I am finally finding the need to replace an old Bell & Gossett size H-1 pump originally installed in my home on it's original construction date of 1951. Yes, the date code on the unit confirms this: 11-L corresponds to Nov. 1950. This is a 1/6 H.P. motor, 3.1 amps full load. The pump plate shows H-1, B-4, 11-L (last one stamped), and on the casting, P5-41. My problem stems from the fact that I am having trouble determining a modern equivalent, with only partial success searching the B&G knowledge-base. On one posting from someone starrting with a nearly identical unit, they recommend using one of their Series 100 pumps, elsewhere I find reference to using the Model HV (similar, but considerably more powerful in terms of overall pump curve). My home is small (about 1100 sq. ft.), located in the SF Bay Area (mostly mild winters), and has copper hydronic-heat tubing embedded in the concrete slab. I can see two 3/4 inch lines on the supply side, and five 1/2 inch lines on the return, but I have no way of determining loop lengths, or overall pump head. The boiler is an American Standard, about 75,000 BTU. From this I would imagine I need about 8-10 GPM flow.
Anyone care to comment on how to choose the pump size?

Thanks in advance.

 

I'd say check out the Grundfos 15-58 ... it has three adjustable speeds, so you can adjust across a wide range.

It is possible to measure pump head by taking pressure readings right at the suction and discharge of the pump... but you would have to install some way to read the pressure first... only to rip it out and install the new pump.

The NEW (redesigned) TACO 007 would be my second choice. Old stock would probably work fine as well, but the newer pumps have a much flatter pump curve and flow a bit more.

You may have to do some minor re-plumbing (flanges, etc) to fit either.

 

You want about 2-4 feet per second flow rate. In 3/4" pipe, that's about 3.2-6.5 gpm. In 1/2" pipe, it's about half that, or 1.6 to 3.2 gpm. Higher velocities can cause vibration, noise and pipe erosion. If you really have no way to figure up a resistance curve (and it sounds like that's the case), then probably a 3-speed like the Grundfos 15-58 as Trooper suggests would be the way to go. You probably only need speed 1 or maybe 2 for that system.

You could also stop by your local plumbing supply house and see what they suggest based on the old pump.

 

Thank you for the input; I was originally really leaning toward one of the B&G 3-piece oil lubricated pumps like my original one (I must say I was impressed with a 55+ year life span), but I will consider one of the other Taco or Grundfos pumps you mentioned. Any experience with typical lifespan on these type of units, or are they too new to really get a sense? Am I likely looking at a 5-10 year lifespan? I like the cartridge design of the Taco, and the three speed aspect of the Grundfos. I do hope that they are truly quiet, as the pump is just outside my kitchen in the garage.

Also, I am quite surprised that only one of these small, relatively inexpensive, low power consumption units will have the capacity to replace my big old 1/6 HP pump. Really? I thought I was looking at $200-$500 for an equivalent pump to my old one, but it would appear to be under $100 for either of these recommended units. Yes, I am aware of the need to do some re-plumbing work, as most of the modern units seem to use 6 3/8" flange to flange spacing, versus my old unit's 8 1/2" spread.

Sadly, I am not getting much support from asking around at local plumbing supply houses (not many hydronic heating systems in the area, I suppose).

Any further thoughts on how to estimate total pump head? I want to be double extra sure.

 

Crude guesstimate. Say you have 1100 sf on a 50'x22' slab. Say the tubing in the slab is 1/2". Say the tube spacing is 12 inches. Say it's all one zone. Call it about 1200 feet of tubing including the arc of the turns. Add a few more "equivalent feet" for fittings, near-boiler piping, etc., so say the total equivalent length of the whole thing is 1300 feet.

At 1.5 gpm, the head is around 4.1 ft. At 2 gpm it's 6.9 ft, at 2.5 gpm it's 10.3 ft.

If that's in the ballpark, you could do this with a Taco 005, 006, or 007. Or the Grundfos 15-58 on low (speed 1), which would automatically hedge your bets (you could turn up to speed 2 if the slab isn't performing adequately).

Yes it is amazing what these little pumps can do. And they are quiet. I doubt you will get 50 years out of a new pump, but I wouldn't rule out 20.

It would be helpful if someone else here could check these numbers, but I think these sound reasonable. Some pictures of the near-boiler piping might also shed insight into what you've got for total tubing, fittings, etc.

 

I'll chime in on the 15-58 because you really don't know what you have for obstructions in the piping.

I'm not so sure that I would be worried about the lifespan, 50 years is a very long time and the difference of pumping efficiencies compared to your 56 year old buddy at 3.1 amps or even a new Series 100 at 1.75 amps compared to a 15-58 on speed 2 at 0.7 amps means the pump will pay for itself in electrical savings quickly. My first 15-58 replaced a Series 100. At 10¢/kWh it paid itself off in less than 2 heating seasons. Running constant circ for a 200 day heating season it saves $58/annum. That said, a 15-58 is still not a very efficient pump compared to some of the newer emerging electrical motor technologies so when it does get replaced, you may have something like a 0.2 amp pump, but that won't be needed any time soon.

Use isolating flanges to allow a quick pump changeout next time. If you want to really know what the head curve is on your system, think about installing a pressure gauge at either end of the pump or bypass piping around the pump with a gauge in the middle and a ball valve on either side. Then you can get a pressure differential by the valve at one end being open and then switch it around - no need to worry this way if the pressure gauges are calibrated to be totally identical. Running the pump at all three speeds, note the differential, and by using the pump curves from the 15-58, you'll then be able to plot a curve for your system resistence.

BTW, the 15-58s are quiet! On speed 1, dead silent - even up very close. On speed 2, you have to be very close just to hear it. The sound that it does make is a very onubtrusive white noise tone. On speed 3 it sounds far busier than speed 2, but still it is silent once you're a couple of steps away.

 

The 007 on my system is over 25 years old, and still hummin'...

I've seen others as old and older still in service.

Another way to determine proper flow...

Temperature measurements. You want to shoot for something around a 20* differential between supply and return. If the flow is too lazy, you will have a higher delta, vice versa for too much flow.

Another rule of thumb that's handy... figure on 1 GPM / 10 KBTU . Your estimate of 8-10 GPM is probably close, if not a bit on the high side, based on your 75 K boiler.

 

Will disagree with Trooper on two points.

For radiant, I believe the "standard" delta T is around 10F. If you measure the delta T, I would get the system going and give it a couple days to charge. Big concrete slab might take that long to stabilize.

There ain't no way you're gonna get 8-10 gpm through 1300 feet 1/2" pipe. At 8 gpm, the head is 733 ft. Even 100 ft of 1/2" tube has a resistance of 60 ft of head if you try to push 8 gpm through it.

 

Every part for that circulator is still available. Are you sure that "B-4" is not "P-4"? I can't find anything about a P5-41, but I did find a listing for P5-4.
Both of these would use a 1/6 HP, 115 volt, 1725 rpm motor. The B&G number for the bearing assembly is 189100 (old # P7Z-4351). The Sid Harvey # is B70 (new) or B70-R (remanufactured).

 

>>There ain't no way you're gonna get 8-10 gpm through 1300 feet 1/2" pipe.


His description made it sound like there were 5 loops though... two 3/4" supplies and 5 returns... I'm certain that whole slab isn't one loop. If he can't push 8 GPM (total) through the boiler, how's the heat gonna move ? Need the water to move the heat... and less delta means MORE flow ...

 

Good points. Split, a ~250 ft loop of 1/2" has 13 ft of head at 2 gpm. Not unreasonable. So say five loops at 1.5-2 gpm each. There's your 8-10 gpm through the boiler, right? I don't quite get the two 3/4" supply and five 1/2" return. Manifold in the slab? Where?

I agree about the delta T. This system may not have been designed for 10F. Could be rather more depending on the loop layout.

I'm guessing there's larger piping near the boiler, too. 1" perhaps?

Wouldn't it be great to know a bit more about that old pump? Great that parts are still available.

 

Again, many thanks for the additional input. I am close to buying the Grundfos 15-58, but am still trying to muddle my way through my own estimate of pump head calculation.

I agree with the comment that there is no way that 8 GPM is going through a single 1300 ft. loop of 1/2" tubing, but I will restate the situation again here.

Upon further study (and some more careful measurements), I see that the initial outlet (supply side) from the boiler is 1 1/4" copper pipe, that soon splits into TWO 1" copper pipes dropping into the slab, and the return is as I said before, FIVE 1/2" copper return pipes. I found an excellent tech document on the Grundfos web site (search for L-TG-PG-001.pdf), that has some very helpful flow vs. pump head calculations for PEX and copper pipe (see pages 39 and 40 for copper), as well as some excellent overall pump theory information. If I guessed that I needed about 7-8 GPM OVERALL flow rate (based on 1 GPM per 10,000 BTU of the boiler), this would mean about 4 GPM per 1" supply line, and about 1.5 GPM per 1/2" return line, right? I think I can presume that each 1/2" loop probably does not exceed 200-250 ft. (300 ft. at most, but not likely). What I still don't follow (and would appreciate any commentary on), is how to add it all up--is this a simple sum of each loop?

I will try to post again with links to photos, if anyone still thinks it will be helpful.

 

Yes, please do post links to photos. We love pictures & can often learn more about a system by looking at a few shots than by reading page after page of text.

 

Ask and yee shall receive.
OK, for the big photoshow, go to:

http://i203.photobucket.com/albums/aa56/jsbeddow/IMG_4238.jpg
http://i203.photobucket.com/albums/aa56/jsbeddow/IMG_4237.jpg
http://i203.photobucket.com/albums/aa56/jsbeddow/IMG_4236.jpg
http://i203.photobucket.com/albums/aa56/jsbeddow/IMG_4235.jpg
http://i203.photobucket.com/albums/aa56/jsbeddow/IMG_4234.jpg
http://i203.photobucket.com/albums/aa56/jsbeddow/IMG_4233.jpg

Please try to ignore the clutter and general disrepair of this system--you will notice that the pump body is just hanging there, not bolted-in for example. As well as few other dubious plumbing and electrical no-nos. Remember that I am taking this spring/summer opportunity to repair/upgrade this system. I will most likely get rid of the old style expansion tank, in favor of a bladder type. This old style expansion tank should not have been used with the automatic air bleed currently installed, if I understand correctly, as it will lead to a water logged tank (not that I ever had a problem in the 15 years I have lived here). I will probably also install a real air separator as well, but am willing to accept advice on that point as well.
Thanks again.

 

Oh, the clutter!

If this system has worked fine for 15 years, then my first inclination would be to get the necessary parts for the existing pump and call it a day. Maybe put a new 15-58 in there, but that's as far as I would go. If there are other significant issues you want to address and this is part of a long-term plan because you're staying in the house for years to come, then I would start at square 1: a heat loss for the structure. The Bay area has about 2500-2900 heating degree days per year (for reference, that's less than half of what it is here in southern New England). Unless your house has almost zero insulation, I can't imagine why you'd need a 75k BTU boiler. Of course a 50 yr old boiler might be running at maybe 60-70% efficiency, so it might only be yielding 45k-53k of usable output.

A long-term plan might include doing the heat loss calculation, adding some insulation and doing some other thermal improvements, and seeing where that puts you for a new boiler down the road. If you have gas, then a small modulating/condensing boiler would probably be just the ticket. Considering the improved efficiency, the payback time on a new boiler might not be longer than you'd find acceptable.

Question: what water temperature does this system run at? Concrete slabs generally use pretty low supply temps, in which case I'd be concerned about condensation in the boiler. But if it's 50 yr old, then I'm guessing it's built like a tank and has been dealing with any condensation, or it's running at a pretty high temperature.

 

In your application, I think I would keep the conventional tank & do away with the auto vent. In place of the auto vent, you could put a small baseboard type manual bleeder if you so desire. The conventional tank, if piped with a steady rise from the boiler to the tank & no horizontal piping, will act as an air separator. Other than that, I would agree with Xiphias, particularly about a more modern boiler.

 

That's cool... that system is VERY similar to the one in the house I grew up in. I just knew there'd be balancing valves (img_4237) !

I'm down with the others, just fix what needs to be fixed, and save up some coin for a new system.

Just out of curiousity... how easy would it be to retrofit the home with fin tube baseboards ? There's a possibility you may need to address that issue at some point.

 

As for keeping the original pump: yes it really is a Model H-1, made all the way from the 1930s up through about the early 1950s, and the modern day equivalent is likely the Model PR (a bit like a Model 100, but capable of more pump head). I probably could repair it (it needs a new coupler and motor mounts), as about 5-7 years ago I set it up with a new mechanical seal section (sorted out from a Grainger catalog without getting into details of exactly what pump I was working with). But, I think (as a group), you have convinced me to pick up and install a Grundfos 15-58, if for no other reason but the considerable electricity savings and variable speed aspect.

For the boiler: I know I am probably living on borrowed time on this unit, and I am aware the efficiency is low; but hey, it works, and as mentioned, it is built like a tank. I generally run the temps fairly low (and modify them slightly as the season warrants); maybe 120F to 135F in general. When I first moved into the house, it was set up to run hotter, but it took me a few years to get a feel for how best to run this unit. With it running hotter, the concrete slab would effectively swing back and forth in temperature too much, with too much lag: now with the above temperatures, I basically can set up my (programmable) thermostat to try to achieve the desired overall temperature during the night, and actually turn off (by setting back a few degrees) during the day. Effectively, I can get a near constant temperature all the time this way.

My gut feeling from holding the supply and return lines in the past was that I had a considerable temperature delta, easily twenty degrees, but I can't really be sure. At least I have a better handle on what to look for now.

Tell me more about condensation problems--what should I be looking for? Wouldn't this only occur in situations where the boiler is located in a truly cold environment--mine is in an attached garage space, with fairly moderate temperatures (partially elevated by some of the copper tubing from the boiler crossing the slab on the way to the rest of the house).

The "auto" vent I am using actually has a way of closing completely by rotating the body---I could just do that, if the consensus is that I don't need to upgrade to a "real" or "enhanced" air separator system. I am curious though, why the backing away from diaphragm type expansion tanks? Are they really so troublesome or prone to failure?

Again, I really appreciate all the advice given. Many thanks.

 

You're welcome.

You would look for condensation damage (corrosion) in the flue and in the heat exchanger. There are a lot of variables that go into whether, when, and how much a boiler will condense. Generally speaking gas-fired boilers start to condense at return water temps <135F. Oil is a bit lower, around 105F IIRC. The problem (in a conventional boiler) is the system is running at such low temps the flue gases never get really hot and they condense. The condensate has a pH of about 4 or so, which is basically acid rain. Eats cast iron and most everything else. Condensing boilers (and modulating/condensing boilers) take advantage of the heat of condensation to extract a few more percent of efficiency by designing the heat exchanger so there's an additional pass to promote condensation and heat extraction.

So if you are condensing, then I'm guessing the reason the boiler hasn't failed is that you aren't condensing that much, and/or this 50s vintage boiler has such incredibly well-built and thick castings that you could go on for years before it gets eaten up.

I think Grady's suggestion about the air tank is good and follows your reasoning about the boiler "it works." If it ain't broke, don't fix it. Simple is good. Why start messing with all those components if you don't have to.

 

Personally, & I think the balance of the group concurs, diaphram tanks in general are preferable but in certain cases, like yours, the conventional type is better because of not needing to add anything else. As others have said, if it ain't broke, don't fix it.

 

Start thinking about your next boiler. Better to have a plan in case something happens. Hopefully you can coax a few years so that the price of modulating condensing boilers drop. They're way overpriced (and oversized) in North America.