vacuum reduces when vac pump is isolated

[bill1983]

Whilst evacuating a small hermetic system, consisting of a 1 hp condensing unit with hermetic compressor, 20 feet of tubing, an evaporator with txv and accumulator, I noticed the following: At a pressure of 2000 microns, when isolating the vacuum pump, which was being used through a manifold set, the pressure on the system reduced by over 200 microns. Initially the vacuum pump was connected to the suction and liquid valves at the exit of the condensing unit, ie after the liquid receiver and before the suction stub on the compressor. The micron guage was connected to the receiver valve which had been closed to make it the furthest point from the pump. After noticing this phenomenon several times, I used the vac pump in several different ways to try to determine the cause; I was unable to. What was even more curious was that upon restarting the pump, the vacuum actually rose to its pre-isolation value. I have been unable to determine the cause of this. Can anybody out there shed light on this?

[Magoo]

If vacuum gauge is electronic, it is possibly that transducer is contaminated with oil,giving inconsistantancy, or else replace batteries as another possible cause.

[lowcool]

gday bill.quite likely moisture.was 2000 micron the deepest it would pull ?

[Magoo]

Hi Lowcool, Bill said it dropped 200 microns , not increased 200 microns. Me to as well confused, Bill may have developed a natural vacuum re-generation system... [ joke ] .
Normally only seen natural vacuums in the heads of clients, defying nature when it comes to paying for service, and they cannot remember you being there.

[bill1983]

hi magoo, vac guage is electronic but sensor was cleaned before use.
lowcool, eventually pulled 300 micron total vac so moisture was removed totally.
thanksfor showing interest guys, keep thinking, there must be an answer.
how can a vacuum reading reduce when the vacuum pump is isolated, or how can you carry on removing moisture when the process is stopped?
is it possible that moisture emerging from the oil is being reabsorbed? the system was under a stable temperature of 32c to prevent the water from freezing.

[Gary]

At a pressure of 2000 microns, when isolating the vacuum pump, which was being used through a manifold set, the pressure on the system reduced by over 200 microns.

So... the pressure in the system was 2000 microns. Then you closed the valve on your gauge manifold, which isolated the pump from the system. Makes sense so far.

But then the pressure in the system either went to 2200 microns

OR

the pressure in the system went to 1800 microns.

Which is it?

Or is it the pressure in the pump that changed?

[monkey spanners]

I have seen this too, only I had the electronic vac gauge mounted on the pump as I'm lazy....
Would be interesting to find out the cause.

Jon

[lowcool]

gday magoo i did read it twice and three times today,confusion still reigns supreme,i only took it in the context that if a vaccum reduces,it is a loss in this addled mind still to me if i had a 27" vac and it reduced to 26" bloody micron values but i do find it the easiest to work with.

another natural vaccum ive found is with consultants and architects,the old told ya so phrase comes to mind.
have a good weekend fella
bottoms up!

[bill1983]

thanks for joining y'all. to clarify. the vacuum reduced to 1800 microns from2000 when i isolated the pump. it took around 40 seconds to achieve this new level, which is why i thought that it must be due to moisture being absorbed. low cool you think you're confused. this was at 11 at night how do you think i felt even when i woke the next day?

[nike123]

I would say that possible reason for that behavior is puling vacuum from low and high side. Because of internal differences of restrictions in system, at point where you measuring vacuum there is lower vacuum ( higher absolute pressure) then in some other points in system.
When you close isolation valves, pressures in system equalize and you get higher vacuum (lower absolute pressure) then before equalization.

[Gary]

I would say that possible reason for that behavior is puling vacuum from low and high side. Because of internal differences of restrictions in system, at point where you measuring vacuum there is lower vacuum then in some other points in system.
When you close isolation valves, pressures in system equalize and you get lower reading then before equalization.

Or perhaps a little refrigerant and/or moisture bubbling up out of the compressor oil. The oil absorbs both and will reluctantly and slowly release them under deep vacuum.

[nike123]

Or perhaps a little refrigerant and/or moisture bubbling up out of the compressor oil. The oil absorbs both and will reluctantly and slowly release them under deep vacuum.

I think that will rise pressure on isolation and not lower it.

[Gary]

I think that will rise pressure on isolation and not lower it.

Exactly so. Seems I am still confused... lol

[nike123]

Exactly so. Seems I am still confused... lol

I edited little my post for some clarification. Maybe that helps.

[Yuri B.]

Maybe it has smth to do with the working of the electronics in the gauge itself (not with the refr system). Its "kink" so to say.

[Yuri B.]

Or rather of a particular vac gauge.

[littleyapper]

weird one all right ... my only suggestion is that the suction accum might have had a better vac and when you stopped the pump the negative pressure in that brought down the rest of the system.... or tranducer acting up... try a good old fashioned vac/torr guage and see if it's the same ... what make of vac guage are u using?

[Peter_1]

Whilst evacuating a small hermetic system, consisting of a 1 hp condensing unit with hermetic compressor, 20 feet of tubing, an evaporator with txv and accumulator, I noticed the following: At a pressure of 2000 microns, when isolating the vacuum pump, which was being used through a manifold set, the pressure on the system reduced by over 200 microns. Initially the vacuum pump was connected to the suction and liquid valves at the exit of the condensing unit, ie after the liquid receiver and before the suction stub on the compressor. The micron guage was connected to the receiver valve which had been closed to make it the furthest point from the pump. After noticing this phenomenon several times, I used the vac pump in several different ways to try to determine the cause; I was unable to. What was even more curious was that upon restarting the pump, the vacuum actually rose to its pre-isolation value. I have been unable to determine the cause of this. Can anybody out there shed light on this?

Just joining this thread and I'm happy this is brought up here: I'm an assessor for the F-gas exams in Belgium and I saw now +/- 400 candidates passing their exams from which most followed the pre-exam courses.
In this course, they all can do 2 complete recover cycles. So I saw at least 1000 vacuumcycles the last 6 months.
Well, I each time explain this phenomena and I promise each time a crate of Stella (finest Belgium beer ) for the one who can explain this.
When they can't reach a certain vacuum, I advice them to isolate the pump and watch for 30 seconds. Then teh vacuum drops and rather fast. Once the vacuum has dropped with isolated pump and stays on a certain level, switch on the pump again and continue pulling deeper, starting from this new 'strange' deeper level.
This phenomena only happens with electronic meters (I tried 4 different brands of meters, Refco, Yellow Jacket, Inficon, ITE) but not with a precise absolute analog one. We use a very precise analog one to control all that electronic stuff.

[Peter_1]

BTW, we placed the sensor already everywhere on the system with the same strange result.

[nike123]

BTW, we placed the sensor already everywhere on the system with the same strange result.

It seems that we need to analyze principle at which these electronic vacuum gauges working.

[Peter_1]

Even this isn't true Nike: the Inficon works with a Wheatstone bridge, other uses a Piranni gauge as sensing element, others a piezoresistive sensing element,...

[nike123]

Piezoresistive is also resistor in Wheatstone bridge.

I am confused why pressure sensing piezo elements acting same as thermal conductivity sensing elements (pirani and thermistor).
Are you sure that both types are acting same?

Anyway, I am sure that there is no same vacuum level at all points in system during vacuum pump operation (otherwise there is no movement of contaminants in system), and that some pressure equalization is present when pump is isolated.
Since, in this case, vacuum pump has higher vacuum level than furthest point in system (where vacuum gauge is connected), it is normal for me that, at moment after pump is isolated, all points which are distanced from vacuum pump are at lower vacuum level (higher absolute pressure) than those closer to vacuum pump. And, after equalization of pressures in system, it is normal that at these distanced points vacuum level rise (absolute pressure fall). It could be checked by positioning two vacuum gauges (one at furthest point from vacuum pump and one at closest point to isolation valve. Then we could compare rise and fall of pressures during equalization.

[nike123]

When they can't reach a certain vacuum, I advice them to isolate the pump and watch for 30 seconds. Then teh vacuum drops and rather fast. Once the vacuum has dropped with isolated pump and stays on a certain level, switch on the pump again and continue pulling deeper, starting from this new 'strange' deeper level.
This phenomena only happens with electronic meters (I tried 4 different brands of meters, Refco, Yellow Jacket, Inficon, ITE) but not with a precise absolute analog one. We use a very precise analog one to control all that electronic stuff.

I did not registered/noticed this phenomena, I will check this on my next few vacuum jobs. But I think that could also be explained by pressure equalization in system. Since pressure is equalized, there is no possibility for rise in absolute pressure at vacuum gauge point if you first turn on pump, than pump reach same or higher vacuum level like in system and after that you open isolation valve. Absolute pressure could only fall down further, but if you again isolate pump, it will fall again during new equalization.

How much is precise your analogue one?

[bill1983]

i used yellow jacket vacuum guage for the procedure.
if the vacuum equalises when isolating the vacuum pump due to restrictions in the system, then would this cause an equal rise in vacuum (increased micron value)? if it is due to moisture/contaminents being reabsorbed then surely these would be released upon restarting the pump from the lower micron value, returning to the original pre isolation value. the time taken for both reactions is roughly the same.
peter, what level of vacuum were your students pulling when you saw the same phenomena? were the level rises and falls the same as mine or different at different levels of vacuum?

[lowcool]

what was the name of the mercury job that was around during my apprentice days that got canned due to being not so enviroment friendly?

[Yuri B.]

You draw the air at point A, vac gauge is set at point B. The space at the point A near the pump is more rarified than at the point B so the air flows from B to A.. You isolate the pump, the flow of air from the point B to the pint A continue to flow yet for some 40 sec due to inertia rarifying the space at the point B yet for some 40 sec. The vac gauge is showing it to you.

[lowcool]

maybe the pump is to big for the application

[nike123]

Here is some PDF about book " Vacuum for Service Engineers " from nice guy named Rat on HVAC-PROTech Forums.
Credits to him.

[Peter_1]

..
How much is precise your analogue one?

I don't know for the moment but in my opinion, much more precise than ALL that electronic stuff.

How I know this?

We once coupled all the vacuummeters we had on that moment (+/- 12 or 13 pieces) on a 7/8 tube with shraeders soldered around it and we vacuumed the 7/8 with a 3/8.

You can't believe the difference we measured and all those meters were only months old. We know now which are prone to failure, which ones has the best repeatability (very important),

Same comparable practical experience with electronic sniffers.

[Peter_1]

....
peter, what level of vacuum were your students pulling when you saw the same phenomena? were the level rises and falls the same as mine or different at different levels of vacuum?

We have to pull legally (EN378) lower than 290 Pa but we advise to lower than 100 Pa because you can vacuum the small systems for the exams easily with a tinny leak easily to 250 Pa. With a leak, you never can go lower than 100 to 150 Pa. So this phenomena happens lower than 150 Pa.
Stopping for example at 120 Pa, isolating the pump will initiate a drop of +/- 15 to 20 Pa and then stops.
Level stops on say 100 Pa, you re-engage the pump and you then can pump down further.

What I noticed is that this phenomena is the most
strong as soon as you notice the decrease of pressure is slowing down or even stopping.
You then can trigger a further pressure decrease with performing this technique (but not always)

[Peter_1]

Here is some PDF about book " Vacuum for Service Engineers " from nice guy named Rat on HVAC-PROTech Forums.
Credits to him.

This link seems not working and I was interested in it

Not Found

The requested URL /forums/Vacuum for Service Engineers was not found on this server.
Apache Server at www.refrigeration-engineer.com Port 80

[gwm121]

I have had this..several times.
1. Manifold Gauges connected to low side torr guage on manifold. Vac pump to guages. 3 way valves off as system not yet commissioned.
2. vacuum drops to say .7 torr and wont go lower.
3. shut guage valve...and amazingly the vacum slowly goes down to .5 torr.

[Peter_1]

We also use a technique I never saw before while performing leak checks on a pressurized system.
We can see within minutes (sometimes 30 seconds) on our small systems if there's a leak.

[gwm121]

message to peter 1
May I ask if ur an fgas examiner where one could get study notes and lots sample questions to prepare for the 2079 test please.

[nike123]

This link seems not working and I was interested in it

Not Found

The requested URL /forums/Vacuum for Service Engineers was not found on this server.
Apache Server at www.refrigeration-engineer.com Port 80

It is fixed now!

[Peter_1]

I have the Belgium questions but those are completely different than those used in the UK.
And they're in Dutch and soon in French.

There was a poster some weeks ago - but this post seems erased - who tried to sell the questions but he then promised to post them here. We're still waiting.

[nike123]

I don't know for the moment but in my opinion, much more precise than ALL that electronic stuff.

How I know this?

We once coupled all the vacuummeters we had on that moment (+/- 12 or 13 pieces) on a 7/8 tube with shraeders soldered around it and we vacuumed the 7/8 with a 3/8.

You can't believe the difference we measured and all those meters were only months old. We know now which are prone to failure, which ones has the best repeatability (very important),

Same comparable practical experience with electronic sniffers.

I measuring my vacuum with Refco VG64 and I wondered many times if it is acting on me or it is OK. With some experience I found that, if sensor is cleaned before use, it shows me always there is leak in system or at gauge and/hoses (if leak is present and I skipped leak test for some reason).
With +-10% (of reading) accuracy of this instrument I taught that these drops in reading after isolation are result of that and system pressure equalization (and maybe time needed for change of temperature of sensor). Now, after you presented that that phenomena is visible on all type of instruments except on analogue one, I am confused.
Usually, digital instruments reveal more than analogue if sampling rate is good.

[nike123]

You can't believe the difference we measured and all those meters were only months old. We know now which are prone to failure, which ones has the best repeatability (very important),

Could you name best ones, and what you could say about performance of my Refco VG64 (if it is compared).

[MechAcc]

what was the name of the mercury job that was around during my apprentice days that got canned due to being not so enviroment friendly?

McLeod Gauge

[MechAcc]

If you are using valve core removing tools to isolate the system it may be there is a small amount of seepage around the valve shaft. When you rotate towards close it may be sealing off.

Next time you vacuum seal around the shaft with some vacuum grease and see if this phenomonon stops.

[nike123]

Here is some of discussion from guys on HVAC-TECHPro forum where I linked this thread.
I cannot link it because board require to be registerd and loged in in order to see posts.

Mark Beiser

Without getting into the exact mechanics of how the vacuum sensors we use work, mostly because I don't like talking out my ass , it is because the sensors don't measure pressure directly.
The Thermocouple and thermistor sensors in the electronic vacuum gauges we use measure molecular activity. There is a direct relationship between molecular activity and pressure, so the electronics are able to convert the measured molecular activity into a pressure reading, usually in microns, but some gauges can give other units of pressure.

Now in the specific instance in the post you linked, when the pump is running, gas molecules in the system are exiting the system at the ports the vacuum hoses are connected to.
As gas molecules are removed from the fixed volume, the remaining molecules move to take up the void left by the exiting molecules.
When the system is valved off from the pump, no more gas molecules are being removed from the system, so the remaining gas molecules in the system stop moving to fill the void.
The molecules are still bouncing around, but they are no longer moving in the direction of the exit points, so there is less movement overall.

The net effect of this is that very often the vacuum sensor will detect more molecular movement, and thus a higher pressure, when the system is running than when it is off.

In my personal experience the effect can often be exaggerated by where and how the vacuum sensor is connected to the system.
I used to see it almost every time I pulled a vacuum back when I was using a sensor that connected inline with the vacuum hoses, where the actual physical sensor was less than an inch from the gases flowing in the hose.
My current vacuum sensor connects to the system with a brass tube and ball valve setup, so the actual sensor is 6 or 7 inches from the service port is connected to, so I don't usually get the little drop, or as pronounced of a drop, when I valve the pump off from the system.

He had the vacuum sensor connected to the king valve on the receiver, so essentially it was connected to the main exit point of a component that has a rather large number of gas molecules leaving it, which would affect the reading of the vacuum gauge.
If he had the vacuum pump connected to the SL service valve and the king valve on the receiver, and the micron gauge connected to the LL service valve, he may not have gotten a noticeable drop in the the vacuum reading when he valved off the pump.

That is my simitechnical answer, I'm to lazy to type out the really technical answer, and the simple answer doesn't fit the audience.

Hvacrmedic

That's an interesting theory. I don't get it though. The "activity" that you're speaking of is just the kinetic motions of the molecules, which translates directly to temperature, and for a given density, it also translates directly to pressure. A reduction in KE per molecule for an ideal gas is equivalent to a drop in temperature. For ideal gasses the two are actually one and the same with a correction factor applied. So the only way to reduce the pressure on the vacuum sensor surfaces, when dealing with a rarified (or very nearly ideal) gas, is to either reduce either the temperature of the gas or reduce its density or both. Stopping the flow into the vacuum pump will not accomplish either of those things in itself.

Now here's an alternate explanation, take it for whatever it's worth. (BTW, I've seen this behavior many many times, but like most here, I've never seen a solution to the problem offered.)

While pulling a vacuum the oil sump in the compressor is cooled as are any other collections of liquid, be they either oil or water. The sump is at a temperature lower than that of the tubing and connecting hoses after the vacuum pump has been allowed to run and pull a vacuum on the system. While the pump is running there are more molecules leaving the compressor sump than entering it per unit time. It is in fact this "evaporation" that reduces the oil temperature in the sump. This effect is more pronounced when refrigerant is suspended in the oil. As the refrigerant boils out of the oil, the oil is cooled. This stream of molecules leaving the oil sump generally goes directly out and along the suction line back toward and past the suction service valve. Some of those molecules bounce out the service port opening and on into the vacuum pump. Though there are molecules bouncing along the suction line in the opposite direction (toward the compressor) too, the flow rate in that direction is smaller. This is uderstood in the fact that if these flows were equal an opposite, there would be no net evaporation from the sump oil.

When the vacuum pump is valved off, the flow of molecules from the sump to the vacuum pump ceases, and the gas in the system begins to seek an equilibrium state. The flows in both directions are still unequal, but are reversed in relative magnitudes. The colder molecules leaving the oil have a lower dispersion rate because they are slower. Now, instead of evaporation from the sump oil taking place, condensation replaces it. The hotter molecules dive into the oil at a greater rate than the colder molecules escape. It's the reverse of evaporation. The density of the free gas in the system decreases, and thus the pressure also decreases. Equilibrium is reached when the density of the gas in the system reduces sufficiently to reduce the rate of flow into the sump oil to a value that exactly equals the rate of flow out of the sump oil. Because the gas is rarified, this equilibrium process occurs rather quickly.

Comments?

Hvacrmedic

Behappy wrote:

Mark Beiser wrote:
So under that theory, what causes it when you get a drop in the micron reading immediate after valving off a system when you were only pulling a vacuum on new refrigerant lines and evap coil that have never had refrigerant or refrigerant oil in them?

The equilibrium process of the system. You can only "pull" a vacuum from one end of the system. Just like kids, it takes a couple of minuets for those peskie molecules to stop jumping around.

That's a different thing that what we were discussing. That equilization from the far end of the system to the near end generally leads to a rise in micron reading. That happens because the pressure near the pressure taps where the pump is connected is lower than the pressure at the far end of the system. But we were discussing the phenomenon of a drop in the micron reading after the pump is valved off. When I've seen this I've noted that a rise occurs briefly before the reading drops to a value lower than it was before the pump was valved off.

Another tech was helping me last week, maybe the week before last, on an a leak repair job. The leak was on the condenser coil header, so I had to recover the entire charge to do the repair. When I pulled to 500 microns and valved off the pump he said "the reading is going up". I said, "just wait a minute and it'll drop back again even lower than it was". It did exactly as I had predicted. The other tech just scratched his head and asked me how that could happen. I didn't have an answer for him. I just said "pfm, that's what it is." I suspected as a poster in the thread on the other forum did, that the oil was absorbing some of the vapor after the pump was valved off, but I couldn't figure out why it would do that. To be honest with you, I didn't really have a theory until reading this thread and realizing that it wasn't just me who was seeing this. I really hadn't given it any thought since that day a week or two ago when I last observed it. I threw the above theory together just before posting it to this thread. Like I said, I won't swear that this is what is happening, but after looking it back over I think it's probably worth considering.

Here's a question. Suppose you have a refrigerant cyclinder half full of liquid refrigerant. Now you apply heat to the upper portion of the cyclinder so as to heat the vapor above the liquid, but not the liquid itself. What happens to the mass of the vapor above the liquid, does it decrease, or does it increase? I would think that the vapor density would decrease, which means that some of it would be forced to condense into the liquid below. This is basically the same principle that I gave, except the liquid is oil rather than refrigerant. When the vacuum pump is valved off, there will be a migration of warmer vapor from the suction line into the compressor can because it is no longer being impeded by the counterflow from the refrigerant that is evaporating out of the oil and leaving the compressor. The warmer vapor above the colder oil will result in a net migration of vapor into the oil with the result of a reduction in vapor density in the system, and thus a lowering of the pressure in the system.

Mark Beiser

hvacrmedic wrote:
I can't think of anything specific that we might check or do to try to rule out one cause or another. I doubt they're going to award a Nobel or anything to the first guy that pins it down though.

Well, they gave Nobel prizes to Carter, Arafat, and Gore, so you never know.

The reality is, it is probably a combination of both, and maybe more things tossed in that we didn't cover. Or maybe we are both wrong, and are over thinking it way to much.