| Pages:
1
2 |
dann2
International Hazard
Posts: 1523
Registered: 31-1-2007
Member Is Offline
Mood: No Mood
|
|
Hello,
This I believe,
In the UK (and Europe?) most if not all domestic refrigerator pumps are piston pumps unlike the USA
where they are all rotary pumps (stater + impellor).
The USA pumps do a better job at pulling a vacuum, thats not to say that the piston pumps will not pull a vacuum. The vacuum will not be as good.
Dann2
[Edited on 24-8-2010 by dann2]
|
|
|
peach
Bon Vivant
Posts: 1428
Registered: 14-11-2008
Member Is Offline
Mood: No Mood
|
|
Have you cracked them open to have a check of that?
From the schematics of pumps I'd seen, I'd been expecting to see the piston vibrating inside a solenoid, yet it was on the end of a camshaft stuck to
the top of the rotor.
Quite a lot of them are coming from the same people as well, like HSH, who all seem to use the camshaft / piston layout.
I would be interested to hear some numbers on them there'z USA pumps, as I have checked multiple UK ones with both a mechanical and digital filament
gauge, so I'm very sure of the numbers the ones around here can do.
|
|
|
Contrabasso
Hazard to Others
Posts: 277
Registered: 2-4-2008
Member Is Offline
Mood: No Mood
|
|
Whatever the nature of the pump I bought from the fridge breakers, I'm very satisfied with the vacuum it pulls for reduced pressure distillation.
|
|
|
peach
Bon Vivant
Posts: 1428
Registered: 14-11-2008
Member Is Offline
Mood: No Mood
|
|
Glad you've got one you like contra. Especially for £25, versus £2.5k
Some of them do this odd trick, particularly once they've been left on for a few hours under vacuum, where the pressure goes into a cycle of floating
up and down a bit every few minutes. The first few times you try fractional distillation with it, keep an eye on the thermometer and look for any
signs of the solution bumping, or trying to (e.g. bubbles disappear a little, then return with vigor).
|
|
|
aonomus
Hazard to Others
Posts: 361
Registered: 18-10-2009
Location: Toronto, Canada
Member Is Offline
Mood: Refluxing
|
|
I just got ahold of a LG QA series rotary compressor, in the only online documentation I could find, it detailed a few issues.
1. Duty cycle: recommended 6 shutdowns/hour, each lasting 3 minutes/shutdown
2. Exposure to atmosphere: minimize it, the manual sounded like its lubricated with n-BuLi, and the universe would implode if you didn't keep it under
N2.
Also:
- Has anyone found it necessary to add a run capacitor to their compressors? I would suspect the load is lesser on a vacuum pump vs a compressor pump
- What about duty cycle? Without cooling gas (ie: refrigerant), won't the motor windings get much hotter? I tested the pump for a few minutes, outer
casing hit 80degC.
- I managed to achieve <29" Hg vacuum with poor sealing on the vacuum fittings
[Edited on 27-8-2010 by aonomus]
[Edited on 27-8-2010 by aonomus]
|
|
|
peach
Bon Vivant
Posts: 1428
Registered: 14-11-2008
Member Is Offline
Mood: No Mood
|
|
Awwww gawd, not "hg.... [opens google]
Right, that sounds about right. I've usually seen them around that, 30mBar. One of them I could never get to below 100mBar, but the others are closer
to 30.
When I was taking them to bit recently, I tried emptying the oil out and squirting some BOC Edward Ultragrade 19 in instead, and managed a constant
15mBar, but the pump was running warmer with the thicker oil.
I supposed one could try even thicker oil and simply fan cooling the pump. Given how well they're sealed (welded), provided care was taken around the
electrics, the entire thing could go in a water bath. Never tried it however, and seems like quite a lot of effort for no real difference in it's
performance with regards to chemistry. It'll never get low enough for analytical gear, and it's not a massive change in temperatures for distillation
etc.
At their normal 30 odd mBar, that's 97% of the atmosphere left. Changing the oil squeezed another 2% out. About 1/50th better.
About them getting hot, I doubt they're designed to be left hot for so long, but they can manage 12h runs without burning out (with their thermal cut
off cut out), and keep doing it. I've not had one go from overheating. If you live somewhere where it's 30C everyday, lucky you, but the pump may go
quicker. 20C is about the warmest it gets here most of the time.
But, if a lot of volatile solvent goes through, a hundred ml's or so of something like DCM, it seems to rinse the oil cushion off the piston and it
temporarily seizes.
Some possible solutions;
a.) Use some ice in the water for the condenser to stop the solvent first
b.) Leave the pump running for a while, open to the atmosphere, when you're finished, to replace the oil cushion
c.) Leave it upside down for a while to encourage the oil back into the piston
d.) Try a bigger starting capacitor, to overcome the higher resistance of the oil less piston (that may burn out the windings, but it's also likely
the windings are a lot heavier than they need to be, for duration, and the capacitor already on there is the smallest they can get away with for cost
reasons)
e.) Cut a hole in the top and manually flick the camshaft round to recoat the piston
The only time I've had one die from solvent abuse was after being incredibly lazy and removing a lot of DCM without a condenser, then immediately
turning the pump off.
|
|
|
aonomus
Hazard to Others
Posts: 361
Registered: 18-10-2009
Location: Toronto, Canada
Member Is Offline
Mood: Refluxing
|
|
The compressor is a taller cylinder, not a pancake type. Rotary vane, for R22, new old stock. The 'manual' states a cutout at around 120degC, and I'm
sure that it would reach that point without much trouble. I won't be doing too many reduced pressure distillations, or pulling too much solvent
through the pump. My main concern would be the temperature, and the flash point of solvents (hot surface in contact with solvent vapor at the exit
port means bad things.
I'm thinking of making a bunch of aluminum plates and clamping them on as makeshift heatsinks, anything to help wick away more heat from the casing
including fan-cooling. I mainly wanted more confirmation that running it so hot wouldn't cause immediate seizing or failure. I did have the pump trip
the power bar and had to move a little bit upstream to get power straight from the outlet. Batteries on my clamp meter were dead so I couldn't measure
the current usage either.
Finally, it would appear that the start capacitor is internal and integral, though 3 terminals exist for start, run, and common, and no run capacitor
exists. I'll have to pick one up next time I'm at the surplus place, but the motor runs fine without it, and maybe find a better vacuum gauge.
I'll probably bolt the compressor down onto a piece of plywood with a metal sheet to help reflect heat away from the wood, and add on a cooling fan,
etc.
Any solutions for a makeshift oil-coalescing trap? I was originally thinking one made of steel wool or another high-surface area material for the hot
mist to condense and drip down onto, though that might never allow droplets to merge and settle for recovery.
|
|
|
peach
Bon Vivant
Posts: 1428
Registered: 14-11-2008
Member Is Offline
Mood: No Mood
|
|
If it cuts out at 120C, it's going to be fine.
Flashpoint is the point at which a solvent forms a flammable mixture with the air and will ignite when a spark or flame is around. Autoignition is the
point at which it'll spontaneously burst into flames when mixed with air.
Don't over complicate and end up out of time, like a fly hit with the wasp killer, start by silicon / epoxy fixing a big, cheap computer fan to the
top.
It tripped the breaker? If it was going to trip it, I'd expect it to go when you flicked it on. That may be due to a lack of starting capacitor or it
may just need running in to lower the starting friction. If it went half an hour after switching it on, that would be a bit odd to me.
If it's like the ones here in Europe, inside the black shell is full of oil, gases under odd pressures and gets hot. The capacitors are always on the
outside, or not there.
Oil leakage won't be much of an issue, if one at all. First of all, if it still as it on there, leave the looped bit of copper on the outlet side. The
air coming out is barely warm on the ones I have. The steel wool idea should work. For the ones I've used, I've seen about one or two drops come out
over 12h, and no visible mist whatsoever. With your's being a rotary, I'm not sure how those are designed inside for fridges. If it has an oil pump
spraying back into the rotor, mist may be a significant issue (report back on that, I'd like to know).
[Edited on 27-8-2010 by peach]
|
|
|
aonomus
Hazard to Others
Posts: 361
Registered: 18-10-2009
Location: Toronto, Canada
Member Is Offline
Mood: Refluxing
|
|
Sorry for the 2-3am vaugeness/verbal ambiguity. The compressor tripped the thermal breaker on the *power bar*, not the upstream 15A panel breaker. I'm
a little unsure as to whether the thermal cutout is integral with the unit, or if its a external clamp on device. I have some big microwave oven fans
I can mount to really put some air onto this thing (at the cost of noise), we shall see just how effective that is (I'm going to tape on a
thermocouple and take readings without and with the fan(s)).
As for the oil mist, when the pump shell is cold, no mist emerges, but once it warms up, like a real rotary pump, it really starts spewing the oil as
a mist. The pump wouldn't get that hot in normal operation anyway, so the real question is whether or not I can capture the oil loss on the output
side and occasionally return it to the input. As far as I know from the drawings I've seen, LG has an inverted rotary design (rotary on bottom, motor
above), meaning the rotary compartment is submerged under oil at all times, and any oil that is lost to the output tubing draws in an equal amount
from the internal reservoir.
I really want to make this work as there is a whole pile of these compressors at the surplus store, stringing together 1+2 for a 2 stage system would
be an interesting experiment to see what ultimate vacuum I can pull. They are for R22, so they will likely never see use in refrigeration ever again.
|
|
|
watson.fawkes
International Hazard
Posts: 2793
Registered: 16-8-2008
Member Is Offline
Mood: No Mood
|
|
Also on that site is an article called Vacuum on the Cheap, which talks about the switch, by and large, from rotary to piston pumps for refrigeration. Pretty much everything in that
short article addresses something that's been raised here.
Changing focus ... Quote: Originally posted by peach  | | At their normal 30 odd mBar, that's 97% of the atmosphere left. Changing the oil squeezed another 2% out. About 1/50th better. |
That's right if you're focused on the pressure against evaporation, as is the case for solvent stripping. If you are trying to
evacuate a chamber, it's three times better. I mention this because the articles in The Bell Jar are focused around evacuation, not lowering ambient
pressure.
|
|
|
peach
Bon Vivant
Posts: 1428
Registered: 14-11-2008
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by watson.fawkes | | That's right if you're focused on the pressure against evaporation, as is the case for solvent stripping. If you are trying to evacuate a chamber,
it's three times better. I mention this because the articles in The Bell Jar are focused around evacuation, not lowering ambient pressure.
|
Because the evaporation in a flask is only concerned about one dimension, up?
Interesting page, I think I may have seen it before. I have also noticed a lower pressure with two of the piston kind in series, somewhere around 15
to 10mBar.
[Edited on 27-8-2010 by peach]
|
|
|
watson.fawkes
International Hazard
Posts: 2793
Registered: 16-8-2008
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by watson.fawkes | | That's right if you're focused on the pressure against evaporation, as is the case for solvent stripping. If you are trying to evacuate a chamber,
it's three times better. I mention this because the articles in The Bell Jar are focused around evacuation, not lowering ambient pressure.
|
More or less, that's right. In
particular, it's the interaction at the liquid (solvent) -- gas (atmosphere) boundary. That's a two-dimensional boundary, so the orthogonal complement
is one dimension. You can call it "up" if you want. The important thing is that the pressure near the boundary changes the rate at which solvent
evaporates. The pressure near the boundary doesn't change the partial pressure of the solvent at the boundary. What it does change is the rate at
which gas molecule near the boundary diffuse away from the boundary, which enables new liquid molecules to hop the boundary and go from liquid to gas
phase.
|
|
|
aonomus
Hazard to Others
Posts: 361
Registered: 18-10-2009
Location: Toronto, Canada
Member Is Offline
Mood: Refluxing
|
|
I found this interesting video on youtube with a high speed camera shot of the vane/rotor in a rotary compressor. This might explain why we see
fluctuations in pressure that make it unusable for vacuum distillations.
Edit: the link would sure be useful... derp.
http://www.youtube.com/watch?v=w_hpMZY7HbM
[Edited on 27-8-2010 by aonomus]
|
|
|
peach
Bon Vivant
Posts: 1428
Registered: 14-11-2008
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by watson.fawkes | | More or less, that's right. In particular, it's the interaction at the liquid (solvent) -- gas (atmosphere) boundary. That's a two-dimensional
boundary, so the orthogonal complement is one dimension. You can call it "up" if you want. The important thing is that the pressure near the boundary
changes the rate at which solvent evaporates. The pressure near the boundary doesn't change the partial pressure of the solvent at the boundary. What
it does change is the rate at which gas molecule near the boundary diffuse away from the boundary, which enables new liquid molecules to hop the
boundary and go from liquid to gas phase. |
You're a fountain of knowledge Watson, indubitably.
| Quote: Originally posted by aonomus | I found this interesting video on youtube with a high speed camera shot of the vane/rotor in a rotary compressor. This might explain why we see
fluctuations in pressure that make it unusable for vacuum distillations.
Edit: the link would sure be useful... derp.
http://www.youtube.com/watch?v=w_hpMZY7HbM
[Edited on 27-8-2010 by aonomus] |
That's a neat video.
I've also seen pressure fluctuations on the piston kind. I don't know how they float around on those rotary fridge pumps but, on the piston kind, it's
a very slow oscillation, minutes long; even though the piston is moving a lot quicker than that.
It would help if my vacuum gauge would work as a datalogger. It's brand new but, as normal, I had it in bits to look at it yesterday. If I can find a
suitable point to remove the readings from, I may be able to get it datalogging and could plot graphs of the pressure. My dream is to have a cheap
datalogger for the still head, maybe another thermometer, pH, pressure and perhaps flow rates, all in one box and one application that will work on
old computers for the lab.
These fridge rotaries are essentially backwards to the way I've seen dual stage rotaries designed, where the vanes are usually inside the rotor and
pushed out against the walls of the chamber. In BOC Edwards', there's squishy springs between the vanes. In Alcatel pumps (the one's with the orange
heads), I've seen them using nothing other than the centrifugal force of the rotor spinning to flick them out.
He's got a video on there of pumping water through them. Wonder how long they last doing that and what pressures you can get.
[Edited on 28-8-2010 by peach]
|
|
|
aonomus
Hazard to Others
Posts: 361
Registered: 18-10-2009
Location: Toronto, Canada
Member Is Offline
Mood: Refluxing
|
|
Wouldn't be too hard to set up a datalogger with a computer, it just wouldn't be intrinsically safe. Arduino + AD595 + K type thermocouple in the
still head, and some pressure transducer by Omega by the vacuum port on the receiver adapter. Too bad the pressure transducer is a $200+ part.
|
|
|
densest
Hazard to Others
Posts: 359
Registered: 1-10-2005
Location: in the lehr
Member Is Offline
Mood: slowly warming to strain point
|
|
It's pretty easy to isolate transducers especially after signal conditioning. Analog Devices & Texas Instruments/Burr-Brown sell inexpensive
isolation amplifiers with pretty good accuracy.
Arduinos can work well, though you still have to isolate the RS232 or USB or whatever.
Pressure transducers don't have to cost $200 if all you need is 0 to -1 atm with 3 digit precision. If you need ion gauges or such, they're more
costly. A Dwyer 4-20 mA vacuum transducer costs $114 from Grainger, and that's retail. *waves hands* automobiles need vacuum sensors for the control
computers. I'm sure they're as inexpensive (cheep cheep cheep) as possible!
If you need ion gauge sensitivity that's another matter.
|
|
|
peach
Bon Vivant
Posts: 1428
Registered: 14-11-2008
Member Is Offline
Mood: No Mood
|
|
Isolation isn't hard at all, as densest says.
I'm currently talking to one of the manufacturers of a vacuum sensor about the price of them now. And another guy on sparkfun who's interested in
building a little, cheap interface box for them.
My vacuum gauge has a hot filament in it, it's digital, displays about 9 different units, reads to 0.001mBar and cost £85, so there's a cheaper
source than Edwards out there.
I've seen the Arduino's on sparkfun, but never had a go with one. I'll have to investigate.
|
|
|
watson.fawkes
International Hazard
Posts: 2793
Registered: 16-8-2008
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by densest | | It's pretty easy to isolate transducers especially after signal conditioning. Analog Devices & Texas Instruments/Burr-Brown sell inexpensive
isolation amplifiers with pretty good accuracy. |
These days, it's cheap to isolate after digitization. AD 595
to condition the thermocouple itself, a small AVR to digitize it, and then an optocoupler pair on the I2C bus.
|
|
|
aonomus
Hazard to Others
Posts: 361
Registered: 18-10-2009
Location: Toronto, Canada
Member Is Offline
Mood: Refluxing
|
|
Did a little more searching about automotive vacuum sensors. Manifold absolute pressure (MAP) sensors are the ones to get. They give a analog voltage
output proportional to the absolute pressure on the port, both above and below atmospheric. A ADC sensitive down to mV should be fine for using as a
pressure meter.... thanks for the tip densest. I'll start looking for cheap MAP sensors.
Note: don't confuse MAP with MAF sensor, some cars use one or the other
|
|
|
densest
Hazard to Others
Posts: 359
Registered: 1-10-2005
Location: in the lehr
Member Is Offline
Mood: slowly warming to strain point
|
|
Hmmmm... From www.mouser.com, Freescale Semiconductor MPX2102AS is $7 (plus shipping). Output is analog, 40mV from vacuum to nominal 14.7PSI. An inexpensive
zero-drift op amp will put that up to 0-2.5 or whatever the ADC wants. That particular one isn't available until Oct. 20, though. Other similar ones
should work as well.
Auto sensors are fairly expensive new (super-wide temp range, etc.) but should be practically free from junk cars - once you know what they look like!
[Edited on 29-8-2010 by densest]
|
|
|
peach
Bon Vivant
Posts: 1428
Registered: 14-11-2008
Member Is Offline
Mood: No Mood
|
|
Here's some more, new in the box...
I'm not too sure about how accurate these are going to be once they hit fridge pump levels and below.
I'm also absolutely sure I'm no smarty pants when it comes to modding cars, but the vacuum lines for turbos and things typically aren't like the thick
ones found in labs; had a quick google but couldn't find a torr / mbar for the manifolds themselves.
The vacumm to ... rating on these likely means, they won't break under a full vacuum. But they'll probably go fairly squiffy as the vacuum starts
getting harder, or the guys using turbo molecular and ion pumps could use them and save a few hundred / thousand.
Provided they work down to fridge pump levels, or a bit below, they'd be good for those users and the aspirator guys. Scraping the bottom of the
barrel at 0.1mbar and below isn't really that essential for distillations, and is more of a problem for me a lot of the time (which is why I'll
purposefully use fridge pumps over a rotary sometimes, and save the rotary for things like driving off the very last few drops of volatiles).
I asked some of the guys on eBay who sell silicone vacuum hose about using it at 10 and below mBar values, hoping it'd be more flexible and easier to
use than the thick orange rubber I have now, but they said it'd probably collapse.
[Edited on 29-8-2010 by peach]
|
|
|
watson.fawkes
International Hazard
Posts: 2793
Registered: 16-8-2008
Member Is Offline
Mood: No Mood
|
|
The odd thing about these is that their maximum pressure rating is specified at slightly than atmospheric
pressure, not more. I'm assuming this isn't a problem for device lifetime, but it is a little funny to me.
I couldn't determine what the packaging material is made of, but it is injection molded; there's a maximum dimension for flash listed. I'd guess the
package material is a weak link in using one of these for solvent stripping, as there's certainly some solvents that will degrade the package quickly.
You could protect it with a U-trap, but then you'd need measure the manometer you've created.
The other packaging material are a "stainless steel metal cover" and a "silicone gel die coat", neither of which seem potentially as problematic as
the plastic case.
|
|
|
Jimbo Jones
Hazard to Others
Posts: 102
Registered: 15-10-2009
Member Is Offline
Mood: No Mood
|
|
http://www.instructables.com/id/convert-a-tire-inflator-type...
Sounds interesting. Any input from someone with more experience in this field?
|
|
|
deathmetals
Harmless
Posts: 4
Registered: 14-10-2010
Member Is Offline
Mood: No Mood
|
|
I worked with those at my internship.
We did an indurence test with them they can fly appart with in a month. After 10min an hour 9hours a day. but some work longer a lot
You probably will need to change the oil its that chinese grimy black oil. It will work better with new oil.
I also found difrent engines with in the same batch we ordered.
They where in the same casing tough
The test where done bu letting the compressor go to 3 bar (if i rember correctly) and back again.
Also found that some cant get to above 10 bar or the manometer was broken
Never used them to pull vacumes tough.
Tldr: the quality varies alot but if u can get a good one it may be very useful.
Edit: if they fly apart, they fly apart hard! Plz build. Somekind of protection or casing wile you run these things. a part went about 0.75 cm in to
some building wood.
[Edited on 10-4-2011 by deathmetals]
|
|
|
Jimbo Jones
Hazard to Others
Posts: 102
Registered: 15-10-2009
Member Is Offline
Mood: No Mood
|
|
I plan to buy some of the powerful brands and give it a try. Maybe some tuning will fix the things, but before I get the air compressor these are just
some wild dreams!
|
|
|
| Pages:
1
2 |
![[*]](images/xpblue/default_icon.gif){kind=icon}
