self priming pump loosing prime

Assume you've already positively proved that the suction line on the recirc pump is free of leaks

Correct, free of leaks

Add a baffle between the two intakes.

OP,
How do you know it's losing prime if it runs constantly? Have you actually witnessed this or is this an assumption based off of a wacky pH reading being caused by an air pocket in the line as a result of losing prime?

Ideally it would run continuously. We have tee'd in a water line to allow us to prime it at startup, and after priming are able to get it running for about a day or so, but eventually it looses prime. I've never witnessed it personally, just reports from others who have.

OP,
Understood. Are both pumps at or above grade or submersible? I'm guessing above grade.

How did you check for leaks? Was the system vacuumed checked?

You mentioned the larger pump cycles regularly. I'm would assume there's a manual switch for this pump. Can you get eyes on the liquid level and watch what happens when the pump is kicked on and off? This should confirm the need for 3DDave's suggestion. There are some other things that could possibly be going on, but I would start by a confirmed vacuum check then an actual visual check.

It'll be interesting if it's just the bearings are failing and it has a thermal trip, cools down after 5 minutes, and then picks up like nothing was going on.

Perhaps get a camera on the thing and be prepared to fast forward through 12 hours of not much going on.

Van you draw this?
Photo?

What's a sump crock?

Height of pump above water level?
Depth of submerged of the inlet pipe

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

I was thinking the same 3DDave, not as likely as a small vacuum leak but a possibility. LittleInch, I had to google sump crock, looks like just prefabricated sump.

The larger discharge pump in on a frequency drive and ramps up when the high limit level sensor is made. Runs for a timed duration, and then ramps down. The suction pipe has been inspected for leaks multiple times in the past. New teflon tape and pipe dope is applied at all connection. The maintenance tech believes he has done all he can to assure the suction is leak free. That said, we don't deal with self priming pumps alot, and he didn't vacuum test it, so a leak is possible. I don't know the distance the suction pipe is in relation to the liquid line when the discharge pump finishes running.

What I'm calling a sump crock is really a stainless steel basin. It's sealed up, and a confined space, so will take some work to get a scope in there and make a drawing. The motor states it has thermal overload protection.

OP,

Let me give you one more possible scenario I have seen. Your large pump was likely sized for some controlling case. Something like an upset or abnormal condition, a vessel overfill, a 100-year rain rate...etc. The pump also needs to handle day to day influent, condensate, rainwater, washdowns...etc.
The larger discharge pump in on a frequency drive and ramps up when the high limit level sensor is made. Runs for a timed duration, and then ramps down. Can you confirm the logic you provided?
The VFD may have been installed due to the two different flow rates I mentioned, and the logic may be slightly different then you stated. I am wondering if the logic is, as long as the high-level senor is made, then pump will run at full speed and once the high level is no longer made, then the pump runs for a timed period, then ramps down. If this is the case, then your high-level switch may be sticking causing your sump to be sucked dry but once the pump kicks off, you have backflow refilling the sump, so if someone looks in it at a later time, it all looks good, and the only indication is that your small pump has lost prime. there's also the possibility that a maintenance person adjusted the timing, not realizing the purpose of the logic.
This goes back to actually getting eyeballs on it or a camera as 3DDave suggested.

Type of pump?

If it's self priming how does it lose prime?

What's the self priming mechanism?

Sounds like this mysterious crock runs out of water....

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

"The larger discharge pump in on a frequency drive and ramps up when the high limit level sensor is made. Runs for a timed duration, and then ramps down."

Why timed duration for the pump to stop? Why not a low level signal? The timed duration may be too long, level may have dropped below suction intake for the recirc pump at the time when this discharge pump stops ? Are the intakes for both these pumps at the same elevation in this SS basin ?

I was wrong, the high limit is really just a emergency limit which during normal operation is never used. There is a bubble tube, which is really doing the controlling. It turns the pump on around 30 inH20 and shuts it off around 20inH20. One thing I did note is the the discharge line for the recircualtion pump is a few inches away from the suction for that pump, so it's basically discharging water back down right atop the suction line, so perhaps that's leading to air entrainment. I think for a plan, I'm going to try and take the pumps apart and do a RCA on them, then get the crock drained completely and measure heights of both suction tubes, and bubble tube. I think I'm leaning on aiming the discharge for the recirculation line at the crock wall instead of at the suction line.

OP,
If no vacuum leaks
If pump is electrically and mechanically sound
If the controls are working properly
If you're sure the smaller pump is just running dry due to the larger pump.
Another consideration, Is this a new problem? What has changed? Did it start happing as it warmed up (assume northern hemisphere)? Are the lines traced? Is the influent warmer than it was?

You might want to check your suction height, fluid temp and vapor pressures. If things are marginal, you may be cavitating and losing prime, if exceeded you're flashing in your suction line.

OP, think your comment re the re-circ pump is your answer, it's entraining air into the suction of the pump. Overcome that situation and your problem will go away.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

Speak to your instrumentation engineer also, that bubbler tube could be unreliable for level measurement. Rarely if ever used. Find some other level device. Or it could be the bubbler gas that is finding its way into the recirc pump suction ?

Good point, bubble tube air going into the suction will be on the list of things to monitor.

Pump is also loosing prime because the nozzles on the pump are oriented to enable vapor locking of the pumping chamber. Exit nozzle should be at the high point of the pump casing so that any bubbles in the feed get to escape without collecting in the pumping chamber.