the "boil off" of the refrigerant occurs, the system would then run fine as described. Of course if there is scale build up this would create an even greater chance of migration even with a pump as the water temperatures (due to scale insulation) would be likely below design usage.

So run some LSD (liquid scale dissolver) thorough the HRU, neutralize it and see what happens. All the HRU systems we have installed are installed with a pump and a VFD for demand purposes of the system. Or in the old days on houses we would install an in line Gurndfoss / B&G fractional pump, as the HRU is usually under a house, in a garage or out doors on a roof for small commercial facilities.

Weighing late on this, but my question is, Was this system serviced before the onset of the problems of shuddering? Could be Non-Condensables were introduced in to the system during the servicing and causing flashing during the system start-up then settles down after the gas cycles through the system.

Did you get it working?

Thanks for all of the suggestions. Sorry I haven't answered back sooner but my computer has been down. No, the system hasn't been fixed. One thaing I did notice is that the vibration is more severe, and lasts longer, the colder the temps (below freezing) are. I'll get a picture of the unit and see if I can figure out how to post it. And, I'll get in touch with the system manufacturer. Thanks again.

Don

I'm an HVAC contractor. I have a few suggestions. IF I was working on this unit the 1st thing I did after talking with you is to check to make sure all the isolation/hand valves were fully open (and I know I'm assuming they were installed, possibly they weren't). If they're gate valves bottom them out, fully open them then close them down 1/2 turn. 2nd I'd really want to get a good look at the valve that allows the water to flow into/through the heat exchange coil. The majority of time this is were your problem will be if nothing else has been altered on the system and now you have this situation. All this other talk about water boiling, delta, etc. is noise. These types of machines come with many different types of valves. Some have a manual lever that you can actuate back and forth to "exercise" the valve. Doing this several times (like 20+) will often clear the problem. The premise here is a build up of dissolved solids gumming up the valves moving parts. They work fine for a few years (depending on water quality and usage) then after being idle for a few months you go to use it and this suddenly pops up. You should also look for a high point where any trapped air might migrate. If the system was designed and installed properly some type of vent or mechanical connection should be there to bleed any air. Good luck.

Yes as mentioned before, descale, move the warm water through the heat exchanger and this will cause the the "Migration Vibration" (sounds like a good movie name) to be eliminated.

Back wash the unit, descale the unit and flush the unit. You could even do my old timer trick if you want to have fun. Get a rubber garbage can, put a piece of EMT in the can and fill the can full of water. Isolate the heat exchanger drop a submersible pump (plastic impeller) in the water and pour in LSD (Liquid Scale Dissolver) available from NuCalgon. Have the soda ash on hand to neutralize the mix and let her rip. Use the supplied PH strips to keep the descale mix hot (it will neutralize automatically when eating scale) and watch the EMT tubbing in the can. After a bit (length of time determined by exchanger size) the EMT will begin to change color, as soon as you see it begin to turn copper neutralize the mix and dump the can. Add clean water and flush the system out. Once the exchanger is free of scale the acid picks up a bit of the copper tube and the chemical reaction has it stick to the EMT. At this point if you could open it up you would see very shinny and clean tubes.

I have been doing this on large tonnage chillers for 30 years and it works every time (with a copper / brass exchanger) of course how long you need to descale is dependent on the exchanger size and pump flow 15 - 20 GPM is good for a 200 ton machine. It was this or the old messy Tube Rodder before the eddy current test as scale would give false readings on the eddy.

So have some fun with it and get the "Migration Vibration" Gone.