Garn water temp - difference between front and rear?

[Gooserider]

Not a Garn expert, never seen the inside of one, but I don't understand why the Garn would work any different from other storage setups, where mixing is generally considered undesirable... I would also think it best to get the hottest supply water from the top, and return the cooler water to the bottom, hopefully in a way that would discourage mixing...

It would seem that unless there is hardware inside the tank that would be in the way, that it should be possible to add an extension to the top intake pipe with minimal draining. Just pull down low enough to get that connection above the water line, and pull the fitting. Make a setup with an extension that you can screw in, then mount it in the hole. I would probably go in the top manhole and try to rig up some support for the pipe as it goes by.

I'd probably try to make the extension long enough to almost reach the front of the tank, cap it, and drill a series of holes in it, possibly making the holes larger near the far end, such that the area of the holes was far greater than the area of the pipe. Position the extension so that the holes are on the top side. Alternatively, not put a cap on the end, and then you'd be pulling from the same general area as the front thermometer dial...

Either way you would be preventing any potential short circuiting, and hopefully pulling from the hottest portion of the tank, without inducing any added turbulence.

Gooserider

 

[foxt]

The garn shouldn't work any different than our propane tanks!! we might have 180* on the top and 120* on the bottem and can still feed the baseboards with hot water.
leaddog

That's what I thought would happen. I get the top/bottom stratification; what is puzzling me is the variation that apparently sets up in at least a couple of GARNs front-back after the burn is done and the tank has been supplying heat for a few hours. I bet that you don't see the same front-back variation in your propane tank after it is charged to temp and as it draws down on the stored BTUs. When my GARN is behaving like any other storage tank and is drawing down the stored BTUs, not only does it set up top-bottom stratification, but there can be as much as a 20* delta between top front and top back of tank.

That difference has to be due to something unique about the way that the GARN behaves when it is just a storage tank - most likely related to the presence/geometry of the burn chamber at the front of the tank? Otherwise, like you said, it's just a big storage tank after the burn cycle is completed, and it should behave like one (no front-back variation). What am I missing?

Tom

 

[TCaldwell]

leaddog, i agree on the tank stratification, after the tank is fully charged. i think it takes alot of gpm for a short period of time[ length of burn] to fully charge the non pressureized tank .most time i wait till 135deg to fire, 2000 gal up to 200 deg. not sure that ceramic thermocouple cover will last, it looks like the probe tip is not protected and the lead wire not having a protective sheath, unless it is already outside of the boiler at that point. the other part that fails on thermocouples is the potting material that covers the connection of the probe to the leedwires, needs to be temp rated. sometimes it is worth having a thermocouple built to your needs, not expensive, espically if the install is difficult

 

[Rick Stanley]

Just "for what it's worth" info, thought it was interesting.....................................

The fire had been out for 12 hrs, the pumps had been running as needed for 12hrs (continuous for last 2 hrs).

I calibrated my 2 temp gauges on supply and return of garn and the 2 on the hx (one in from garn one out to oil boiler). I adjusted all four so they read the same as a digital kitchen thermometer w/meat probe.

The gauge on front of garn said 168 2 hrs before any other readings were taken. After the pumps ran continuous for the 2 hours, it read 162, so I was moving heat into the house. Warm and cozy.

Anyway, what's interesting is that with the garn/front gauge reading 162 (12" below water surface), i reached through the manway and stuck the meat probe in there and it read 171 (2" below surface 48" back from front gauge), the gauges in the supply AND return on the back of garn read 120, the two gauges in the house read 125 in from garn and 130 out to oilboiler. So I had reached the tipping point I guess and was backfeeding btu's the other way it seems.

I have the oil burner disabled so I'm gonna just let the pumps run and see if I can EVER get he heat that appears to still be in the garn, the front of the garn.

So, Foxt, I'm more interested than ever in finding out what kind of submersible pump you dropped into your garn. I have a couple of those black plastic submersible sump pumps kicking around here. Is that what you used? wondering about the high temps???

Anyway, like I said, for what it's worth

 

[foxt]

Rick,

The pump I dropped in there for that experiment was indeed one of those black submersible sump pumps. IIRC, it was a Flotec 1300x, I think it's something like 1400gph max. As I reported, it equalized front/back temps within minutes. Having said that, I am sure it also disrupted whatever vertical stratification existed, and I don't know if that is a good or bad thing, or if in the long run it hurts more than it helps.

You're seeing a much bigger delta between front/back (almost 40*) than I think I have ever seen on my WHS2000. I think my max has been about 25*. Can you confirm that when you are nearing the end of the burn cycle all of your temps are close (mine are)?

As you said, and for me as well, this is all about getting the most out of that tank between burns. I'm surprised that the more experienced GARN crowd hasn't been as bothered about this as me (and maybe you). I keep wondering if it's more of an issue with the way that our systems are configured, or if we're operating under some flawed assumption that such a big difference between front and rear temps means we're not getting the most out of the GARN when it is in storage mode?

Tom

 

[Jim K in PA]

Tom,

I have been following this thread with interest. I have not noticed much of an issue with my system. I am running a P/S piping arrangement right off the GARN. The primary loop is all 1.5" black pipe, with ~ 7 elbows, and only about ~20 feet of pipe. Have not calculated the head loss, but it ain't much. My primary pump on that loop is a nice fat Grundfos UPS 43-44 3 speed unit running on low. Even if I have 5-10 feet of headloss, she is still circulating 20-30 gpm. Even if the GARN had no extension pipe into the return line (which it does), I doubt it would be short circuiting.

Even without a continuously running pump, convection will eventually equalizes temps horizontally, while stratification takes place vertically. I cannot fathom how you are getting 170 degree readings at the manway, and 120 degrees drawing off the top of the tank 4 feet away. I am at a loss there.

I have not checked temps in the piping much this year since things have been running so well for me. I re-piped my in-house distribution and included a temp/pressure guage just downstream of my circ pump in the house. I rarely see much more than a 5-7 degree difference between the front temp gauge on the GARN and the temp gauge in the piping. And there is a FP HX in between them, too.

With outside temps at night still at or just above freezing, I am running my BB temps down to 130 degrees F quite comfortably.

Doubt this helps, but not sure what to offer.

 

[foxt]

Jim,
Thanks for that. Just to clarify - the max front/back delta I see is about 25*, I think Rick reported that he saw over 40* this morning. He and I both have a single loop off of the GARN feeding our distribution system via fphx. You, and I think most other GARN owners, are p/s. I keep thinking that's the difference that is contributing to this somehow. Maybe we dump a higher volume of cooler return water into the tank at a slower rate than you do, and perhaps that's driving it. The configuration of the tank internals at the front might be contributing as well, somehow trapping the warmer water up near the front gauge. At one point, I thought that the actual "hot spot" at the front might be extremely localized (as in just right there at the gauge), but Rick's analysis this morning tends to indicate that this is not the case.
I don't know what to do about it, other than resign myself to running the front temp down no lower than 140-150 (which supplies 120-130 at the rear), which just means that I might fire more frequently than I have to. I don't think I want to run a circ in the tank because it will disrupt whatever vertical stratification I do have. I could repipe to p/s, but I am not really sure it's worth a try (lots of work at the start of the heating season for a fix that might not really correct the situation).
So, my last hope was that there were others out there besides Rick and I that have seen this and knew why it happens, and could provide enough info upon which I could decide a course of action. The folks who have responded have been great, don't get me wrong, I'm just surprised that there isn't anyone else besides Rick that has seen this ....

Tom

 

[heaterman]

I don't have an answer other than to say you should measure the temps at your different locations with the same thermometer to ensure an accurate and meaningful reading. There's too much variation in different types of thermometers to give you a reliable indication of what is really going on.

Now....... I spoke with Martin regarding tank turnover during firing as well as stratification during the "off" cycle. He told me their testing has shown the entire tank will turn over in a matter of minutes once even a moderate fire is established. Their testing regimen uses 8 type T thermocouples (accurate to within 1/2*) spaced from top to bottom They originally built the 3200 with an integral circ thinking that the big tank would need help equalizing temp during the burn cycle. That proved to be false. The tanks turn over far more quickly than you can pump even at 40-50GPM.

As to vertical stratification during the off cycle he said a lot depends on flow rate and temp drop the heat emitters are able to generate. They have seen units with very low flow rates (<6gpm) develop thermoclines of nearly 60* within just a couple inches. Water above the hot/cold interface could be 160* and below it only 100-120*. I have seen this phenomena in radiant floor systems with a mixing valve in place and now that I think about, on one that we did that had a section of snowmelt incorporated in the design. Water coming back to the Garn was less than 90* IIRC. Interestingly, when the owner would fire that particular unit you could see the tank "blend" as the fire became established. The temp reading would be at 120-130 at the start of the burn and within 5-10 minute it would drop 10-15* then start to come back up. This is because the colder water on the bottom was being mixed with the hot stratified at the top. Same thing happens when you start one up with cold (<70*) water in it. They seem like they sit there forever and all of a sudden the temp starts to climb.

He also said that they used to hear occasionally of some units/systems with front to back variation in models prior to when they added the return "injection" pipe but never to the extent you are describing. Maybe 4-5* at most. He suspected that the variation was coming from different types of thermometers and suggested that you use the same one everywhere in order to get a valid set of data.

 

[Rick Stanley]

I let my pumps run all day. When I got home I had 120 coming into the house and the front garn gauge was just below 140. I built a fire and just checked it after 20 minutes of strong burning (stack temps over 400*) and I show almost EXACTLY 144 on both gauges.

 

[nt3041]

Hi;

I have a Tekmar 150 installed in a well next to the Garn supplied gauge. The 150 is read in my kitchen area about 60 feet from the boiler. Both read within one degree of each other. My main circulator operates on demand,ie on a call for space heat or DHW.

Regards

 

[foxt]

Thanks for the replies.

I am pretty sure that the variation isn't due to the different temp gauges/devices - as Rick reported, I too see front/back within 1-2* of each other at some stage of a burn and for a while thereafter. The differential only develops after btu's stored in the tank have been drawn down for a while.

For grins, I measured the elevation of the front tap for the gauge, and the rear top supply. The front is maybe 5 inches higher up the tank (as near as I can tell, I'm not able to get to the actual tap at the rear without some work, so I am measuring the black iron pipe that runs horizontally off of it for a while).

Martin (and Steve, I think), reports that there can be huge temp differences at, or even within a couple of inches of, the cold/hot vertical barrier that develops over time. So can Rick and I just be seeing this thermocline in action as it gets to the top levels of the tank (front gauge is in the hot layer, rear supply is in the cold)? And might that dramatic vertical differential be due to presumably lower flow rates in our configuration vs. the flow in the p/s plumbed installs? Since the supply is slightly below the front gauge, and there's what, another foot of water above that? is it possible that I'm not effectively pulling heat from that uppermost layer of water?

If so, that would beg several questions:

1. do the p/s plumbed units also see such a strikingly thin hold/cold interface layer at some point? From the feedback we've seen here, it doesn't sound like you can observe it at the top level of the tank, but does it develp down lower? Or is it less of a dramatic discontinuity?
2. presumably that layer rises through the tank as the btu's are drawn down? does it rise faster or slower in a p/s system (assuming it develops there at all) vs. a single loop system?
3. all other things being equal, which config yields more usable heat between burns?

These may be up for guessing vs. observable, but if the answers pointed to p/s providing a longer storage cycle as compared to single loop, I'd bite the bullet and replumb. Or do I increase the flow rate of my circ to try and pull water from that top hot layer?

Tom

 

[allan]

I piped my boiler so that I pull the supply water off of the back of the boiler near the top of the tank and I piped my return water to the front of the boiler about mid way up the tank. My tank is 10.5 feet long by 7 feet in diameter. I notice something different then what you are experiancing, my gauge will read say 195 after a burn, but my supply water is actually reading something over 200 degrees. My supply water is always hotter then my gauge on the front of the boiler. My supply line is maybe 10-inches higher then my temp gauge in the tank. My unit stores about 3000 gallons of water and so far I'm lossing about 25-30 degrees in 24 hours while heating my 2500 SF house and heating my domestic hotwater for 5 people. I think that if you would take your supply from the rear of the boiler near the top and return the water to the front of the boiler closer to the bottom, you should get good mixing of your tank. I'm using a B&G Pl-36 pump that runs 24/7. When there is no load on the heat exchanger, I have about 1 degree loss between the supply and return water at the boiler which is about 85 feet from the house. When I have a load on the heat exchanger, I have about a 10-15 degree delta between the supply and return line.

 

[brad068]

You know guys, the garn heat exchange tubes are set in the tank at different levels and two passes are right close to the bottom. With that in mind that setup is the same as if you were charging a tank with an intank exchanger which most that I've seen are set close to the bottom to get maximum stratification. This is the same setup only using flue gas instead of water so why wouldn't the tank turn itself over naturally during firing?

One thing that I noticed when I switched my supply return around is that I get hot water sooner during the burn, but now that I have a bigger circ running parallel with my inducer timer the hot water isn't as "soon" as it was.

 

[heaterman]

You know guys, the garn heat exchange tubes are set in the tank at different levels and two passes are right close to the bottom. With that in mind that setup is the same as if you were charging a tank with an intank exchanger which most that I've seen are set close to the bottom to get maximum stratification. This is the same setup only using flue gas instead of water so why wouldn't the tank turn itself over naturally during firing?

One thing that I noticed when I switched my supply return around is that I get hot water sooner during the burn, but now that I have a bigger circ running parallel with my inducer timer the hot water isn't as "soon" as it was.

I was wondering how that was working out for you. So at present I assume you are supplying from the bottom tapping, returning on the top?

 

[Rick Stanley]

would a rig like this help with taking temps at different places on a heating system?
http://www.amazon.com/gp/product/images/B0017L9Q9C/ref=dp_otherviews_z_5?ie=UTF8&s=hi&img=5

 

[brad068]

You know guys, the garn heat exchange tubes are set in the tank at different levels and two passes are right close to the bottom. With that in mind that setup is the same as if you were charging a tank with an intank exchanger which most that I've seen are set close to the bottom to get maximum stratification. This is the same setup only using flue gas instead of water so why wouldn't the tank turn itself over naturally during firing?

One thing that I noticed when I switched my supply return around is that I get hot water sooner during the burn, but now that I have a bigger circ running parallel with my inducer timer the hot water isn't as "soon" as it was.

I was wondering how that was working out for you. So at present I assume you are supplying from the bottom tapping, returning on the top?

Presently, supply from top. Last year I supplied from the bottom.

 

[heaterman]

Hey Fox!!

I was at my favorite farmers place all day Thursday and Friday and I think I observed what you are talking about. This guy has two very abused 2000's which are piped together via a common manifold. They were an absolute mess and we were cleaning them inside and out. We drained, power washed and flushed one, filled it and fired it up to 200 in the morning and were doing the same with the other one while the single boiler kept everything heated in the barn. I would say that we had the first unit up to 200 by 1PM and it was sitting there with a 300,000 btu load on it for about 3 hours after that when I looked at the temp gauge. It still read 188-190* and I'm thinking...........we just took about 8-900,000 btu's of heat from this thing and it still shows almost 190*. So I got out the Fluke temp probe and started checking. I went to the barn and checked supply water temp going to the indirect tank and sure enough, it read 146.4*. Went back to the Garn Barn about 100' from the mechanical room and checked the supply on the manifold (2" steel) 146.8*. Then I took the infra red and shot the back of the Garn (we had the insulation off) and starting at the bottom took measurements about every 6-8". They were as follows from the return port up. 135, 135, 137, 138, 140, 140, 143, a couple inches below the supply port I hit 146, a couple inches above it the temp read 163 and 6" above it read 187. Sounds like what you are seeing.
I would call that pretty good stratification. I was interesting to see that the temp gouge on the Garn hardly budged during that time period with that size load on it and that's what got me thinking about your scenario. Now here's the wild thing about the stratification and it matches what Martin told me about tank blending when there's a fire in the Garn. I stoked it back up again with a good size load of wood and kept an eye on the tank thermometer. About 5 minutes into the burn the temp gauge still read 178-179 but while I didn't actually see it happen about 15 minutes later the gauge read 150. Either the return water hit the thermometer or else the tank turned over. The temp gauge stayed at 150 for about a half hour with a raging load of wood in it before it started to climb again. I would guess that it took that long for the entire tank to blend plus that fact that the beast was still carrying a good size heating load.
I don't know if that matches what you're seeing but it was interesting to see that stratification in action.

Now, you should know that we have a 1/2hp Taco 1400-70 circ in the main loop, which according to Taco's spec's and my head calculations should be cranking close to 45gpm through the Garn. Definitely not a low flow situation.

 

[foxt]

Hey heaterman,

As always, so glad you are here! I think you have now seen what I have seen, and what Rick may see. What is curious about this is that others have not seen it, or is that just because they haven't been looking for it? Or maybe they fire before the thermocline reaches the supply port? If we are right and can attribute the front/rear differential to the existence of a very thin boundary between hot and cold water that moves up the tank as btu's are drawn down long after a burn, a couple of things come to mind:

1) I assume this very sharply defined stratification is a good thing, right up until the moment that layer passes the supply port? In other words, I shouldn't keep the tank mixed during storage mode?

2) Is there any way to get to the btu's stored in the upper reaches of the tank once the thermocline gets above the supply port? Or is that warmer water just going to sit there and taunt me?

3) Since the tank mixes very well during a burn, does it matter where that tank gauge is mounted? Can I drop it below the supply port, maybe off of one of the bungs for the electrical elements? I want to be able to measure the temp in the tank so that I don't overfire, but I also want to reliably know when the tank needs to be refired, ideally when that thermocline hits the supply port?

4) my root problem is that my GARN circ is controlled by a temp sensor at the front bung next to the temp gauge - since we are pretty sure that, at least in my case and now maybe in your farmer friend's case, this location will continue to report temps above my set-point even though the temps at the supply port can be 25* lower or even more, any advice as to how to work around this?

5) and finally, if this condition is directly related to configs that are not p/s (not quite sure how your farmer's setup if plumbed), do I replumb to a p/s or not bother?

Tom

 

[heaterman]

Hey heaterman,

As always, so glad you are here! I think you have now seen what I have seen, and what Rick may see. What is curious about this is that others have not seen it, or is that just because they haven't been looking for it? Or maybe they fire before the thermocline reaches the supply port? If we are right and can attribute the front/rear differential to the existence of a very thin boundary between hot and cold water that moves up the tank as btu's are drawn down long after a burn, a couple of things come to mind:









Tom

1) I assume this very sharply defined stratification is a good thing, right up until the moment that layer passes the supply port? In other words, I shouldn't keep the tank mixed during storage mode?

I don't think a person could keep the tank blended with less than 80-100gpm of flow. Good stratification will keep the temp supplied to your system consistent until the cooler water reaches the supply port. Then it's time to fire

2) Is there any way to get to the btu's stored in the upper reaches of the tank once the thermocline gets above the supply port? Or is that warmer water just going to sit there and taunt me?

There is no way to get at that last 100-150 gallons of water that I can see. Other than welding in a new supply port at a higher location. I think that one would want a little cushion water level wise over the top of the port. If it was too close to the top of the water you might run into a situation where you would create a little whirlpool and start sucking air into the line.

3) Since the tank mixes very well during a burn, does it matter where that tank gauge is mounted? Can I drop it below the supply port, maybe off of one of the bungs for the electrical elements? I want to be able to measure the temp in the tank so that I don't overfire, but I also want to reliably know when the tank needs to be refired, ideally when that thermocline hits the supply port?

I think I'd leave that one where it is and put an additional one at a lower point if you wanted to track the thermocline. If your only gauge is at a low point you won't be able to tell how much hot water you have left.

4) my root problem is that my GARN circ is controlled by a temp sensor at the front bung next to the temp gauge - since we are pretty sure that, at least in my case and now maybe in your farmer friend's case, this location will continue to report temps above my set-point even though the temps at the supply port can be 25* lower or even more, any advice as to how to work around this?

If you could enlighten me as to why you are controlling the Garn circ in this manner maybe I can offer a suggestion. Shooting from the hip, I'd say get a strap on aquastat and install it on the supply pipe.

5) and finally, if this condition is directly related to configs that are not p/s (not quite sure how your farmer's setup if plumbed), do I replumb to a p/s or not bother?

The condition is related to the fact that the Garn stratifies very well. Piping configuration would have virtually no effect on it that I can envision.

Now, assuming that your Garn circ is in a main loop feeding a heat exchanger, does the aquastat near the thermometer location kill the pump at low temps and if so why?

 

[foxt]

If you could enlighten me as to why you are controlling the Garn circ in this manner maybe I can offer a suggestion. Shooting from the hip, I'd say get a strap on aquastat and install it on the supply pipe.

Now, assuming that your Garn circ is in a main loop feeding a heat exchanger, does the aquastat near the thermometer location kill the pump at low temps and if so why?

Not sure I understand this last one, but I'll explain what I have and maybe that will answer both or your questions ...

GARN circ supplies fphx that's plumbed in to my original manifold distribution system in series with backup fossil boiler. When the tekmar controller in the distribution system decides that my manifold has a boiler demand, I use the sensor at the front of the GARN to determine if the GARN circ should come on, or if the oil burner should fire. When I first set this all up, I ran with the GARN circ setpoint at about 125* above/at which the GARN circ is enabled (I can use water down to 120* in the house, it's all radiant floor). With a setpoint at 125*, and with the thermocline, the supply temps from the GARN would actually fall below 100*. Over the course of maybe a week, I kept boosting the setpoint until I discovered that the lowest I could go was 150* to maintain supply temps above 125*.

Since it does indeed work this way, and since there doesn't appear to be a way to get the remaining btus out of the tank, maybe I should just leave it alone. On the other hand, I would like to control the system a little bit more directly (i.e. base the control on actual supply temps from the rear vs. this empirically derived value from the front). I thought about just strapping the sensor to the supply pipe, but when the circ has not been running for a while, won't that give me a false low reading - or does all of that steel/iron conduct heat well enough that if I can attach the sensor to within an inch of the back of the tank on the supply pipe I'm good to go?

If your only gauge is at a low point you won’t be able to tell how much hot water you have left.

Actually, I guess the only temp that will really matter is the temp at the supply port, right? When that's below my setpoint, I fire, and when it's above 190* I don't. Makes the front gauge seem kind of useless, no? I wonder why that bung is not at the exact same elevation as the supply port?

The condition is related to the fact that the Garn stratifies very well. Piping configuration would have virtually no effect on it that I can envision.

This is the part that still confuses me - so there's a pretty good chance that other GARNs are doing this, but folks haven't noticed?

Tom

 

[jebatty]

Jumping on to monitor. Working with an organization with the 3200, and actual supply temps are a real issue, so this is interesting. Also, fphx installed to handle 950,000 btuh with a 30F temperature difference. This concerns me, thinking that it should have been sized at a 20F temperature difference. F/B temperature in the Garn may impact this for getting usable heat into the system.

 

[Gooserider]

Speculating a bit, but the reason they might put the front gage lower than the supply port could be to give you some "inertia" by getting you to fire while the tank is still hot enough to be supplying acceptably hot water at the supply port... The sloping thermocline seems to screw that up, but it seems like a reasonable idea in theory.

As to temperature monitoring in general, my understanding is that a strap-on aquastat or other sensor will pretty much track the temperature of the water going through the pipe it's strapped to, as long as it's reasonably well insulated...

If I was wanting to monitor the temperature of the tank itself, I would try to get a sensor attached to the wall of the tank itself... If I was feeling really ambitious, what I'd think would really be neat would be a few strings of DOW (Dallas One Wire) temp sensors attached to the tank in different places with thermal epoxy. Depending on how many strings one ran, and how many sensors were used, it would be possible to do a very precise map of the tank showing just where the thermocline was, and the various temperature distributions in the tank...

Gooserider

 

[foxt]

Speculating a bit, but the reason they might put the front gage lower than the supply port ....

Just a note - front gauge is higher than the supply port. If you use the front gauge to determine to fire, you may actually be firing too late if the thermocline is at the lower supply port ...

Tom

 

[Gooserider]

Speculating a bit, but the reason they might put the front gage lower than the supply port ....

Just a note - front gauge is higher than the supply port. If you use the front gauge to determine to fire, you may actually be firing too late if the thermocline is at the lower supply port ...

Tom

OOPS!! :red: Me bad....

Gooserider

 

[Jim K in PA]

Interesting. I would think that with Steve's customer's high-flow setup that there would be enough mixing to disturb that pocket of warm water up at the front wall temp gauge. Apparently not! That front bung is about as far away from the return bung as you can get.

Jim - your project with the 3200 - keep in mind that like any storage system, you are NOT going to have a constant supply temp like you would have in a constant firing appliance like an oil or gas furnace. You can set up a mixing valve to reduce output temp to something below the max temp, but that only works until your tank drops below that point.

I think the lesson here is that no matter what piping arrangement you have, circulator controls should be based on temps in the PIPING, not on what the temps are at the front gauge or in a well attached to the tank. I have a temp well installed in the second bung on my WHS2000, but I only use that for remote temperature display. I have Tees with temp wells installed in the primary piping for (as yet unused) control circuits.