I'm confused

Mark J Strawcutter

they do include the well, choice of 1/2in sweat or NPT

Mark

Eric Taylor

Boiler Flow Rate

Heat transfer should go up with flow rate in a given heat exchanger if all else remains constant. They tought me this in school and I have applied it several times in my short career as an engineer with predictable results. But now I have a situation at my house that seem contrary to this proven rule.

Single stage gas burner and one 007 pumping away. My boiler has been working overtime since this cold snap has hit my area and I have been observing its operation. There is a ball valve right on the outlet of the boiler upstream of the resorber and the pump. I found that by throttling this valve just enough to add 3 feet of head to the boiler my flue gas temp drops significantly. With the valve wide open I can't hold my hand on the flue pipe for more than a few seconds when the burner has been firing for a while, but with the valve throttled I can keep my hand there indefinately at all boiler temps (I used to wash dishes for a living). Not the most scientific approach, but the observation is true nonetheless. The distribution system is piped in parallel, so with the valve wide open I have almost no head across the pump and therefore max flow and low delta T across the boiler. Even with the valve throttled I am getting decent flow in all zones and the delta T across the boiler is only a tiny bit higher.

The static pressure in the system is roughly 13PSI at ambient temp and climbs to just over 20 at high limit. There is no air in the system that I can detect, and no operational issues to speak of.

Can anyone offer an explanation for this?

hot_rod

Thats one of those things that makes you go...

Huh...

There are many variables to flue gas temperature, and flue gas temperature without the other variables is really irrelevant. In order to see what is truly going on within your combustor, you'd need to do a flue gas analysis while changing the flow variable. I'm guessing that the efficiency of the appliance actually only changes one or two points, but the only way you'll know for sure is to perform steady state combustion tests.

ME

To Learn More About This Contractor, Click Here to Visit Their Ad in "Find A Contractor"

chris smith

things that make you go huh

my guess would be that by dropping your flow rate the delta t goes up in your baseboard, thus your return water temp. to the boiler drops cooling your flue gas temps down slightly, just a wag

chris smith

paradise porter maine

Boilerpro

My swing at it

if you have a sectional boiler, you have several heat exhangers working under on hood. If the water velocity is too great, it may be just shooting past the inlets and outlets of the sections closest to the supply and return and they get very little flow, causing less heat transfer due to higher average water temps. Most boilers have big passages inside and it is easy to over pump them. If you reduce flow to more reasonable rates, water flow can spread across the sections and pick up heat more effectively. Just a try at it.

Boilerpro

Steamhead

\"The distribution system is piped in parallel\".......

was this originally a gravity system that now circulates with a pump? If it has huge mains (2-inch or larger) and radiator lines (1-inch to 1-1/2-inch) it almost certainly is.

It is way too easy to over-pump a gravity conversion. There is almost no resistance in those big pipes to the load they must carry. The only real work the pump has to do is move the water thru the boiler. Your pump is probably oversized, since the water picks up more heat in the boiler when you slow the flow down.

Bell & Gossett had a formula in the 1940s that still works well for sizing circulators for gravity conversions. Measure the actual amount of radiation in the house (not the boiler's rating), then go to http://www.heatinghelp.com/newsletter.cfm?Id=125 for an easy-to-use chart based on this formula which will tell you how much flow you need.

If you're not sure how to determine the amount of your radiation, get a copy of Dan's book "E.D.R." which has charts for almost any radiator you might have. Order it on the Books and More page of this site.

To Learn More About This Contractor, Click Here to Visit Their Ad in "Find A Contractor"

Eric Taylor

More details

I think BoilerPro's take on this might be correct. The boiler is sectional cast iron and I have seen the exploded view pic in the documentation. Very large passages with the sections in parallel. I can envision the flow short circuiting through one section because the supply and return appear to be on the same end of the block.

Steamheads take also has merit, but this is not a gravity conversion. At least not anymore. Remember that I'm the nutbag that dumped all of his rads! I ran the new distribution piping in parallel because a loop system was not possible and a one pipe, monoflow system wasn't going to save any pipe or labor. I was also concerned with supplying one of the larger risers with diverter tees. Each room is on its own circuit and I have added TRVs to the second floor (TRVs are sweet!). The near boiler piping is 1-1/4" and then it tees into a 1" supply and return header that gets reduced again to 3/4" as it gets farther away from the boiler. Though the pipes are not 2" anymore, the flow characteristics are still similar to a gravity conversion system.

At first I thought I had a slug of air trapped in the boiler and changing the pressure was compressing the air and giving me more heat transfer area, but there is zero air in the system. I can reduce the system pressure to atmosphere by draining less than 1 gallon of water i.e. the volume in the expansion tank.

Last night I did the test again. I had left that valve throttled all day and when I got home I cranked the stat and let the system come up to high limit (185*F). I was able to keep my hand on the flue pipe as long as I cared to. Then, with my hand still on the flue pipe, I opened that valve fully. Within less than 30 seconds I had to remove my hand or get burned.

So my 007 is too big for the system? At zero head it should develope only 20gpm. I have a total of 12 3/4" circuits in parallel. Assuming perfect balance and full flow thats only 1.6666gpm per circuit. But I do not have perfect balance (at least I doubt it), and I am happy with the heat output of all circuits. I could try a 005, but when I do get air in the system I have found that the resorber removes it faster when that valve upstream of it is throttled just a bit, and I like that. I think I will keep the 007 and leave that valve throttled.

Do boiler manufacturers spec flow rates and pressure drops for their boilers? I know the Vitodens max out at 7gpm, but the design is very different from a sectional. I also see many install photos here that have five or more circs tied to the system. I assume that a situation can arise when all of the zones are calling for heat at the same time, or do the controls stage them? Even with pri/sec pumping the boiler flow rate would change when all those pumps are on at the same time unless a restrictor of some kind was installed in the primary loop, and then zones could starve for flow and cause pump cavitation.

This is more thinking then I wanted to do this morning!

Eric

Eric Taylor_2

Steamhead's table

I just checked out that newsletter about sizing pumps and it looks like I should be just under 15gpm. I doubt I match either the modern or the gravaty case exactly, but the output of the boiler is 85.3mbtu and I ran enough baseboard to load the boiler that high. So the 007 is 25% oversized and throttling or replacing it would result in better operating condidtions.

I love this web site and all of you folks that contribute!

Eric

hot_rod

Instead

of throttling, it's always best to size the pump closer to the load. Consider the 15-48 Grundfos with 3 speeds. This allows you to adjust flows without throttling and wasting energy. With so many pump choices on the market why not pick the right one instead of throttling an oversized one?

The best way to know how much heat you are sending into the system would be to install a flowmeter and a couple accurate thermometers. If you know the actual flow rate, and the delta t, you can calculate exactly how many BTUs you are are moving into the system.

hot rod

To Learn More About This Contractor, Click Here to Visit Their Ad in "Find A Contractor"

Eric Taylor

Flow Meter

Yup, I have been scrounging around looking for a flow meter that I could install and good temp gagues. Quality new instruments are big money! I have been guestimating flow with measuring the head across the pump, but this is not the best way. I wouldn't mind a varible speed pump as you suggested, but I suspect it would set me back a few bucks and pay for itself in a decade or two. Even if I went with a 005, I doubt that the energy consumption would change much. I'm guessing that the 005 is just a 007 with a smaller impeller. The 007 is 1/25th of a horse power at .6 amps if I recall. Not much of an energy consumption there. Works out to about 80 watts peak and less if RMS voltage is used for the calc.

I would still like to know how the pros spec boiler flow rate. I have Siegenthaler's book and he seems to skip that one. My boiler's manual has a great table for pressure drop vurses flow rate up to 100gpm, but dosen't specify what the desired flow rate/pressure drop is. 100gpm would REALLY be causing me problems! Are there rules of thumb or are you more concerned with system flows than boiler flows? I am very curious.

Mark J Strawcutter

primary/secondary

> Do boiler manufacturers

> spec flow rates and pressure drops for their

> boilers? I know the Vitodens max out at 7gpm,

> but the design is very different from a

> sectional. I also see many install photos here

> that have five or more circs tied to the system.

> I assume that a situation can arise when all of

> the zones are calling for heat at the same time,

> or do the controls stage them? Even with pri/sec

> pumping the boiler flow rate would change when

> all those pumps are on at the same time unless a

> restrictor of some kind was installed in the

> primary loop, and then zones could starve for

> flow and cause pump cavitation.




Don't think so. With P/S flow in one loop does not affect the flow in the other.

In the simple case where the secondary flow is twice that of the primary, you just end up with half the secondary flow rate (or an amount equal to the primary flow rate) going "backward" (from the perspective of the primary loop) thru the common piping.

Mark

Mark J Strawcutter

thermometers

Honeywell/Braukmann makes some nice 2" and 2.5" dial thermometers that mount in a well. I paid about $20 ea not including the well.

Mark

Eric Taylor

P/S

I'm starting to see the light. As long as the primary loop is sized to handle that reverse flow while still maintaining forward flow through the boiler I can see it working. Flow in both directions in the same pipe boggles the mind just a little, but there is no reason that it can't work that way. Vitodens' low loss header is a good visual aid too.

If you guys keep answering all of my questions like this I will soon be able to effect a career change to hydronics! Believe me the thought has crossed my mind.

I STILL want to know how you size the primary pump with consideration to boiler flow rate. Or is it done by the btus required and the pipe size? Seems I have some more reading to do.

Aidan (UK)

Primary/Secondary mixing

I'm with Mark. The primary/secondary system should ‘de-couple’ the pumps; flow variations, or even switching pumps on/off in one circuit shouldn’t affect the flow rate in the other.

I think that the reduced flue temperature is due to P/S mixing.

If the primary flow was originally higher than the secondary flow, then the primary return would consist of all the secondary return water, mixed with the portion of the hotter primary flow water that hadn’t been circulated in the secondary systems. Throttling down the primary flow rate has reduced the portion of hot primary flow water in the primary return, the boiler return temperature reduced and the flue temperature reduced.

If the secondary flow were the higher, you’d just increase the portion of re-circulating secondary water. In either case, the primary return temperature would drop. The dT across the boiler should increase, due to the reduced flow rate. You’d need to check secondary flow and primary return temperatures. Bet ya 5 quid.

Mark J Strawcutter

sizing

the common piping needs to be sized to handle the larger of the flows (pri or sec) so that when that loop is running alone you have enough capacity.

I strongly recommend Dan's "Pumping Away" and "Pri Sec pumping made easy" books.

Boiler/primary pump sizing - start with the flow. Boiler may require a minimum, else BTU of boiler and delta T (commonly 20degF) can be used to determine flow. Flow determines pipe size. Pipe size, length, fittings and boiler all determine head requirement for the primary loop (this is the part I've been having the most trouble with).
Then find a pump whose performance curve will give you the desired flow at the required head.

The head thru a residential boiler primary loop is pretty low. I suspect those in the trade have a couple of pumps they use depending on boiler size and throttle if necessary.

Mark

Eric Taylor

Whats a quid?

Just kidding. Its a pound right? We started talking about P/S and got off an a tangent. The system I started this thread with is direct pumped. I was going to go with P/S but I ran out of time on this job and it started getting cold. I think we resolved the issue of flow rate and flue gas temp with BoilerPro's theory that I was bypassing some of the sections in the boiler by over pumping, and Mark was setting me straight about P/S pumping and the decoupling factor.

Never in my wildest dreams did I think that residential heating was so complex! And don't even get me started on steam!

hot_rod

More thoughts

The pump I mentioned and a couple good thermometers would be under $150, even at street prices. www.blue-white.com has inexpensive flowmeters if you really want to mess with this.

I'm not sure how you came up with the 15 gpm needed if your boiler is 83,000? At a 20 degree delta t around 8 gpm would be the number.

Conventional cast iron boilers don't have much pressure drop. In fact the Viessmann bi-ferals posted here often operate without a boiler pump. Typically a cast iron boiler will just run up to the limit setting and shut down if the pump should fail.

Not so with copper tube, which is why the CORRECT flow switch is critical in their application. You'll know when you are looking at a copper tube boiler that is firing without flow!!

When you start dealing with copper tube or condensing style boilers with small passage ways, you need to be careful with flow rates. Generally the installation manuals will be real clear about the required flow rates at various delta t's.

The delta T YOU chose will dictate the flow rate need through the boiler. Here is an example from a copper tube boiler manual. Note how the flow rate effects the pressure drop and pump sizing criteria.

Siggys "Modern Hydronic Heating" is a text book. Virtually all the answers you need can be found there. However I find his columns are much easier to apply to real life installations and troubleshooting, not being an engineer The columns use the theories and calculations in Modern Hydronic Heating and apply them to the actual job.

Go to the RPA site and order the "Hydronic Design Toolkit" software also. It has all those formulas in an easy to use program.

Dan, ME and other hydronic writers do the same. They show the contractor how to apply the numbers to the task at hand. When you have time browse the www.pmmag.com website, and other trade mag sites, and read the archived columns. I'll bet you find one that matches exactly what you are doing. This one shows the difference in pump energy consumption rhttp://www.pmmag.com/CDA/ArticleInformation/features/BNP__Features__Item/0,2379,83636,00.html.

I am constantly reviewing these gems of information, and have many bookmarked on my 'puter. Priceless info is a click away these days.

hot rod

To Learn More About This Contractor, Click Here to Visit Their Ad in "Find A Contractor"

Eric Taylor

head requirement

Now you have me wondering about the head requirement of the primary loop. I'm with you regarding flow requirement, but would you need extra head to provide for the mixing? Pipe size, length, fittings etc. might be vastly dominant, but mixing the primary and secondary fluids would also consume flow energy and thus head especially if the two fluids are moving in opposite directions. Now I'm really getting into the nitty gritty.

Mark J Strawcutter

be the water

You're beyond my level of understanding now. If I had to take a guess :-) I'd say the mixing doesn't consume enough energy to matter in a residential system. Now if we're talking about 12in loop piping it might be a different story. Then again, relative to the other energy requirements it might still be small even there.

Mark

Eric Taylor

15gpm

I just looked at the table in the newsletter that Steamhead wrote. I made some assumptions that my system is neither a gravity conversion or a modern system, but has characteristics of both. I am leaning more toward the gravity conversion side of things, i.e. more flow than a modern system, but that might not be the best assumption.

http://www.heatinghelp.com/newsletter.cfm?Id=125

I'm between 63.3 and 96 mbtus and closer to 96 than 63. The guess is that I need more flow than a modern system and less than a gravity conversion and I squinted and saw "just under 15gpm". Squinting again I think I could go as low as 13 or so. The kicker is that my delta T is small with the current system operating either with the throttled flow or full flow. I haven't got exact measurements because of my lack of good gagues, but all of my poor measurements point to less than 10* delta, and it chose me. I replaced the distribution system but kept the near boiler piping when I ran out of time. Where did you get 8gpm? That would give me about .6666gpm in each circuit and make it rather difficult to entrain any air. Not that I couldn't move the heat with that flow mind you, but if I'm on the hairy edge regarding flow then system balance becomes a larger issue. I am on my way home now, so I just might play with that throttling valve again and see what happens at even lower flows. I'll grab a couple of thermocouples and some electrical tape while I'm at it.

More tomorro.

Eric Taylor

I tend to agree

At low flows typical of residential systems I don't see much mixing energy either. It just popped into my head when I read your post. I suppose if EVERY detail is engineered to exact soloutions and there is zero flow energy to spare it might crop up at design conditions, but it would probably be insignificant otherwise.

Steamhead

Eric, if you want to experiment

get a Grundfos 15-42 that has the 3-speed controller. Then get an infrared heat gun like the Raytek I posted a while back. This will let you make and measure changes easily.

Remember that the best flow rate thru fin-tube heat transmitters such as baseboard or convectors is more than for cast-iron radiators. This is because those big old lumps of iron have much more heat-transfer surface in them, and the slower rate takes advantage of this.

To Learn More About This Contractor, Click Here to Visit Their Ad in "Find A Contractor"

Online Sheriff \"Murph\"

ME HITS IT !!!!!

20* DELTA TEE, get the combustion peaked at steady state, what you do with the rest of your feul savings is just fluff !!


Murph' (SOS)

Mike T., Swampeast MO

Time

You're homing in on the ideal flow rate for the installation at maximum output.

You're approaching the point where the water stays in the boiler just long enough to pick up the maximum BTUs output by the system.

The drop in flue temperature {should} equate well with increased efficiency.

Aidan (UK)

Doh!

It can't be that then. I hadn't followed the description. It must be as Boilerpro/Steamhead said, you're getting an increase in the boiler efficiency by reducing the pump flowrate. Now I'm wondering if my own boiler's doing this.

Where should I send the fiver?

Mark J Strawcutter

You're headed in the right direction

and toward the conclusion that boiler flow and system flow requirements don't always agree. Primary/secondary to the rescue!

Consider a primary boiler loop with properly sized circulator and bypass using ESBE TV style valve if the boiler doesn't have it built in. Secondary system loop with it's own properly sized circulator, TRVs on all the rads, and a pressure differential bypass.

Lastly, run the secondary on constant circ. Outdoor reset controller with water temp sensor on secondary return. The outdoor reset controller connects to existing boiler controls in place of thermostat - existing control handles burner and primary circ.

Mark

Boilerpro

But, Steamhead.....

The chart calls for more flow through a converted gravity system than a baseboard system. Uh oh! Here we go again!

Boilerpro

Eric Taylor

Don't worry about it

You can keep the fiver. Not that I couldn't use it, but it wouldn't be a fair bet given that you didn't have all of the info at the time

Eric Taylor

yup

I did some more reading last night in Siegenthaler's book, and came to a similar conclusion. If I design for a 10* drop then I need 17gpm to move 85,300btuh and this flow is probably a bit high for good boiler efficiency. At 20* drop then I get what HotRod said above, 8.53gpm. That is too low a flow for my system piping. If I go with a 15* drop then I need 11.37gpm. I believe this flow will work for both the boiler and the system piping rather well, so I can keep it direct pumped. It just so happens that with that valve in the throttled posistion that I have been using, I measured a Delta T across the boiler last night of almost exactly 15*! Since I was home alone last night I walked around in my skivvies and found the comfort level in each room to be extreamly high at these conditions--not that comfort was low before, but I can feel the difference. Oh, and that 007 at 1/25th hp is only adding 30 watts of energy to the fluid! The hp rating is net output and the electrical rating (115V @ .7 amps) is gross input. electric motors are not very efficient! Therefore the throttling isn't wasting much energy, but I am paying for electrical losses running a larger pump than I need to. I think I am really starting to get this stuff! Now I need to calculate my system's flow curve and see if I can get a pump that will give me 11 or 12gpm at the right head.

Now I am sure that part of the problem with the old gravity conversion system that I ripped out was over pumping. The other BIG problem was direct pumping, and another problem was poor radiatior sizing and placement. Not to mention the fact that almost every rad was painted silver and very obtrusive. I most likely could have kept the rads IF I switched to a proper P/S design and shuffled the rads around to better meet the load in each area. I don't love that old iron the way you folks do though. I still don't buy the claim that 40% of the output of a CI rad is radiation. I know three people that have CI rads. Two are gravity conversions and one is a one pipe steam system. I recently observed one of the gravity systems and the steam system at max operating temp and I just don't feel the radiation untill I am very close to the rads. Even the steam rad operating above 200* and hot all the way across is belching heated air above it while the radiation is barely detectable untill I stand within 4 feet of it. Convecting that hard in one spot gives rise to some major stacking. Finned tube wall to wall distributes the convective currents over a greater area (volume?) and the result is lower heated air temps leaving the convector and less stacking. Sure I could have kept my rads, but I STILL don't believe that I could have been as comfy as I am now.

Got off on a tangent again!

Thanks again to all of you who so willingly share your knowledge!

Eric

Steamhead

This Paradox

is easily explained. A gravity conversion has more water in those big pipes, so you need a bit more flow to move it all. But you still get that higher flow rate with a small pump, because there's so little effort needed to move the water.

BTW- got the NRF-9 and the Delta-T on my system is now up to 15 degrees. I wonder if the head in my system even approaches 3-1/2 feet?

But the smaller circ helped solve a long-standing puzzle- the rad I had installed in the back bathroom (converted from a pantry) never worked quite right. It's connected top and bottom on the same side, and somehow I had gotten the pipes reversed- the hot water was coming in the bottom. This was easier to spot with the slower circulation. I reversed the pipes and now it works great!

To Learn More About This Contractor, Click Here to Visit Their Ad in "Find A Contractor"

Boilerpro

Oh, I agree, but the table calls for more flow, not...

a bigger pump. I'm not disgreeing with the table, but more flow in the system pipng also means more flow through the heating units....in this case rads, than you need for a convector system. I don't think the rads need the flow, its' just needed to help balance the system...through enough water at the system and its bound to hit all the rads. And I do agree, alot of gravity conversions are way overpumped. I've replaced a few 1/6 hp 2 inch circs with NRF-22's on systems with only 600EDR of cast iron.

Steamhead

Bet they were line-sized

someone saw 2-inch mains and installed a 2-inch circ. Kind of like line-sizing steam traps, eh?

With 600 square feet on a gravity conversion you could use an even smaller circ. The NRF-22 has a capacity similar to the Taco 007. Try a Taco 005 or even an NRF-9 on the next one of that size. It should work even better.

You need more pump on a newer forced system because there's more head to work against. The flow rate you'll get is lower, but there's less water in the pipes so the system will respond as quickly as a gravity conversion with a small circ.

To Learn More About This Contractor, Click Here to Visit Their Ad in "Find A Contractor"