Advanced Tinkering - 2023-08-04
I received this decommissioned helium leak detector from Pfeiffer Vacuum. Helium leak detectors are used to detect the smallest leaks in large vacuum systems. For this purpose, helium is released around the flanges and weld seams, for example. If this helium enters the vacuum system through a leak, the detector can detect it with its mass spectrometer. At the beginning it looked as if the device was in good condition, but then I made an unpleasant discovery... Join my Patreon and support my projects! Your contribution means the world to me and helps bring my ideas to life. I truly appreciate your support! https://www.patreon.com/AdvancedTinkering Vacuum Hax0rs Discord server: https://discord.gg/2c7SQtfG
Hi! I do Helium Mass Spectrometer repair and maintenance for a living.
These older Pfeiffer units are somewhat of a collaboration between them and Inficon. The spectrometer tube and calibrated leak are both Inficon models which were licensed to Adixen-Alcatel which would later have their vacuum division purchased by Pfeiffer.
For a hobbyist/home user having an Inficon spec tube is great as you can expect the filaments to last basically forever. They run reasonably high voltages but incredibly tiny amperages compared to every other brand and also have a far larger ion beam of relatively lower strength.
The big failure points on those units are the pumps (those mini rotary vane pumps are awful in general) and the pirani vacuum gauges. You had oil in the valve block because the anti-suckback on the pump failed. This means that if it isn't rebuilt every time you turn the unit off you run a huge risk of pulling oil from the pump to the highest point of vacuum, which is your turbo.
You mention at one point using a silicone tubing. Do not do this. Silicone and helium mass specs do not mix. Silicone is highly permeable to helium and also hangs onto it an releases it as a virtual leak relative to its density. You should always be using Buna-N (80 durometer is recommended) or Viton for seals and Tygon or metal tubing for hoses (hardware store vinyl tubing does work but usually has lots of mold release and outgasses like crazy for the first few hundred hours).
I'd be happy to answer any other questions you have on this. It's always good to see people interested in Helium Mass Spectrometers.
Thank you a lot for all the advice and information! I will check the suck-back valve of the rotary vane pump.
@@AdvancedTinkeringThis comment said what I was suspecting.. that the oil from the roughing pump might have been pulled backwards into the turbo pump when the unit was shut off and allowed to stand with high vacuum in the chamber.. I was unaware of an "anti-suckback valve", so I was going to suggest a procedural solution - "never do that" - have a vacuum relief valve in the experimental apparatus, run this devise for the duration of an experiment cycle, then end the cycle by opening the vacuum relief valve before powering down this unit.
Videos of this calibre can never be too long! This was a fascinating and nail-biting teardown and repair of a piece of equipment I had no idea even existed. Thank you so much for taking the time to explain so much of what you were doing.
this piece of equipment is awesome!
As a lab manager dreading the day I will have to do maintenance on one of our Pfeiffer mass specs, this video is extremely valuable!
Thank you!
There are way to view people out there able to repair things. But too many able to throw stuff away for buying new.
I'm very jealous of your new toy! Excellent job on the repair, I was on the edge of my seat watching you disassemble and reassemble the turbo. Quality entertainment, haha.
Your videos are always fascinating. I'm impressed you got that pump stripped down and put back together, that looked seriously fiddly. The part you didn't know the name of is called a cable gland in English. Looking forward to the next video already
i would not want to know how much this thing would have cost, its always a gamble with rarities like this, either you end up needing to save in the 4 figures, or find one floating on a (for youtube to remain unnamed) site for 50€ from someone who has no clue what it is.
ive also never seen such a cute turbomolecular, its a very very compact setup that way.
amazing video as always! I do love learning from reverse engineering high tech stuff, espeically if its still in one piece haha
What a great win !!! Many thanks for sharing with us.
I hoped to see the pcb and the electronic because I enjoy seeing scientific equipment and measurement circuits. But I am happy for you since dismantiling the control corcuits was not needed. Enjoyed each second of the video, it is great
Thank you! I'm glad you enjoyed the video!
You may actually get lucky in the future. According to the error log of the machine the 24V supply for the mass spectrometer is low. It doesn't seem to effect the machine in any way so I decided not to mess with it. But if it fails or some other problems show, I will have to take a look at the electronics.
more or less a scientific thriller, very entertaining. thank you👍👍👍
Thank you! I'm glad you liked the thriller ;)
My reaction from 10:00 is that after disassembling and reassembling a vacuum system, it's very helpful to use a helium leak detector afterward if any troubleshooting is required :P If only you had a helium leak detector with which to troubleshoot your helium leak detector.
As of 42:00 I'm pleased with how it's going, and I think it's probably going to work, more or less. You were as thorough as you could be for a first attempt at reassembling a turbomolecular pump. Also, how hard is it in Germany(?) to make a company so that you have perceived legitimacy dealing with companies?
49:45 cable gland, cable grommet and/or strain relief are phrases I would use for that thing
Congratulations! It's a leak detector! I would advise you not to break the display like I did when I was an undergrad by dropping a wrench on it, but since you've got an older model without a glass touchscreen it should survive modest wrench drops. I will also warn you that having a helium leak detector will tend to make you forget about other methods of leak detection, because the helium leak detector is both more sensitive and easier to use.
A more real piece of advice I would give is to put a port with an isolation valve (ie adequate to stand between atmosphere and 1e-4 mbar vacuum with sufficiently low leakage) branching from the line between the diffusion or turbomolecular pump and the backing pump of your working vacuum system, and also have the option to valve off the backing pump. Maybe even with a ballast chamber in the middle of the line. This lets you keep the system under high vacuum with either no backing pump, the helium leak detector as the backing pump, both pumps, or even no pumps for hours at a time if the ballast chamber is large enough for the combination of leak rate, outgassing and compression ratio of the high vacuum pump in your system. Since this is all on the rough vacuum side of the high vacuum pump, it can be cheap stuff with buna-n o-rings, little ball valves and ordinary pipe fittings, or similar improvisations as you desire.
Haha, I thought the same thing during reassembly. But if I understand the system correctly it should be able to "diagnose itself". If I release helium around the sealing surfaces of the machine while it is testing I should be able to detect if helium enters the system.
It is not that hard to register a company. Actually fairly simple. But it comes with added bureaucracy (mainly tax releated). At some point I will have to register a company due to the YouTube revenue. But I'm still far away from the threshold which requires me to register the channel.
I am impressed! (Even editing a 1 hour long video💀 respect!)
Amazing piece of equipment. Can't wait for your sneak peak project ;)
Once upon a time I was a young engineer and I specified, installed and qualified a GEA system for two such MS devices.
I‘m pretty sure it was the same manufacturer but the turbo was external and we had it integrated in to the DCS of the plant.
It was very useful for detecting Si-Oil leaks and also gathering data on our products that somehow nobody wanted (if validated: never touch!).
I love this type of kit.
Thank you for sharing and warm regards to Elias 😊
A side note on oil..
I recall walking up to our screw compressors (yes, the Japanese ones by Mycom..) as I had heard a slight howling noise.
Instead of clear oil I found a whitish emulsion in the freaking sight glass.
As we were in the crucial steps during drying I didn’t hit the E-Stop, I notified my boss and asked our vendor if they „had one one stock, we may need a drop-in replacement very soon“.
We ran that compressor for weeks on end, it howled and sung churning „milk“.
I can’t remember what made that oil go white but it was something to do with water.
That‘s perfect when you‘re using oil as a lubricant and a sealant. Great stuff.
We had no deviations, pump-down was completed within the vendors specifications - we then found out, that a stretch of piping was not large enough on the suction side.
We also had the gunk analysed in the pump, until today I wonder where that water came from.
It was literally H2O in oil with minor impurities but not what one would expect from the Ca-laden cooling water used in the jacket of the compressors.
Anyhow - those compressors are expensive but nothing compared with our revenue/batch or a lost day manufacturing. We had two compressors per unit, they were used for the condensor in one circuit and for temperature ctrl of the mantle and shelves in the unit.
If anyone needs a compressor that will compress just about any gas and also tolerate a bit of water of unknown origin, then head over to Mycom.
You‘ll find a video on PascalAIR here, -100C with ambient air as heat transfer medium..
Very interesting rebuild, great job!
I have 2 old turbopumps that I scavenged in working condition many years ago, but I have not dared to try to start them before cleaning. We used to rent part of a small clean room in the basement of the then new MC2 center at Chalmers University here in Sweden. The room was used soley for reconditioning their turbopumps and similar stuff. The cleaning and service of these seemed to be a science in itself. They oven baked everything before assembly.
You mentioned silicone tubing, it was tried as a return system to He liquifier in the building but was found to be useless, very permeable for He so it is perhaps not the best choice in Your case. I have a small Edwards He sniffer that has proved very useful. Many polymers leak He as crazy, has been a useful demonstrator for some of my work on toxic chemical transfer systems. It has different applications and for a vaccum systems You need to slightly over pressurize the system with He and look for leaks with the detector. Vaccum seals etc may behave differently under overpressure which can be a problem.
Your unit looks really wonderful, very pretty build! I would nevertheless perhaps consider lifting out the oil pump and place it on the floor, this may reduce future oil ingress to the turbopump. But I guess the vacuum line needs to be fairly short for prompt response of the detector.
As a Pfeiffer Vacuum Service tech i said "No no no!" quite a few times. :D
Haha, I'm sure you did :D. I hope it wasn't to hard to watch ;)
That little screw you had in middle of oil cartridge to help lift it threw me off at first. I never saw that being put back, but it was never there to begin with :)
Impressive repair job! well done.
Thank you!
49:50 Grommet, gland, wire nut, strain relief; take your pick.
awesome content! glad to gain better understanding of vacuum systems
Das nenn ich mal Puzzeln für Fortgeschrittene. Froschlers Ölbanhammer war auch dabei. Great work.
So glad I found this channel. The name of the channel is so accurate too haha
Haha, thank you! I appreciate it!
There will be a new video today :)
Cool amazing job great work
the turbo molecular pumps are really beautifully machined :)
english word for that @50:00 is "cable gland" which is a cable strain relief device
Excellent video! I would've cleaned and flushed the rotary vane pump, but if it works fine without, that's also alright.
A lapel mic doesn't need to be wireless to work fine; if plugging a wired one into your camera directly doesn't work you can connect it to your phone and record your audio on that and then put it together afterward. Clapping your hands is a good marker to sync the tracks if your software won't do it automatically.
I'm working is plasma physics field, and my Ph.D. thesis was a new method of using such helium leak detector. Since it just really precise mass-spectrometer for 4 a.m.u. we eventually made it measure, quite precice, the amount of helium atoms (and deuterium molecules) passing throught it. It always strikes me why nobody figured it out before.
Where can I find the community you asked for help when you took the bearing screw out? I am getting started with UHV equipment and such a community would just be amazing...
Anyways; great video you made! Glad I found your channel!
Thanks! I added the link to the discord server in the description of the video.
You're performing CPR on the system. As you found out, and I assume you got the unit for free, it was already dead. You can't 'kill' it anymore. (As one my CPR instructors called it, they're already dead, you can't make them any deader). Had your attempt failed, it would still be fundamentally dead. But now it's alive! Have yourself a frankenstein moment and a beer :) Congrats!
The stators are bent that way to act as a spring sort of. Keeps tension.
I suspected something like that. Thank you for the information!
I work with HIVAC systems in the semiconductor industry. We use helium leak detectors all the time. I found this video very interesting as well as your other turbo videos. Great job!!!
fantastic work
Thank you!
4:43 in and I'm thinking someone changed the filter etc and didn't refill the oil, this was followed by someone running it dry...
8:08 oh... it is SO much worse.
That part at 50:00 is probably what we would call a 'Gland' in English
Nice video, I learned a lot👍
Man, great video ! I was thrilled the whole time.
Thanks! I'm glad you liked it!
This leak detector design is just nuts. Definitely not the german engineering we get used to. The TMP should be always higher than the RVP otherwise this happens. With this layout, under normal operation the RVP slowly fills up the TMP with oil. Without the help of gravity this will happen again. Or try to make a U turn with a longer vacuum hose, where the bottom of the U goes below the inlet level of the TMP.
The oil in the felt came from the RVP of course. I have never seen a Pfeiffer DUO which was not leaky... Repalce it with a Pfeiffer membrane pump, if u have any. A TMP with a drag stage works with a membrane pump as well.
And by the way, replace the rotary vane pump's starting capacitor. In case of failure, this type of starting cap can burst into flames. Trust me, once it almost burned down our lab.
Thanks!
another hint was the color of the hose..
True :D
@@AdvancedTinkering but kudos dude, it takes alot of courage to take a turbo molecular pump apart, re assemble it, and use it.
Great video. Immense respect for being to able to basically take that thing apart.
This video shows validates my immense dislike for oil in vacuum systems ;)
Is that leak detector one where you can switch to use hydrogen instead of helium for more economical testing?
That vessel at the end is indeed a beauty. The number of ports on the top and the big window promise a fancy experiment.
Thanks! :) As far as I know you can't use hydrogen.
@@AdvancedTinkering That's too bad. I know that other Pfeiffer ones can do it, so I speculated it might be somewhere in the options, but it's probably not too big of a financial hit on a hobby scale to use helium.
With a more modern HLT 560 system it's possible to switch to leaking with hydrogen. But the sensitivity is way lower.
9:22 It does not really ruin the TMP, unless it happens when it's at full rpm. Otherwise it just results in an annoying and cumbersome cleaning where the bearing also need to be cleaned very good and then re-lubed or re-greased with the right oil or grease.
If it wasn't contaminated, we'd have never seen inside this amazing device...So nothing is inherently "bad".
... just thinking I was rather pleased when I managed to fix my angle grinder. ;)
I use the same Device. Be careful with the Display. Pfeiffer told me there's no more spare part available anymore. In the way it is mounted it by factory it can be damaged easily.
Awesome video! Cool to see the inside of a turbo molecular pump. Do you think it would be possible to repurpose the pumps to something else or do you think this will be a useful tool to have for future projects?
Thank you! One could probably make a flange that fits to the custom ports on the turbomolecular pump to adapt it to a CF or KF flange. But I definitely want to keep the leak detector. It's a useful piece of equipment and the chances to find another pump that fits the unit are slim to none.
49:40 i usually call that a wire grommet
Would it be possible to modify the System into a "normal" mass spectrometer to let's sas connect to a GC system? And from what i saw in the Video the built in MS only features a magnetic sector?
It's a fixed system only tuned to see He3/He4, it doesn't sweep.
If you had a Residual Gas Analyzer(RGA) connected to a normal vacuum system though, you could. They're tiny self-contained mass specs designed for process monitoring, and they usually sweep from 0 to 50, 80, 100, or 300 amu.
If I understand correctly you could use this or similar leak detector to provide vacuum to a very small system with no other pumps.
Yes, they can evacuate a smaller chamber themselves.
You're right, moreover the system doesn't have to be very small in fact.
But don't expect achieving really high vacuum levels with this approach, - even though the TMP is present in the system its pumping speed is severely reduced by the valve block cross sections. With really small volumes and long times you could reach some 10^(-5)-10^(-6) mbar, but with larger chambers the feasible limit is set by the rough pump with its ~10^(-3) mbar limit.
@markedis5902 - 2023-08-04
Taking something apart and putting it back together so that it works is a right of passage for every engineer
@good-ql6un - 2023-09-19
I work in aerospace and it is the best feeling ever when something is will and truly broken and you figure it out before engineering have then fix it
@darrenconway8117 - 2023-11-21
Taking things apart and not being able to put them back together again is why engineers become engineers.