Before I buy the part: You guys are sure that the Tekmar 361 mix control with sensors, will:
--give full outdoor reset on our zone valve system;
--give the conventional cast iron boiler protection by barring returning water below 130 degree condensing temperature;
--when piped together with a pumped boiler loop through a buffer tank, and a pumped system loop also through that buffer tank?
In this way, a two pump system uses a buffer tank to achieve HYDRAULIC SEPARATION as a result, rather than using the 3 pump system so often seen in primary/secondary schematics.
And the tank [ for me here a re-piped discarded well pump tank] can give an oversized boiler some longer firing runs.
You see, I have no experience with these controls.
Its like buying magic beans. Trust the bean man.

 

How big is the tank?

I'm not totally sold on the 361.

The whole idea of the buffer tank is to store and hold BTUs so why not just use a simple aquastat with a wide differential and then something like a Taco i-series mixing valve to supply the heating system off the buffer tank. The beauty of this solution is that if you want to run more than one temperature, just get another iseries valve and each can have their own ODR curve.

 

What is the boiler and overall system again? Agree with Who. There are plenty of options to explore. The 361 is a good one, but there may be simpler and equally, if not more, effective depending on the specifics of your system, heat emitters, heat loss, etc. etc.

 

Thank you guys for responding. Ideas are so needed. In response, the System is:
Buderus G124X-32 atmospheric cast iron boiler,
with 6'' aluminum chimney liner up 25',
contains about 3 1/2 gallons,
weighs about 340 pounds,
installed 6 years ago.
There are five Honeywell zone valves ,
A single Taco 007 does the job well.
Air removal with B&G IAS and filtrol works well.
Radiators are mostly smaller thin tube cast iron [20" wide, 25" tall, 3 tubes deep...that I salvaged and used to replace the 1940 in-the-wall fin/tube stuff. This added heat output I think. The older basic part of the house is monoflo.
The add-on family room and a bedroom above it have cast iron baseboard.
Probable system volume is 40 gallons.
Soon to add 6th zone with bargain AO Smith heat exchanger and glycol for a few rads in garage shop for occasional use.

The 1940 colonial two story has good attic insulation, R 38 or so. Walls R8. House is 2,450 square feet above ground, and 800 in basement.
House is sided in Alcoa aluminum, is fairly tight per door test, is warm, comfortable, but with slightly noticable temp swings .

The Problem, you may recall, is that I was led by old charts and notions into oversizing the boiler.
I Designed for -20*. It recently fired only 57% of the time at -17*. [ We did not get to -20* with high winds this year, for a perfect test]..

Nor does the boiler get very hot. When I finally put on a dual thermometer, I saw that at -10*, the boiler supply only got up to about to 163* or so with Delta T of 15 to 20*.

The length of the fire runs at -10* were reasonably OK: 8 minutes on, 6 off, 8 on, 4 off, 11 on, 14 off, 11 on, 5 off, 1.5 on, 1.5 off, 9 on, 5 off, 11 on ...for 108 minutes. 55% fire on.
At 5*, but sunny, I got 4's and 5's on, firing 41% of the time.
At 18* I got 3.5, 4, and 5 on. Firing 27% of the time.
Also now revealed is the low return temps at higher outside temps. On a 38* day it was 110* in and 128* out. That's too cool for an outside wall chimney, maybe.

The perceived need: I have the 36 gallon tank,
I can pipe stuff,
there is space for it.
I would like LONGER RUNS .
OUTDOOR RESET on the system side would be very good, for economy. The boiler as such does not need reset. It does not get that hot nor go off on high limit.
The boiler needs PROTECTION from low return temps.

It takes up to 1.5 to 2 minutes to get a smooth bed of flame, and I hate to hear that Effikal damper signalling yet another startup.
I disconnected the end switch on the zone valve on that added on 2nd floor iron baseboard zone. It is hyper insulated and does well without needing its own firing cycle. This helped.

So far I have cozied up to the KISS principle very much. I even have a pilot light.
I don't really want a delicate control that costs big bucks.
I'm was just getting pulled into it by all the optimistic talk . And the magic bean mystery of blue boxes. But I will bite the bullet if its optimal.
Maybe a less expensive ODR control on only the tank [???].
I sure do like the idea of a true temp sensing protection on the boiler.
As others have said, my head hurts.

 

While the buffer tank will give you hydraulic separation, what it will NOT give you is the ability to do FULL system reset. Obviously you can not do full boiler reset... so you need a way to run the two loops at different temperatures. With the buffer tank in between they will both still be at the same temperature.

You need a MIXING point, and that's where that third pump, or mixing valve would enter the picture.

 

Gotta think about control and piping options, but meanwhile play with this:

Vbt = t * (Qh - qload) / (500 * dT)

where

Vbt = size of buffer (gal) [36 gal]
t = desired on-cycle (min)
Qh = energy source output (BTU/hr) [for G124X-32 = 110,000]
qload = rate of heat extraction from buffer (BTU/hr)
dT = temp rise of tank between on/off cycles (F)

If you want to see what your 36 gallon buffer will give you for on-time at various heat loads (cold vs. warm outdoor temperatures), rearrange the terms so it's like this:

t = (Vbt*500*dt) / (Qh - qload)

 

I see what you mean. You woke me up.
Just because the buffer tank sits there acting as a fat happy junction between a boiler loop and a system loop, and turbulent mixing will naturally occur there, it doesn't mean one has a measured controlled mix. It's just heated boiler water and return water meeting in a fat spot. Hydraulic separation, as such, just means an available even distribution pressure I guess.
.

I re-read "Rediscovering Hydraulic Separation", a neat two page piece by John Siegenthaler. Figure #10 had beguiled me. I should have focused on the mixed zone in Figure#5. Please check that out. I confused the basic topic there with reset.

However, there, in #5, if you carve away the unmixed zone you get, in sequence, a boiler, boiler loop, a buffer tank , and a system loop and injection pump, doing the task needed..supplying reset water to the emitters. Is that not a picture of what I need?
And, surely, that Taco i-series valve is a nice candidate. But a number of writer seem to prefer VS pumps over mix valves.

If that picture is correct, so far here,
there are two fun questions.
1. The ideal controller[s] to reset the System ,and to protect the boiler?

2. And this question, what if the mixing were done upstream of the buffer tank? Then both the buffer's content and system would contain reset water? Hmm, not so good; there would not be the desired long firings to deal with an over-sized boiler.
But there would be fewer BTU,s stranded on occasion. Well, I'll insulate that tank well.

The aquastat idea is also interesting. I've got to think.

I think this theme is recurring these days.
Good night.
joew

 

Joe, one thing to keep in mind is that most of the boilers in that article are 'mod/cons', and don't need any boiler protection, and indeed WANT the coolest return water they can get.

You are working with a conventional boiler.

Have you looked at Tekmars website? There's one essay in particular that may speed your learning curve:

Mixing Methods and Sizing of Variable Speed Injection Pumps

The title is a bit off the mark, they do discuss mixing valves as well.

In any of those diagrams, you can replace the CSTs on the boiler loop with a hydraulic separator. (have you looked at "Boiler Buddy"?)

 

NJ Trooper,
OK I did the homework you all recommended.
The 3 pumps, if used , would be this:
The boiler loop pumps into and out of one side of the buffer tank. NO CST's there. Simple. As seen at "Boiler Buddy".

Then on the other side of the tank I have to get Closely Spaced Tees off the tank. Two closely placed holes [nipples] in tank side? Or,
I could run an "outrigger", that is , a vertical pipe running from the top port to the bottom port, and then take off with CST's from the outrigger, horizontally. Any thoughts?
Those CST's begin and become my mix loop with its variable pump.
Then off the mix loop are a second set of CST's to the system loop with its own pump.
Is that it?

This is fun, and difficult.
Joew

 

JoeW here, with an update. 3/10/09

Dan ,at Hot Water Products [Boiler Buddy] , kindly responded, almost immediately, to my query along the lines of my just previous post here.
He stated the outrigger would work but is not needed, stating that since the Boiler Buddy buffer tank is itself a hydraulic separator, one can just use the tanks OWN ports [ albeit several feet apart] for the take-offs to the mixing loop. Then pipe westward to the system loop and its own pump.
The buffer tanks other two ports of course pipe eastward to the boiler.
[Or whichever points of the compass you have].

Dan added that it would be good at least in the case of a cast iron non-condensing boiler , as here, to use a differential T-stat in the buffer tank's well , wired in line with the signals from the room thermostats, so that the boiler would not fire until the radiator system had drawn enough BTU's off the buffer tank to take the tank down to ,say, 145*. [A point that keeps boiler above condensing]. This would use the buffer tanks ability to make for efficient long firing runs.

Then, he added, you would use the appropriate Tekmar, Honeywell, etc. to control the mixing and the boiler protection.

Dan will post me his sketch, and may consider tweaking their literature for mixing illustration.
Fun stuff, again. joew

 

Dan is gonna post it here? that's cool...

I don't know if this link is gonna display, but let's try:



I have to think about the differential t'stat Dan was talking about because I don't think it's even necessary. If you wired this up with a Tek 361 for example, you would want to use a FIXED diff of something around 30* (per Tekmar).

 

I'll be expecting a residual check promptly!. Funny, I've been following this for a bit and was wondering when the referencing was going to start.

 

I'm the one getting the resids... I drew it! You just built it...

 

Allright if you insist I'll split it with ya.

 

NJ, why the bypass? The MPO already has one?

Joe, what about a good sized reverse indirect on a fixed wide differential aquastat? Put a good mixing valve on for DHW safety and then you can add as many iSeries 3 way mixing valves as you have separate temperate zones/emitters.

The boiler feeds the indirect which is also the buffer. The ODR for the CH (central heating) is done off the indirect/buffer water tank.

One circ between the boiler and tank. One circ for the / per iSeries zone.

 

Mpo doesnt have a bypass, it has a short tube added on to the return fitting that "injects" the return water "deeper" into the chamber. The i/o manual drawing does suggest a bypass on the near boiler piping.

 

Right... and as it turns out, it's not really needed in this application anyway. During heavier load times, the buffer temp itself doesn't really drop into condensing range. Even if it did, it doesn't stay there very long once the boiler does fire. At least according to data we've gotten from John so far. It will be interesting now that the weather is warming to see if that holds true.

Thing is, when piping it up, for the price of a valve and a couple nips and tees, it's cheap insurance. Don't need it? Close the valve.

 

That seems to be a good outline of the idea.
Nice picture. I still use crayons.
I think Dan of Boiler Buddy would reply to my own e mail.
I will bring it here.

Useful note: Please google "hydraulic separation" together with "Gerry St. Laurent" to see his suggestion there that the cooler system return should go to the top port of the buffer tank, and the hot system supply go out from the bottom, all to cause the tank to mix, which he sees as good in this application. John Siegenthaler posts a reply in which he sees it as useful, I think.
Gerry also, I now see, set forth the idea of controlling the boiler with a differential aquastat in the well of the buffer tank
joew

 

Joe, that drawing that Troop posted is pretty much the system I did at my house. Only difference is that I havent done the DHW yet and my boiler circ is on the return.
One thing I would change on it if I was so inclined would be what you had mentioned. I would have the injection pump(ch supply side cst) come off of the bottom. Not so much for mixing reasons but more to eliminate ghost flow from boiler supply. On the Boilerbuddy 30, the supply side and load side ports are directly across from each other.
I'd be happy to answer any questions you have, as well as I can with my limited knowledge.

 

What is the need for the P/S piping after the buffer tank? Adds one more pump and complexity to the system.

If you have boiler protection, I'd go with a single pump feeding the zone valves. The P/S piping doesn't provide boiler protection, so why bother with the extra stage????

If you use a 361 or TV to protect the boiler (before the buffer) I believe you could achieve near full ODR with a properly sized buffer and boiler.

Pete

 

I'm pretty sure with a properly sized boiler, you probably wouldnt need a buffer tank. The only reason I did the buffer was to combat a boiler that was twice the size of what I needed.
I would also add that my system is a bit more complex than the average bear, with the vsi, sys. circ, zone circs, and when all is said and done a dhw circ. At the time of my build it seemed to be the best way to satisfy the small heating needs of my home. I personally dont mind the slight complexity of it, but others might be put off by it.

 

Who, thanks for suggestion.
1. Regarding indirects, after lurking on The Wall for years I decided upon the experiences related there that indirects don't last all that long. I can get 3 Rheem water heaters for the cost of one indirect. I compared longevity [I get get service out of a water heater of 10 to 15 years] . Also, when a snow laden tree branch kills the electricity, Herself still has a hot shower from an independent water heater. This IS a big deal. So, the water heater is in , it's new, it's done .
Not that a 30 gallon tank-in-tank doesn't remain a VERY tempting choice as a buffer [piped inversely] and water heater. I'd still have the other in readiness.
But then the COST! I have this well pump tank. My friend can add ports. But the commercial ones are still tempting, for sure. Real nice products out there.

2. Say, on the schematic posted above , shouldn't the bypass be downstream of the boiler pump, for a true System bypass?

3. This thread relates in part to the control aspect raised in that of John's project over at Boilerbuff.com. Both sites being new to me.
But here, this is all mid-level mass cast iron.[ to bring that to mind].
But here is the big question now:
Q: With a Tekmar 361, my would-be candidate for control, does the boiler sensor [the protector ] go on
(a) the outlet out of the buffer tank to the mix loop, or on
(b) the outlet out of the boiler to the buffer tank.
The boiler outlet is the usual place. But here, with a buffer being used, to extend the firing runs of an oversized boiler, and the buffer being sort of a load and sort of of a yet unmixed extension of the boiler.?
What's your view?
And does the control give a choice in this aspect here anyway.
Confession here, I have posted this question to Tekmar, who have been unfailingly courteous . I'll share the response.

I do want to protect that boiler. I saw return temps of 100* for close to ten minutes a few days ago from a cold zone of 36 feet of cast iron baseboard.
Good night.
joe w

 

You could add a bypass after the pump, so some of the heated supply water is diverted back to the return on the boiler. A throttle valve would allow you to set the bypass for a reasonable return temperature as the boiler heats up. But, mutliple zones makes it pretty hard to optimize the bypass flow since the optimal bypass throttle settling will change as the zones open and close. I like the Danfoss TV valve, simple and failproof.

Pete

 

Joe, on my set up, I have the boiler sensor on the boiler supply, between the boiler and the buffer tank. I have the mix sensor on the zone supply header. With the 361 the boiler sensor can be placed on the return also, or not be used at all. Different applications for different set ups. With mine, we all agreed it should be on the supply. You can check out almost all there is to know about it if you go to Tekmarcontrols.com and search the downloadable files. Tons of info there.

 

Pete,I have to check back somewhere. Somewhere Siggy faults the Esbe for its inability to sense something, or deal with cold water ,or of some darn thing. [That first sounds like a comment I get from Herself occasionally.]
Because on the face of it it sounds just right.

The Patient Canadian at Tekmar advises today that on the schematic in this thread that we have so far that the boiler protection sensor would go on the outflow from the buffer. He wrote,
" Hello again Joe, see attached A 361. It shows sensor locations. Sensor must be placed on the mix system pipe, and boiler sensor on either boiler supply/outlet (boiler Load reset), or return after the mixing (boiler enable), in your case after (downstream) of the buffer tank.
Yes and the Blasts of arctic air will stop soon, it is almost the Equinox! " [Tekmar quote ends here].

So, in effect the buffer is controlled as , in effect, part of the belly of the boiler.
This thought struck me last night at 2:00AM before the response and it bothered me. Perhaps after a long warm period, or on a start up, it would take quite a while to heat that 36 gallons in the buffer to above condensing.
Must relocate the math for 108K BTUH moving 68* water to 130*.
Maybe we have it all wrong. John's mod/con is a different scenario. Our iron would not like this. Maybe the buffer tank has to be a secondary circuit with CST's separating it from the boiler. That way the boiler gets its own little pumped loop and gets the return sensor.
Siggy has it somewhat that way in an article: Preventing Short Cycling.
[IMG]http://www.radiantandhydronics.com/Plumbing
/Home/Images/1107-portal-GF-Fix-lg-edit.jpg[/IMG]

joe

 

Here is easier way to image that buffer set up in the just prior post.
http://www.radiantandhydronics.com/P...ix-lg-edit.jpg


joe

 

Joe, my boiler is not a mod/con. it is a burnham MPO84 CI boiler with a beckett burner.

 

The first premise of this thread is that the installer, the author here, OK, me, put in an oversized boiler.
The gas consumption, the envelope's R factors, the design temp, all pointed to about 96,000 BTUH. Well, the net 109K Buderus was the nearest next up size. And it fires about 60% of the time near design. It does not get very hot either. Manual J,as they say, exaggerates the need.
So, the desire for the buffer for longer more efficient runs.
And, while I'm into it, why not full ODR on the system side? That requires the downstream CST's.

I just today posted [and cannot find it] that Tekmar's patient man says to put the sensor for boiler protection on the buffer output portal to the system, if using the diagram we have on this thread.
Am rethinking it all. Maybe bes to put buffer on a a secondary loop. Easier to protect boiler. Maybe easier to ease it into operation from a cold start. Lotta cold water in a 36 gallon tank installed onto a 40 gallon system. Siggy does P/S here in a number of diagrams.
joe

 

hmmmm... I don't think I agree that the sensor should be downstream of the buffer. I don't think you want the sensor on the MIX BRIDGE... there are going to be times when that variable pump is not running, or running slowly... and that sensor won't accurately sense the temp from the boiler. No... put it on the supply pipe out of the boiler.

 

I guess it depends on what you want the bypass to do. As shown, it's a BOILER BYPASS. This would afford some protection to the boiler by decreasing the flow through the boiler and increasing the delta T across the boiler by raising the supply temp, in turn increasing the average temp of the boiler.

A SYSTEM BYPASS works by raising the temp of the return, DECREASING the delta T across the boiler, but also raising the average temp of the boiler. In most cases, you do NOT want to divert flow from the system as this can cause uneven heating due to reduced flow in the system. In THIS case though, it might actually be a better choice... because since the system is P/S piped, the system bypass will NOT divert flow from the heat emitters... it WILL divert flow from the buffer tank though, but this might not be a bad thing. It bears some consideration.

 

The amiable Texmar guru advised this:

"Joe, see attached A 361. It shows sensor locations. Sensor must be placed on the mix system pipe, and boiler sensor on either boiler supply/outlet (boiler Load reset), or return after the mixing (boiler enable), in your case after (downstream) of the buffer tank.

Yes and the Blasts of arctic air will stop soon, it is almost the Equinox!"

joew

 

Sorry, not buying it. You can look at that PDF file all you want and you will not find one instance of the BOILER sensor being on the MIXING BRIDGE, which is in effect where it would be if it were downstream of the buffer tank. The MIX sensor of course has to be on the system loop.

If in your mind, if you replace that buffer tank with a pair of CSTs and think of it as a standard P/S mixing arrangement, you would NEVER place the sensor on the middle loop...

He's wrong, sorry.

 

Obviously the mix sensor has to go where it has to go. I would be interested as to why he suggests what he does for the boiler sensor.
On my system, when the boiler first fires, the max differential between my supply and return is at best 15-20*. This number quickly decrease as the boiler/buffer water heats up. Due to constant boiler water circulation, because of a pretty constant heat call, the diff. is usual 0* once the boiler shuts down. What I'm getting at is that there isn't that much of a difference no matter where I would put the boiler sensor.
But, putting it after the buffer isn't giving you true boiler temp. Its giving you tempered boiler temp.

 

Boiler sensor on the return and 10% boiler enable might (I'm guessing almost definitely) yield a longer runtime. Assuming that the injection loop and circulator are sized properly.

Using no boiler sensor at all might also (almost definitely) yield a longer runtime.

Both of those options basically let the boiler bounce off its aquastat and the control is doing only two jobs:

1) closing a contact (Boiler) that allows the boiler to fire; and

2) serving as the mix temp controller.

With the boiler enable option, the boiler will be allowed to fire whenever the injection pump output is >10%. Because the goal is to keep the injection output between 5-10%, that ensures the buffer will be fairly hot to keep the injection rates low.

With the no boiler sensor option, the boiler will just sit there and fire on its aquastat as long as there is a mix demand. The injection pump will modulate speed based on whatever the buffer tank water temperature is.

For the no sensor option, using an adjustable differential aquastat would prolong the runtimes (e.g., get one that will do a 30-50F differential, say from 110-160).

 

I think, after looking back, that maybe someone here accidentally got off on the wrong tack.
Amiable Canadian Tekmar Guru is corrrect.
Having the boiler sensor "downstream" of the buffer in the return flow is to have it at the boiler. That would be right, no?
The problem lay in his compound sentence that tried to say to much in one sentence.

 

You don't want the sensor on the return in an application where the 361 would be the only control. You want the sensor on the SUPPLY side, UPSTREAM of the buffer tank.

Placing the sensor on the return in this type of system is not appropriate.

With it on the return, the 361 has no control over boiler temperature, that task is relegated to the boiler aquastat.

Why use an additional aquastat with adjustable diff? With the boiler sensor on the supply, you already have that built into the 361 control.

 

The 361 has a adjustable diff of 2-42 degrees.

 

With a buffer tank and variable speed injection, I would try no boiler sensor at all, and use a wide-diff aquastat on the boiler. Assuming the injection rates and distribution supply temps were suitable.

 

Xiphias, did you mean to say that you would try a wide-diff aquastat on the _buffer tank_, controlling the boiler?

What would be the advantage of doing this over using the boiler sensor and the wide-range a'stat that's built in to the 361 ?

Using the one in the 361 also would give the advantage of NOT running to the high limit of an add-on fixed a'stat every boiler firing.

Or, even placing the boiler sensor in the well on the buffer tank instead of on the boiler supply pipe?

 

I think it would be cool to try no boiler sensor, period. The boiler aquastat would be in/on the boiler. Use the stock one, or preferably swap it out for something that can do 30F or more.

The idea would be to let the boiler just do its thing, working against the buffer. Say it would cycle between 120-170F. That would essentially be what's in the buffer tank.

Meanwhile, the VS pump handles the space heating load. It modulates speed based on ODR, indoor sensor, and the temperature in the buffer.

The boiler basically just sits there and is decoupled from the space heating function. It's allowed to fire so long as there is a call for heat, but its cycling is not determined by anything but the aquastat.

The result, with a properly sized buffer, would be a fairly predictable and constant on-time for a given heating load (which would determine how fast the buffer gets drawn down).