But, don't upset the second contractor. If you press him too much you may have to find a third. :-(

I know. I'm trying to stay on his "good" list!

So the system has been pretty normal for a week, here are some numbers. I took daily readings of pressure and temperature around the same time. Sorry about all the dots.  All the spaces kept disappearing.

DATE ....PSI OUT.... PSI IN ....LWT ....EWT
12/21 .....62.8 ........67.7 .......31.3 ....34.7
12/22..... 54.8 .........57.5 ......31.6 .....37
12/23
12/24 .....59 ........... 63.5 ..... 30 ......34.7
12/25..... 62............ 66......... 28.9 .... 33.4
12/26..... 65.3 .........69.7 ...... 27.1 ....33.8
12/27 .....SYSTEM LOCKOUT LOW EWT COIL FROZEN
................30 ...........59.9 ........NO WATER FLOW

I put the system on a/c. Within 15 seconds, the coil unfroze. Turned off a/c, waited about 15 minutes, turned heat back on.
...............31.3 ........ 36.7 ........35 ........38

That's where we're at. I don't really know how to interpret the numbers, except that loop EWT gradually decreased slightly over the week, but pressures continued to climb, although delta remained about the same. Am sending this on to my contractor as well. If anyone is really familiar with CM/GeoMax, please let me know.

System is completely back to normal now. House at 68. Howling wind outside. With wind chill, it's like -5 F.

To recap, I *did* disconnect the DSH. I figured it couldn't hurt, and might help. That was disconnected last week. Delta of pressure indicates about 12 gpm (3 gpm per loop), 4 ton system.

That data suggests your anti-freeze protection is good down to just 27 or so...but we all kinda knew that.

Ice formation, well, "snowballs" (pun intended). Once some ice forms, it acts as an insulator, impeding heat transfer. That causes the refrigerant pressure and temperature to drop, causing more ice to form. Once enough ice forms to restrict fluid flow, its all over.

Your data on 12/26 hints at incipient icing - delta T rose from 4-ish to almost 7. Delta-P did NOT rise. That tells me that some ice was present on 12/26. Understand that a coax with some icing is not the same animal the flow / pressure tables reference. In other words, you can no longer accurately infer flow from pressure delta since the flow has become restricted. It would be interesting to have had data for pressure across the flow center at the same time...I'd bet a six pack of very fine IPA that it would show a rise just as delta-T across the coax began to rise.

Once flow stops completely, Delta P across the flow center is equal to delta-p across the unit since nothing is moving.

Have you frozen some of your loop fluid in a freezer? Do so, perhaps in an ice cube tray. Dump the cubes in a tall cup with a probe thermometer. Let them thaw a bit. As more liquid forms, stir and note temperature of the solid / liquid mixture. That should show your freeze point.

I'm suprised the loop pumps will build a 30 psid delta-P. It can't be good for them, though. They are cooled by moving liquid. If they get really hot the surrounding flow center could heat up to the point of deforming the plastic pipe connections which could then leak. I hope the pumps are thermally protected.

Workaround: Keep a close eye on LWT. Whenever it hits 28, switch to AC as you did this time. No need to wait a full 15 minutes before returning to heating operation. I'd leave the DSH off until the present cold snap passes. Consider using supplemental heat to unload the system a bit.

For now, you are your own "defrost control board" (a part found on every air source heat pump that switches the unit to cooling to deice the outdoor coils)

Keep us advised...

Curt,
I just started the ice cube test. I took about 3 ice-cubes worth of antifreeze out of the ground loop and stuck it in the freezer. Then I went out the shovel the path and the cars. It took about 30-40 minutes. When I came in and checked on the ice cubes, a layer of ice had already formed and the liquid was full of ice crystals. I put the probe thermometer in and got 28.8 F.
I will continue to let them freeze, then allow them to thaw in the fridge and keep an eye on temperature.

I'm assuming that with crystals already forming at 28 F, that means that when the EWT gets to around 28/29, I'm clearly getting crystals in the loop and those are gumming up the coil?

The manual talks about a jumper that is the low temp cutoff.  If the jumper is connected, the cut-off is 30F and is the setting for water.  If the jumper is cut, the low temp cutoff is set to 10 F, which is the "antifreeze" setting.  The jumper is cut on my system.  But with ice crystals forming at 28, I basically have water in my system?

Crystals may not "gum" coil up, but there is a reason we want folks to keep antifreeze levels such that you are protected down to the teens.
Your loop design may not take temp lower than 28* but your heat pump would run lower so we need to correct this.
J

Posted By engineer on 27 Dec 2010 09:44 AM
Dump the cubes in a tall cup with a probe thermometer. Let them thaw a bit.
As more liquid forms, stir and note temperature of the solid / liquid mixture.
That should show your freeze point.

Just a small nit-pick: Unlike pure water, the temperature of a water/antifreeze
solution does not remain constant as the mixture freezes. The first ice crystals
to form are essentially pure water -- so the concentration of antifreeze in the
remaining liquid rises, and its freezing point falls. For a practical and joyous
application of this principle, Google: Eisbock

Everyone must believe in something...

...I believe I'll have another EKU 28,

Looby

Ok, I dumped the cubes in a cup, mashed them up a bit, and waited for them to start to thaw.  When I had a mostly chunky mix of "ice" and water, I took the temperature:  24.8 F.



Also, I *think*, based on asking a lot of questions, that the antifreeze in my system is: ComStar International brand, 100% glycol "Freeze Safe".  My original installer remembers that it was sold in black 55 gallon drums, and the supplier where he got all the loop tubing from confirmed that they sell ComStar brand glycol in black 55 gallon drums.

(Added: now it's mostly slushy ice, temp of 25.5 F.)

Posted By stuart.wyss on 27 Dec 2010 04:44 PM
Ok, I dumped the cubes in a cup, mashed them up a bit, and waited for them to start to thaw.  When I had a mostly chunky mix of "ice" and water, I took the temperature:  24.8 F.



Also, I *think*, based on asking a lot of questions, that the antifreeze in my system is: ComStar International brand, 100% glycol "Freeze Safe".  My original installer remembers that it was sold in black 55 gallon drums, and the supplier where he got all the loop tubing from confirmed that they sell ComStar brand glycol in black 55 gallon drums.

(Added: now it's mostly slushy ice, temp of 25.5 F.)

So here are two possible explanations:
1. Your original installer did not mix it correctly. 2. It does not contain 100% glycol. Looking at the spec sheet it gives you the mixing table for 100% PG, but it also says that the actual product contains rust inhibitors etc and is a blend, so it does not contain 100% PG. It is made in China, sometimes those little details are lost in translation. Many of the blends suggesting they contain 100% PG do not. Been there, done that....
One of the reasons we switched over to methanol.
http://www.comstarproducts.com/uploads/files/Frsafe5gal.pdf
So Stuart, thanks for the update, but what you really need to do is fixing your anti-freeze issue ASAP. You are putting your piping, heatexchangers, circulation pumps etc at risk!

So...can we take it as read that my freeze protection only extends down to about 27 F? give or take a few degrees?
This would be based on my "freeze test"?

And it should go down to something like 10 F.  Both of the reasons given indicate that the anti-freeze is definitely not correct for some reason? 

I asked my new installer about methanol.  However, Rehau, the company that makes the piping/manifold, said that their product is NOT for use with methanol, only with the glycol mixes.

I could ask my new installer to flush the ground loops completely, and fill them up with a proper mix of glycol from a reputable company?

2 Things here. The heat exchanger is usually a bit colder than your LWT, this is why the freezing occurs there when the circulation pumps turn off. Yes, your antifreeze is off, based on the fact that your source water freezes up in your heat exchanger at temperatures not even close to the "should be" freeze point. And yes, your freezer test confirms this. If you cannot use methanol, have your installer drain out some liquid by pumping pressurized air into the loop, and fill up the void with pure antifreeze. No reason not to use the same product, just use more of it. Assume that it is 60% glycol, and 40 % other stuff, that is what usually the glycol blends end up with. People who where this happened to usually had freeze point around 23F, pretty close to where you are. You can do the math ahead of time to know how much to drain out and how much of a 60% solution to use to refill. We usually shoot for 15F protection, which equals 25% PG concentration. Yours seems to be around 15% right now, which fits the 60% number I gave you. Trust only the refractometer, nothing else!

Ok, the new guy usually uses methanol for all his WF systems, so he had to order some glycol. He's coming tomorrow to put more glycol in or change the mix or whatever.

In preparation, I drained another 1/2 cup of "original mix" and am keeping it safe. LOL, it's not that I don't trust the new guy, I would just like to have evidence of the bad mix before it gets fixed.

Assuming all goes well tomorrow, I could repeat the freeze test with both the old and new mixes again and compare? I could also get hold of a refractometer and measure the % glycol/freeze protection of the old and new samples?

Is there any way that I can or would need to add more glycol myself in the future, or is this something that homeowners (even ones as savvy as I am!) would never do and need really specialized equipment? Maybe I'm feeling overconfident because I have a pressure gauge and I'm not afraid to use it on a p/t port....

If you have someone coming to add glycol, they SHOULD have the proper equipment to determine the percentages!

Bergy

Bergy, I do.  I'm thinking if I have to prove to my old guy the difference (because I don't think the original guy knew how to calculate the %), could I use a refractometer or something? 
I am very happy that my new guy is adjusting the mix to the proper level, and happier as well because he's been nothing short of amazing and patient in getting all the kinks out that the last guy installed.    The old guy couldn't even tell me what brand glycol he used.

The pressure in your system, assuming a closed pressurized system, is rising as the loop field gets colder due to contraction of the loop pipes with colder water/antifreeze. When the loops get colder, the pressure in the system will rise. In the summer, the opposite will happen. As the water gets warmer, the pressure in the loop field will drop. We see this as well as our owners see this since we install PFU-100 on every system that has a pressure gauge, from Flow Center Products. This is just a side note to try to explain something that I'm not sure got answered before.

We typically see a 5-10psi swing on our loops, but we typically don't see a huge swing on temperatures.

Looks like you are headed in the right direction to start with and get the antifreeze right to start with. The next question is what is the loop field doing longterm and are you getting enough heat from the loop field with the REHAU system. I am interested since we do have a supplier that is wanting us to install REHAU, though we've been reluctant since we have had such good luck with standard HDPE loops. Hope all goes well with the antifreeze correction.

Dan

www.cogeothermal.com

Posted By decafdrinker on 31 Jan 2011 06:13 AM
I'm thinking if I have to prove to my old guy the difference (because I don't think the original guy knew how to calculate the %), could I use a refractometer or something?

A refractometer is the way to go. It must be calibrated with distilled water.

Bergy

I'm not familiar with refractometer operation. Will the information be useful if the loop was filled with softened water (salt in the water after treatment) and an antifreeze, or will the results be distorted?

or will the results be distorted

Distorted, but the magnitude of the distortion is negligible compared to the accuracy with which antifreeze is mixed.

Posted By ICFHybrid on 31 Jan 2011 04:07 PM
Distorted, but the magnitude of the distortion is negligible compared to the accuracy with which antifreeze is mixed.

While I appreciate your response, based on other threads I have questions regarding your experience with this topic.

Please explain how you know your statement is correct?  How do you know it is negligible, and what is negligible to you?  How do you know the accuracy with which the antifreeze in our loop was mixed (even I don't know that, and I own the loop)?  Please give details regarding your experience with refractometers and their results when methanol and salt are contained in the water being tested?

I know the protection that is supposed to be in our loop, but I would like to confirm it.  Given my earlier question, as to softened water, I am unsure as to whether a refractometer is the right way for me to go.  I hope someone that knows the answer can advise me.

While I appreciate your response, based on other threads I have questions regarding your experience with this topic.

Based on other threads? The only thing "other threads" might possibly show is that you have a limited understanding of problem solving approaches and that doesn't reflect negatively on me.

As for "this topic" I have a degree in Chemical Engineering. Chem E., you know? We might know something about refractometers.

How do you know the accuracy with which the antifreeze in our loop was mixed (even I don't know that, and I own the loop)?

Well, let's examine that for a moment. If you were to mix up some antifreeze, how would you do it?

Posted By ICFHybrid on 31 Jan 2011 04:57 PM

While I appreciate your response, based on other threads I have questions regarding your experience with this topic.

Based on other threads? The only thing "other threads" might possibly show is that you have a limited understanding of problem solving approaches and that doesn't reflect negatively on me.

As for "this topic" I have a degree in Chemical Engineering. Chem E., you know? We might know something about refractometers.

How do you know the accuracy with which the antifreeze in our loop was mixed (even I don't know that, and I own the loop)?

Well, let's examine that for a moment. If you were to mix up some antifreeze, how would you do it?

Funny that you have included No Answers to my relevant questions on this topic.  I tried to give you an opportunity to explain - My bad.

For the benefit of others, ICFHybrid has no geothermal training, installation or repair experience, or a geothermal system.  Beware of his advise and wasting time answering his question.  Got tree?