Cavitation issues

[warno]

@warno, to also have four 45 elbows, not two. Don't forget about the two you used to make the turn out of the boiler jacket.

Oh yeah. Changed it here.

 

[Marshy]

Now, check how much the NPSH increases if you moved the pump to the floor. Go back and add the distance to "Tank fluid level above or below centerline of pump". You will have to add additional length of pipe as well, possibly an elbow? You could take the opportunity to increase the pipe size as well.

 

[warno]

@Marshy and here is with length added and pipe size increase. I thought it was posted in this thread but i can't find it, does anyone know the NPSH for the NRF-36 I'm running. I checked the manuals that supplyhouse.com offers but couldn't find it. Unless i over looked it.

 

[Marshy]

@Marshy and here is with length added and pipe size increase. I thought it was posted in this thread but i can't find it, does anyone know the NPSH for the NRF-36 I'm running. I checked the manuals that supplyhouse.com offers but couldn't find it. Unless i over looked it.

View attachment 221044

View attachment 221045

I have looked through B&G tech manuals for this pump but no NPSH data was available. I would give their tech line a call and ask if they can provide it.

 

[Bob Rohr]

Marshy, explain to me how 48" of water column give you those pressures? I come up with 1.7 psi with the pump 48" below the water level?

 

[Bob Rohr]

On smaller circulators generally they provide a chart like this. So at 190F you need 2.2 psi on this Grundfos. I've see 4 psi on other brands and models at 190F, depends on the type of pump to some degree.

I'm of the opinion a 4' volume of water above the circ is around 1.7 psi. Hence the cavitation potential in your case.

 

[warno]

@Marshy I called a B&G rep and they told me the NPSH is so small they don't publish it anywhere. He did say that dropping the pump lower would help. I asked if the larger piping would help, he said it may or may not. And i asked if going horizontal with the pump to get about another foot of pipe before it, he suggested staying vertical for fear that elbow right before the pump might cause issues.

As far as the NPSH is there any other manufacturers that publish that number? Maybe we could crodd reference with a similar pump.

Also he said my impeller might be hurt from thus going on and to check and replace if it's bad. Because a bad impeller would defiantly help cavitation.

 

[Marshy]

On smaller circulators generally they provide a chart like this. So at 190F you need 2.2 psi on this Grundfos. I've see 4 psi on other brands and models at 190F, depends on the type of pump to some degree.

I'm of the opinion a 4' volume of water above the circ is around 1.7 psi. Hence the cavitation potential in your case.

I guess I left my brain at home when I was look at the numbers. I have to agree with you. We are probably using the calculator wrong. I'll play with it or have to break down and do it manually (been a little while, would be good practice).

I put in 1' for the water level to CL of pump, 0.1" of pipe, and no fittings, 1" pipe with 0.1" length and 0.1 gpm and it still says NPSH is 4.96 psig lol

 

[Marshy]

I'll have to revisit this later and do it on paper.

 

[Bob Rohr]

Here is the explanation for static pressure calculation.

As we all suspect Warno is on that fine edge with his static and cavitation, he has about 1.7 psi, the pump would like 2. Since he cannot change the pressure change the temperature is his only viable variable.

 

[warno]

If I take a pressure reading on the suction side of the pump it would be vaccuum reading right? Or no?

I have a port on my suction side i can put a gauge on if this information would help us figure this out. And there's another port i can take a reading at the return just before going into the boiler.

 

[maple1]

Here is the explanation for static pressure calculation.

As we all suspect Warno is on that fine edge with his static and cavitation, he has about 1.7 psi, the pump would like 2. Since he cannot change the pressure change the temperature is his only viable variable.

But then he is getting less heat to his storage. His biggest Catch-22.

 

[Marshy]

If I take a pressure reading on the suction side of the pump it would be vaccuum reading right? Or no?

I have a port on my suction side i can put a gauge on if this information would help us figure this out. And there's another port i can take a reading at the return just before going into the boiler.

View attachment 221052

No, very unlikely. If it was then the pump wouldn't actually pump anything it would become air bound. As the fluid enters the eye of the pump the pump is creating a low pressure from the suction of the impeller. That low pressure is localized right in front of the impellar close to the moving vanes. Its just enough for some of the fluid to flash to vapor. It then recombines to liquid when it reaches a higher pressure, usually in the volute of the impeller. The collapse of the vapor bubble gives the noise you hear and can damage the surface of the impeller.

 

[maple1]

If I take a pressure reading on the suction side of the pump it would be vaccuum reading right? Or no?

Now we're getting into Point of No Pressure Change things. This thread keeps getting better.

I had a hard time getting my head around that one. But theoretically, I THINK, if the top of the water in your boiler is the PONPC, and your circ was at that same spot, all the pressure diff generated by your circ, would go to the outlet side of the circ. So if it is very close to it, which yours is, then practically speaking, you should see all the pressure difference, at the outlet. I can picture it better with a closed system, and expansion tank connection - so this is a bit different from that.

 

[warno]

@Marshy so there is a positive pressure reading at the suction side of the pump. I understand with the pump not running you would see the pressure from the water weight but if the pump is drawing water down the pipe is it still reading positive pressure? Would a pressure reading at the suction side of the pump tell us anything we need to know?

Keep tuned in @maple1 we're going to get this. Lol

 

[Bob Rohr]

@Marshy so there is a positive pressure reading at the suction side of the pump. I understand with the pump not running you would see the pressure from the water weight but if the pump is drawing water down the pipe is it still reading positive pressure? Would a pressure reading at the suction side of the pump tell us anything we need to know?


Keep tuned in @maple1 we're going to get this. Lol

If you pull a sub atmospheric condition, water boils at lower temperatures. Once you boil or create vapor hand that induces cavitation.. The vapor pressure is what determines that point, again pressure and temperature relationship. Water will boil at 100° if you pull enough vacuum

 

[warno]

So I would like to switch gears. I'm understanding my pump is going to be flirting with caviation basically no matter what I do besides running lower temperatures. I'll go ahead with the lowered pump location like talked about before and I'll plumb in the larger piping for good measure.

But what I just realized, took me 2 years, is my FPHX is killing the heat transfer no matter what I do. It's only a 20 plate that measures 4 1/4" x 12" x 2".

So my question is is it possible to hit a target of mid 180s with a HX that measures 5" x 12"? What plate count would that take? I haven't seen a 5 x 12 option on flatplateselect.com. I'd really like to stick with getting one from eBay and those are the dimensions that are in my price range for a new HX.

 

[Bob Rohr]

So I would like to switch gears. I'm understanding my pump is going to be flirting with caviation basically no matter what I do besides running lower temperatures. I'll go ahead with the lowered pump location like talked about before and I'll plumb in the larger piping for good measure.

But what I just realized, took me 2 years, is my FPHX is killing the heat transfer no matter what I do. It's only a 20 plate that measures 4 1/4" x 12" x 2".

So my question is is it possible to hit a target of mid 180s with a HX that measures 5" x 12"? What plate count would that take? I haven't seen a 5 x 12 option on flatplateselect.com. I'd really like to stick with getting one from eBay and those are the dimensions that are in my price range for a new HX.

Yes you can size a HX to do that with close approach sizing. Generally you would be running the A side 10- 20° hotter then what you want from B side. But we have been down that road, running 190 just isn't very doable with an open system.

And one key number is still missing, the BTU output of the boiler? So pick a reasonable number, my suggestion would be somewhere in the 120- 140 K actual output. If you can get your hands on a combustion analyzer it would get you a good idea of the burn efficiency, I'd guess 70% when it is in it's sweet spot, with quality, dry wood? So my guesstimate on actual output is based on the physical size of your unit and estimated efficiency. it looks similar to a Royall wood boiler I opened once, and it ran in the 60% range

Most HX programs let you select close approach design, maybe try 185 input on the A side, 140K input, 12 gpm. Enter that into your HX simulation and see where it takes you.

Obviously to get the two temperatures close to one another it takes more surface area.

As someone mentioned earlier the inexpensive brands, even though it looks like then plate count is high, it the surface are (square footage) that you want.

So if you use the simulator you have, note the Sq Ft on the one it selects for you, then be sure the E-bay version has the same area.

The sim you posted on Tuesday was on then right track, dial down the 199/ 190° numbers to 185/ 180 perhaps. The gpm and BTU looked reasonable 10 gpm, 145K

We did a long hand example of HX sizing in Idronics 6 once you see the math involved you learn to appreciate the SIM programs.

 

[warno]

@Bob Rohr I designed my boiler based around a portage and main unit. My boiler is almost identical in dimensions and gallons as the "BL 34-44" on their website.

Here's their site

I can't tell you if my output matches that but the design is basically the same. They have a brochure you can click and look at a PDF to see the different units.

And you're right, that math does make me appreciate the online calculators. Looking at it makes my brain hurt.

If I put in some numbers on flat plate and get this HX

Basically i just take the cubic feet of those dimensions and search eBay for one similar. Is that right?

So in this case i need to search for a .494 cubic feet HX?

 

[jwise87]

I have been slogging my way through the 45 page pdf someone posted earlier in this thread from idronics. If I read it right you can easily find the head pressure of an existing system by checking pressure both sides of the pump, which you guys were just talking about, but since this is an open system you should be able to come close by either using a reading of slightly above 0 or slightly above14.7 for the input side and just taking one pressure reading as close to the output of the pump as possible.

Talking about reading pressure at the input side, even if you insert a gauge and it shows 0 the absolute pressure there is still slightly above an atmosphere, unless your gauge shows a vacuum.

Last I don't think you can use the port at your return because it is after all the "loads" and will probably read some where near your resting pressure in the tank.

 

[Bob Rohr]

36.8 square feet of heat exchange surface is what you need for those conditions. I can't say for certain all brands are manufactured to the same quality or performance.

Ideally you would look for this same data from a HX manufacturer you are considering, to assure they had actually tested and confirmed the performance. That is why just a BTU rating on a HX is a bit misleading. As you know a lot of import knock off stuff is a value engineered product, I use the term engineered with tongue in cheek in this case.

I've cut open quality HX and cheapos and there is a difference in the technology inside.

I would not worry so much about pressure at the circ, it would take an accurate and expensive gauge to read those low pressures accurately. You really can't change that condition very much. We paid $1300 for an accurate certified delta P gauge used for balancing.

 

[warno]

36.8 square feet of heat exchange surface is what you need for those conditions. I can't say for certain all brands are manufactured to the same quality or performance.

Ok here's a 90 plate on eBay for $310 shipped

If my math is right it's got 37.53 sq/ft of area.

 

[Bob Rohr]

that should do the trick.

 

[warno]

that should do the trick.

Now my question is what can expect for pressure drop through that thing? Will my current pump be able to feed this HX?

These are the numbers posted on the eBay listing, but it's saying 26.66 gpm at a drop of .3 psi for a delta of 20 degrees. There's no way I could ever stuff 26 gpm through there.

 

[maple1]

Your dT would come down to what's happening on the other side of the HX. That's kind of a confusing chart, that doesn't help a whole lot to the big picture. But I guess it does provide a pressure drop at a certain GPM. Which one should be able to extrapolate from?