sight glass water high, boiler water low

Christian Egli_2

When traps play Marco... Polo around the pool

This is better than working out a detective story. Here's my attempt at solving this 'who done it':

My finger points to the obstreperous F trap.

But first.

Air handlers can suck lots of heat, just on a flip as soon as the valve opens or the fan goes on. Not that there is anything wrong with that, but this can create big and sudden demands for steam. We thus can easily imagine everything going nice and easy until there is a scrambled rush coming out of the boilers.

You already suspect this, I believe you and I think this makes your system prone to surging.

It seems the piping is already border-line sized to handle the load. A rushing load makes it more critical. A condensate puddle will aggravate everything as it strangulates the gasping steam flow.

There, that's our cadaver.

As an aside, are those air handler coils fitted with a vacuum breaker? If not, when the supply valve is shut a vacuum will prevent condensate from draining. Upon reopening of the supply valve, this situation could aggravate the surging and overload the condensate system.

Back to the boiler room

All the pipes seem small. (I understand there are 2x40 hp boilers, that should be something on the order of 3,000,000 BTU/h and the steam main and each of the boiler take offs are (4 ?) inch. The 4 inch steam main really seems small.

But, it is not necessarily a problem. It just means the steam will travel very fast and careful attention should be given to the pipe pitch and condensate removal.

So it is easy to believe that 1) the two take offs entrain water with them from the boilers, and 2) that the main leaving the header, since there is only one, has an even bigger urge to entrain water.

Your header diameter looks good, this feature alone should help in dropping out the entrained water, and we all suspect there is lots of it....

But where does this water go, where can it go?

Header objection no. 1

There is only one small (3/4?) float trap. This may be enough on a stable system with no entrained water to get rid of, but not on yours and not on start up. Headers work best when piped into an equalizer-like line, no trap, nothing, just back to the boiler.

On start up and at surging times there is a Tsunami wave heading to that little float trap.

Header objection no. 2

The two streams coming from each boiler butt each other while trying to get out the header through the central steam main exit. Fluid flow is much smoother when avoiding these patterns.

The result:

Imagine what is going on in the header. It is a bathtub filled with condensate splashing around from end to end with a big unstable bulge rising under the central main departure.

With sudden demands from the air handlers this whale you've got playing in your header will really perturb the smooth flow of steam coming from the boiler. The flow is all stop and go now. Of course this causes the boiler water level to bounce. The low level shut off trips and that's the end of the story.

More suspects:

Let's round up the Peerless steam take offs. Looking at the third picture called steam piping we see:

The fore take off carries steam towards the rear, down the way it picks up the aft take off, on the same plane level, then the exit from here for the steam is to the top, for the carry over water it is straight forward. The aft tee and the steam exit are near each other (which makes things a little worse).

The steam from the fore take off travels straight through the tee while the steam coming from the aft take off make a right turn on red at the tee. Like us in traffic, this steam has to yield. Tee's set priorities that way.

What does it mean?

Well, during surge times, more steam will flow out of the fore take off. This will also mean that the water level in the boiler will side with this take off and form a bulge under it.

The low level shut off is in the center of the boiler, a neutral position. The glass gauge is on the front of the boiler. The glass gauge would see the bulge while the low level shut off would not.

I believe there is strictly nothing wrong with the boiler trim.

When things shut off, the bulge goes soft, the level goes flat, some entrained water falls back from the take offs and I suspect you find the resting boiler level to be back above the shut off line.

We can all go smoke a cigarette.

One more aside. I see a float trap in the third picture, is that a trap that is at the bottom of the Peerless equalizer line? isn't the equalizer line directly connected to the boiler bottom? If not, this would be a big contributor to your problems. But, to me, the trap seems to do something unrelated.

Lastly, the issue of the condensate receiver overflowing.

I suspect the pump stays off when the boiler goes out. Remember also, the seemingly low level is only dues to bounce, not to actual low water, so there is no call for feeder water.

However, our bathtub full of water in the header still has to drain out (slowly through the smallish float trap, but out nonetheless) This water is hot hot hot, and a float trap will pass hot water.

So, no surprise, the condensate system fills with hotter than ordinary spent steam condensate.

What I would try to figure out:

Open the hand drain on the header and see what comes out. Possibly even install a glass sight gauge on the header to see what goes on inside.

Improve the drain on the header.

Either build an equalizer line directly to both boilers. This is the best way to go. If altitude is a problem, then raise the header to go above boiler water level. For instance, you could remove the main valve on the departing main, gain that bit of height and raise the header by that much, and bend the boiler take off lines as needed. You could also remove the elbows that make this header into a drop header, you'd gain in height and you'd still have a fine header.

Or, improve the trap by installing the super extra big model and putting one on each end of the header. Super big traps are budget busting expensive and inelegant. This could possibly be more costly than moving the header.

On the exotic side, an implementation of the old alternate return trap could easily pump the condensate back into the boilers without using rotating pumps and the dangerous cavitation they suffer from. Pumpless transfer pumps are still built brand new today.

To improve the steam flow from the Peerless boiler to the header and to improve the butt heading of steam, you could install a new line from the fourth and unused valve to the fore take off. the Peerless boiler would now have two lines to carry steam and carryover water with both lines causing identical and canceling bounce to the water line. The Hurst boiler still would have only one line, but since it is much shorter it might work just fine.

There, that's my theory. Hope you find it interesting, I think this thread is. Where are you located?

Happy steaming

RAW

Looking forward to HAPPY STEAMING!!!

First off I am in North Carolina and it is a good thing that it hasn't been too cold this winter.
I found original drawings for the building and believe it to have been design for low pressure way back(as far as the line sizing at the loads).
I went this morning to see what was up after increasing the trap size at the common header. We went to 1.5" (from the 3/4")with a drain line just beyond the trap. Just as I suspected lots of very hot water(and steam I suppose due to the reduced pressure makes that hot water flash). Every time you see the peerless water level go out of the sight glass that drain line would start flowing water and steaming. Knowing what happens there I know that the feedwater temp gets its heat straight from that trap(or the majority of it) thus the problem with the feedwater pumps, not to mention the boilers are sucked down and they can't make up that fast.
Then I go smoke a cigarette or cigar as the case may be.
I like the idea of taking that fore riser straight to the header. Your tee and yeild sign explaination makes very good sense even though we went with a bigger near boiler header.
The trap at the peerless equalizer line was raised to 2-3" above normal water level(per peerless).
You mention an equalizer between the two boilers before the common header. Would you tie the "equalizer" lines together up high as possible?
Wondering now also if feeding the load with another 4" line from a different location might work well?
Sounds like you have been very close to a similar situation or figured this one out from afar.
Not sure what someone suggested about a drop header would fit the bill, even if we had room to make it.
More to follow and thanks for the advice.

Long Beach Ed

Christian Egli

Did you get my note about the Paul Syatem experiments I did. Thought it maight be interesting to you...

Long Beach Ed

Christian Egli_2

Crack down on the offending traps

I busted the word quota yesterday... so today I made a drawing. A picture is worth a thousand words.

Could you post a picture of the drip end of the Peerless near boiler piping? So it is true, the runout does not drip straight into the boiler it flows through a float trap and (it seems) gets pushed up to an overhead condensate return?

If so - and I still have trouble believing this set up - this is a layout you must delete immediately.

Drip this runout straight back down into the bottom of the boiler, no trap, no lift, no nothing to hinder the return of the unboiled water.

The stuff that comes out of the steam pipes so near to the boiler has no business flowing into the condensate return lines. Unboiled water should be treated differently than spent steam. Float and bucket traps show no consideration for heat and deliver lost of flashing steam into your condensate pumps. Dan Holohan says it best: "Steam in the returns eats pumps for breakfast".

After this simple delete job your system will be much kinder on your pumps and the Peerless surging will disappear.

Plan on doing the same for the common header, and on the Hurst. Do you have a picture of its runout drip end?

For an equalizer line from the header:

What is the respective altitude of each boiler's water line? And is the bottom of your common header above this altitude?

You would pipe this equalizer like a wet return and pipe it into each boiler at the Hartford loop level. Check numbers but ballpark size is 2 inch.

See end of chapter 5 in The Lost Art.

It is terrific you have vacuum breakers!

An additional steam main would be welcome. Check pipe carrying capacities in function of steam pressure.

Tell us how the new 1 1/2 inch trap replacing the 3/4 trap works. That should make a big difference.

I can tell you're getting happier by the minute. Are RAW and jgoyer the same?

Looking forward to hearing good news from you.

Christian Egli_2

Find the rubber ducky

For added intuition get a hold of a wet/dry shop vac and play with it while you're taking a bath...

Or use a shop vac to suck up water from puddles on the floor.

Or play with a pool vacuum cleaner.

You can suck air. You can suck air and water together. You can't suck water alone beyond a certain altitude. The slugs of water stall the air flow.

Condensate slugs stall your steam.

Watch the flexible hose wiggle like it just swallowed a live squirrel.

RAW

Not the case at the back!!!!

No, jgoyer and raw are not the same. Similar interest, but from different side of the fence. I don't believe jgoyer will be involved with this forum any longer, unless it is strickly personal.
The runout does go into the boiler. From 6" down to 2.5" after it makes the turn just like your dotted line drawing (straight through equalizer). the trap was something peerless recommended to be sure it wouldn't flood and is about 3" above water line. As for the hurst the runout drip end is actually the second picture above of the hurst and it is on the side. Nothing on the back but drains. The equalizer line(the one with the green wafer check that won't let it equalize and is soon to go away)is right on the side. That riser/equalizer arrangement has to be changed for sure, there is no way any water from the riser can make that left turn.
The water line of the hurst is above the peerless and the common header (bottom) is between the two.
The 1 1/2" trap at the common header worked so good this morning that the feedwater was 200+ and hard to cool back down. The hurst was off on manual low water and the peerless was cycling on/off by low water. That must go back to one of the boilers right away sounds like instead of piped in with the rest of the condensate from the loads? And without a trap sounds like what you are saying?

Christian Egli_2

Spilling the whole can of beans

The altitude on the common header is annoying.

Maybe focusing on producing dry steam before the common header, and leaving that one as is, might be easier. I don't think an extra 4 inch line to the Peerless will be useful anymore, keep the extra hole for an extra steam main.

So, that's the rest of the story on those runnout equalizers. Clearly we have to suspect those wafer check valves are springed so strongly that they do not open on the short stack of unboiled water.

Immediately gut both their innards (or remove the entire check valve if easier) and watch the improvements. We'll all be amazed.

You then should have a pretty good boiler runnout on the Peerless.

For the Hurst, you need to do the same thing. An easy idea might be to unbolt the section from the valve in waiting to the valve on the common header. Then to plug the end where the valve in waiting was and to insert a riser block on the common header valve.

Now, the valve in waiting would be at the same place, just a foot or two higher. From here add new pipe extending above the previous one that is now plugged; extend it to a point upstream of the equalizer tee take off, then add a down pointing elbow tying right into the old pipe. This set up should look like a good runnout.

Perhaps it is easier and best to build a new larger runnout.

At that point, juices coming out of the common header might not be significant anymore to worry about. But you'll still be pouring good unboiled water into your condensate system and jeopardizing your pumps.

You could still install a direct equalizer line from the common header to only the bottom of the Peerless (since it is lower). This would look like a traditional set up, the only risk would be when the Peerless is off and the Hurst is on, you'd be overfilling the Peerless.

But here is where I think you are lucky, you already have an overfill preventing float trap on the Peerless (set at 2-3 inch over normal level) that would kick in and send the extra water to the condensate system. This would only happen on Peerless off times; not too big a problem.

So my suggestions at this point would be to 1) remove wafer checks 2) improve runnout diversion on Hurst and 3) equalize common header untrapped into Peerless only, and keep the float trap on the Peerless equalizer for the special Peerless off times (I would move the now unneeded 11/2 trap to replace the current 3/4 one). In that order.

Check all these ideas out for yourself, but this should make you ecstatic.

Act quick, winter is soon over.

RAW

Not in Kansas anymore!

The wafer check on the peerless is already gone. We removed it when we fixed that header couple of weeks ago to match what peerless said. 2-4" risers into 6" and it takes off before equalizer with 6"(graded back to peerless) half way to common header then it goes back to 4"(original probably should be six into header).Thought even crossed my mind in restricting the fore riser on the peerless right at the boier(that's scary stuff).
Will remove the wafer check on the hurst when the new near header/equalizer scheme is nailed down.
Sounds good to run that common header straight to the peerless and get that outta condensate loop (what if it is more than peerless can stand--1 1/2" trap?).
We have got to slow down the water out of each though!
Maybe that extra feed to the load will help to keep it down in the boilers also.
Will get my welder back out and stop them dang insulators for a while. Won't be able to buy grease or filters in May and June if I don't stop them.
Thanks for the valuable insight!

Christian Egli_2

Stumped this time

Rather than a major repipe, maybe inserting a steam separator will do the trick.

At (I am guessing) 1,500,000 BTU/hr that's 1,500 lb/hr, the 4 inch S3 Spirax separator might be perfect. State supply has it as SG1856.

http://www.spiraxsarco.com/us/assets/uploads/PDFs/TIS sheets/TI-7-001-US.pdf

What's with the insulators the grease and the filters? What happens in May and June? Now you've got stumped.

Best regards

Jeff M

Spirax Sarco Video

Tony,

I contacted Spirax reps and the company directly and had no luck getting this video. If you know of a way I could get a copy, I'd appreciate it.

Jeff

RAW

Running out of $

The seperator doesn't sound like your style. I really think and you do to(?) that it is a piping problem. There have been enough band aids.
The comment about insulators was just a $$$ comment. The more I spend beating this horse the less will be available near year end to do the rest of the maintenance for the many other things that require $$$.
Good things coming soon. More to come. Believe we will save the common header trap for last. Gotta slow down that steam first!!!!!