Plumbing Layout for Viessmann 200-W: feedback please

markp1

Hi,
Here is a logical layout for a Viessmann 200-W with an IWH and 2 finned baseboard zones using the Viessmann 80/60 low loss header. The two zones are using Taco Bumblebee's in Delta-T mode (20 degree). The Grundfos boiler / DHW pumps are fixed speed. All 4 pumps have integral check valves. The top of the low loss header has a caleffi auto-air purge on it.
Anything I am missing, or anything I should do differently? Feedback is greatly appreciated.

Thanks,
Mark

SWEI

Which model of boiler, and what is your estimated design day heat loss?

markp1

Viessmann 200-W B2HA-19 is the boiler. Design day is 5F, requiring 57k BTU (according to room-by-room manual J), requiring 48k BTU by degree day and past consumption history.

Floor 1: 35k BTUH, 132 linear feet of baseboard = 265 BTUH per linear foot = 135F AWT @ 3.5GPM with delta-T of 20
Floor 2: 22k BTUH, 102 linear feet of baseboard = 215 BTUH per linear foot = 125F AWT @ 2.2 GPM with delta-T of 20

Floor 1 has 3 separate home run loops (hence the 3 pipes off the one Taco Bumblebee). Floor 2 has 2 separate loops, but pipes don't split until at the second floor, hence the diagram only showing a single pipe out of the other Taco Bumblebee.

SWEI

57k BTU at a 30˚ΔT is only 3.8 GPM. Given that your baseboard is running at such low temps, you may be able to direct pump this (HX offers about 2.1 ft of head at that flow, and perhaps 2.8 feet for a 25˚ΔT.

markp1

Direct pumping would be great, as it would eliminate the LLH and an extra pump. A larger delta-T (30 degrees) would be more than welcome as well, to get the efficiency higher.

I spoke to Viessmann, and the minimum flow is 0.5GPM and the maximum is 6.2GPM. A delta-T of 30 would be 4GPM on this boiler as well. So it looks like it might work great.

With two zones, how would you suggest it is plumbed? One pump for each zone, or one pump with zone valves?

The part I am fuzzy on is either way, how do I ensure the flow rate of the boiler isn't exceeded?

Much thanks!

SWEI

At those low flow rates, I would use a single pump and zone valves. No sense running a circ near the bottom of its curve all day long -- you lose most of the benefits of variable speed that way.

If you're using a ΔT circ, I wouldn't worry about exceeding the maximum flow rate. At full output, 6.2 GPM would give you a 19.7 degree drop, so don't set the pump that low.

I don't recall you mentioning it, but is any of this baseboard piped in series? If that is the case, a 30˚F ΔT is unlikely to work well.

markp1

Any suggestion on which pump to use / what setting to put it on? A Grundfos Alpha (delta-P) would seem to be a better fit for this by my thinking, but I don't know how to keep it from exceeding the maximum flow rate of the boiler.

SWEI

I suspect ΔT might work better in this system. The MAGNA3 has a FLOWlimit function, but I'm not sure whether the Alpha has something similar or not.

markp1

The magna3 is super expensive... yikes. The alpha only has settings of constant speed or constant pressure, and auto-adapt, no wired connection for a flow limit.

If a delta-T pump, then the Taco Bumblebee would be the choice, but again, I don't know how to protect the boiler from going over the flow limit... any ideas on this?

SWEI

The MAGNA3 is really for commercial applications (or perhaps a palatial house.

Set the 'Bee for a 25˚F ΔT and it will stay under 6.2 GPM in that system.

markp1

Fantastic Thank you! Here is an updated diagram... this look right?

SWEI

Looks good. You might consider adding ball drains on the returns (or at least one on the common return.)

icesailor

I hope that you accidently placed that Watts 9-11 Combo back flow and fill valve unintentionally like that. Its shown backwards. The 9D protects the potable water system from the source of unknown water. The boiler doesn't need to be protected from the potable water in the structure.

If you've been installing them like that, its wrong. Read the instructions.

SWEI

Good catch, Chris. We generally don't install them at all (at least not permanently.) I have a Caleffi setup with hose thread adapters that we move from job to job. Connect it to a drain port, leave it on for the first week or so of operation and then remove it from the premises. On DDC jobs, we monitor system pressure and throw an alarm if it drops more than a couple of PSI.

markp1

Well that's just downright embarrassing on the backflow / fillvalve... I forgot to flip it around in my drawing Thank you for catching it.

Here it is (flipped around correctly) with ball drains added to each return:

NYplumber

With this boiler you would want to be pumping into the boiler, not away from it. Just my 2cents.

icesailor

markp1 said:

Well that's just downright embarrassing on the backflow / fillvalve... I forgot to flip it around in my drawing Thank you for catching it.

Here it is (flipped around correctly) with ball drains added to each return:

The best part of the drawing is the ball valves on either side of the 9-11. So you can service or change the valves without draining the system. Just don't put black nipples between the 1156 and the valve or connecting to the system. It will cause the 1156 to fail. Electrolysis.

Not to be picky, but how well does that overpriced microbubble scrubber thingy work when you install it like that? I never used them because when they leaked or if it was a system that had to be winterized, I adapted a float vent on the outlet for air removal. Have you ever tried to get one of those apart? A lot of people must do it because my wholesaler had replacement repair kits. They work especially well on systems with Glycol in them. When they get the green goop buildup out of the top, time for a 1/2" black coupling, 2" nipple and a reducing ell with a float vent.

markp1

NYplumber said:

With this boiler you would want to be pumping into the boiler, not away from it. Just my 2cents.

I was going off the layout from Viessman (layout 1 in the manual, attached below) which shows the pump pulling away from the boiler. Have you experienced that this isn't a good way to do it?

NYplumber

With a high head loss hx, you should be pumping into.

markp1

icesailor said:

The best part of the drawing is the ball valves on either side of the 9-11. So you can service or change the valves without draining the system. Just don't put black nipples between the 1156 and the valve or connecting to the system. It will cause the 1156 to fail. Electrolysis.

Thanks, I've added a copper MPT to sweat union on the 1156 to make the connection so I won't run into an issue with electrolysis there.

icesailor said:

Not to be picky, but how well does that overpriced microbubble scrubber thingy work when you install it like that? I never used them because when they leaked or if it was a system that had to be winterized, I adapted a float vent on the outlet for air removal. Have you ever tried to get one of those apart? A lot of people must do it because my wholesaler had replacement repair kits. They work especially well on systems with Glycol in them. When they get the green goop buildup out of the top, time for a 1/2" black coupling, 2" nipple and a reducing ell with a float vent.

I haven't had to take one apart, but I consider it cheap insurance for an expensive boiler being connected to a 40 year old piping / baseboard system. You did make me realize though that I didn't have an air vent anywhere in the system, so I've added the Caleffi vent on the top of the magdirt separator.

NYplumber said:

With a high head loss hx, you should be pumping into.

Point well taken, and the drawing updated to the same:

markp1

So I think v2 I uploaded above is wrong, as the Point of No Pressure Change (PONPC), at the expansion tank, has the pumps pumping to it.

I have changed it to connect to the supply side instead of the return side, and here is the updated version. This is also the side that the Viessmann diagram 1 shows the expansion tank. Is this a better way to connect it?

Thanks.

NYplumber

One more item, add a flow switch to cut the system off if the flow slows too close to minimum flow. You dont want to warp that hx if the bumblebee decided to take a few minute break while the boiler goes into high fire. My guess would be to wire it through the "tt" terminals. Viessmann has their own lingo.

markp1

According to the Viessmann tech I spoke to on the phone, the minimum flow is 0.5 gallons per minute. I can't see the taco bumblebee flowing less than that, unless it is dead. I am also unable find a low flow shutoff valve that goes down to 0.5 gallons per minute in a 1" diameter pipe. How important do you think this is to add?

markp1

Also, the Taco Bumblebee always on start goes full speed for 3 minutes in delta-T mode.

markp1

I think I made a fairly major mistake... I did not include the head loss of the boiler itself in the direct piping design. At 58k with a delta-T of 20, that is 5.8GPM flow through the boiler, the boiler head loss is around 14 feet at that rate. If I add that to my loop with the highest loss (8.4 feet), I have 22.4 feet of pressure to overcome at 5.8GPM. A Taco Bumblebee won't come close to providing that. Additionally, the indirect water heater is a Superstor SSU-80, which has a head loss of 9.1 feet. Add the boiler of 14 feet, and at 23 feet, the Grundfos 15-58 won't handle that either.

I think this means I cannot do direct piping (without going to larger pumps). Looks like I should redo the design, using a LLH, and having the indirect water heater off the system side of the LLH as well. Am I on the right track?

icesailor

markp1 said:

NYplumber said:

With this boiler you would want to be pumping into the boiler, not away from it. Just my 2cents.

I was going off the layout from Viessman (layout 1 in the manual, attached below) which shows the pump pulling away from the boiler. Have you experienced that this isn't a good way to do it?

The drawing from Veissmann shows the circulator pumping IN TO the return of the boiler.

icesailor

"" I think this means I cannot do direct piping (without going to larger pumps). Looks like I should redo the design, using a LLH, and having the indirect water heater off the system side of the LLH as well. Am I on the right track? ""

I'm no expert in what Veissmann wants, but tht first one and all the other ones I did, I used a LL Header. I can't see any reason to not use one in every installation made. It eliminates ALL issues if P/S and maximum flow through the boiler. Without a LL Header, what you have drawn really isn't P/S and when the zone valves are opening or closing, it restricts the flow of water through the boiler.

I think that the 100 and 200 use the same piping drawings. If you use their piping suggestions and use a LL Header, you will have no problems getting everything where it needs to go to work properly. If you follow their drawing using a LL Header and indirects, there will never be an issue. When you do it with a LL header, you don't have to calculate the primary flow and head loss through the boiler. Just use the appropriate pump that Veissmann recommends and leave the boiler out of your head loss calculations on the secondary side.

If I was still doing it, and I had my way, every boiler would have a LL Header. The hydraulic separation becomes a wonderful thing.

markp1

Thank you for the guidance / confirmation, much appreciated. Here is the updated diagram, that takes into account the head loss / flow rates required at design day for each loop. Certainly looks a lot more complicated to me than my previous diagram, but the single Taco Bumblebee can't do my whole house, even with the LLH.

How does this look? Thanks again.

icesailor

I might do it something like that. It should work.

RobG

I'm confused? If a single bumblebee would work before, why won't it work now?

MikeL_2

Mark,
My tablet won't let me open your pdf files, but it sounds like you're on the right track. I attached some photos of our most recent install; they may represent what you are trying to accomplish........

markp1

RobG said:

I'm confused? If a single bumblebee would work before, why won't it work now?

Hi Rob,
Great question... maybe I should explain my math (because I might have got it wrong) and my assumptions:


1) Water should move at a minimum of 2 feet per second and a maximum of 4 feet per second, which in a 3/4 copper pipe, means between 3 gallons per minute and 6 gallons per minute.

2) Since all heating loops are 3/4 copper, each heating loop should flow between 3 and 6 GPM.

3) If I balance each heating loop to put out the same proportional heat per the BTUH requirements of each, then the flow rates at design day are: 3 GPM, 4GPM, 3.6GPM, 4.7GPM and 4.7GPM.

4) This means the total flow in the secondary loop is 20GPM.

5) After the heating loops are balanced, the loop with the most head pressure is 9.2 feet, so that is the head pressure for the secondary side for sizing a pump.

6) So if using a single pump on the secondary side, it needs to supply 20GPM @ 9.2 feet of head pressure.

7) A single Taco Bumblebee cannot provide 20GPM @ 9.2 feet of head pressure.

8) If I tried to pump direct off the boiler, 20GPM will exceed the maximum flow rate of the boiler (6.2GPM max).


Where I originally thought I could use one pump, was due to a delta-T of 20, with pure water, at 58k BTUH requirement on design day, the flow rate through all loops combined would be 5.8GPM. (58000 / (20 * 50)). With a max head pressure of 9.2 GPM, which a single Taco Bumblebee could do, and would allow for direct plumbing. However, I didn't add in the head loss of the boiler, which is 14 feet at 5.8GPM, bringing the total loss if plumbed directly to 23.2 feet, which the Taco Bumblebee cannot handle.

Did I do something wrong above?

Thanks again

markp1

MikeL said:

Mark,
My tablet won't let me open your pdf files, but it sounds like you're on the right track. I attached some photos of our most recent install; they may represent what you are trying to accomplish........

That looks really nice Mike, thank you for those. That is very close to my design.

Gordy

This is why low loss header, or P/S piping is important with high head hxers. You can run a 35*'delta at the boiler, and 20 at the emitters. That's why There is options for running higher delta t's at the boiler in most mod/con manuals. Fire tube hxers have a little more latitude when piping direct.