Another Dirty Turbomolecular Pump
ฝัง
- เผยแพร่เมื่อ 26 ก.ย. 2024
- I acquired this decommissioned turbomolecular pump. It's a Turbovac50, the same model currently in use on my vacuum chamber. As it turned out, the pump was contaminated with oil. This video is meant to provide a glimpse into this pump. Unfortunately, the cleaning was not entirely successful. The smell of oil is still noticeable. So, I'll have to disassemble the pump again to thoroughly clean the rotor and ball bearings as well.
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! / advancedtinkering
I can only speculate that those copper fingers are in fact some sort of Shaft Grounding Rings, to protect the bearings. Especially if those bearings are ceramic. It's hard to tell from the footage, but it looks like the rotor of the motor is an integral part of the overall pump rotor. Specifically, the squirrel cage of an asynchronous motor, fitted to the end of the rotor shaft. That will induce circulating currents in the rotor that will escape, one way or another.
This is by far the most plausible explanation for the copper part. Thank you very much!
Yep, i would say the same thing, that is a thing on larger scale industrial motors, some have carbon brushes on the rotor for grounding in order to avoid ballbearing corrosion due to potential diference. You can also go by the shape of the piece, if you have disassembled mains powered electrical devices before, maybe you had notice that screw ground wiring terminals had a special type of washer with sharp edges made to pinch through paint grease or whatever in order to make chassis contact.
They are used to protect bearings from the negative side effects of High Shaft Voltages, that are present from variable speed drives (VSD’s), for both AC and DC motors, as well as eddy currents from large AC motors. Ceramic bearings don't really give a damn about it. But metal bearings do. Most of them are brushed, but for smaller motor assemblies like this, a copper ring like shown here is sufficient.
@@tiagoferreira086 these are just locking washers btw. has nothing to do with grounding and chassis contact, but can be a neat side-effect, generally you dont want such a connection with only a few small points pressing on the metal, but the full surface.
Same thing I was going to say 👍👍
I used to work for a company that built these pumps, serviced many. Some where very badly damaged when the bearing failed and all the turbine blades had been stripped off in the subsequent crash. We call those pumps, "Corn Cobs" as the stubs on the core resembled an eaten corn cob. The precision and beauty of these devices was always amazing.
Turbo salad...
That’s awesome! So what’s your idea about the mystery part he found within it? Any insight??
The copper ring keeps the bearings from pitting due to static discharge. Latex gloves have natural protein that can contaminate your work. When you spun the bearings by hand they sounded dry. I don't know how much lubrication is needed for bearings in this application. Pumps of this design should run hot on the working side as it helps pull a deeper vacuum. I've seen some pumps use a blanket of foil to ensure they stay warm enough. If you want to cool any part of this pump cool the bearing end.
So nitrile gloves then?
Do they even sell latex gloves these days? I only see nitrile anymore…
I think the copper piece is for heat dissipation. The bearing is in vacuum and can overheat if there is no path to cool it. The copper fingers provide a path for the heat to diffuse out of the pump and into the bottom housing.
Why am I watching this? I’ve never even heard of one of these before today
it sucks
I’m watching because I’ve heard of them and I desperately want one but can’t afford them so I live vicariously through strangers on YT….😒
Some life, eh…?
The copper piece looks like it is designed to accept induced eddy currents. Possibly a shield to protect the electronics from the induced AC from the windings.
From my experience, the Turbovac 50 does not need additional cooling, it will happily run at 50c operating temp, above that its a problem. Once the bearings have distributed the grease after sitting for a long time, and with minimal gas load, it shouldn't go above 40c. I also had a smell issue coming from one pump, it was degraded bearing grease, it went away after A LOT of heating and pumping, the smell was like something rotten... Turbovac 50 is a great little pump, it has a very rigid rotor setup, so you can move it while its running, unlike the HiPace80 that will self destruct if you sneeze near it...
That is very good to know, thank you! I assumed that cooling was necessary as my other Turbovac50 has a cooler installed.
I worked at Solyndra, we used TMPs in Panel Framing Leak Detection of the silicon oil filled solar modules, unfortunately when a leak was detected it meant the TMP on that set of four modules was now contaminated with silicon oil.
At first we would partially disable the pumps and rinse with IPA but as more Panel Framing tools came online we were overwhelmed with oily pumps so it got to where I would just cap and fill half way with IPA and shake for a few minutes, dump and repeat then bake out and test.
It worked very well and good thing because we would have had to have a great number of pumps on hand if it didn’t.
Nicely done video 👍
so these pumps are basically the compressor section of jets engines
Pretty much, except far higher RPMs
Not really..turbo molecular pumps operate under a decent vacuum, usually a vane pump to about 100 millitorr. Running a turbo molecular pump at atmospheric pressure would stall it. The motor has insufficient torque.
If a leak happens at speed, the blades can be damaged.
@@keithjurena9319 but if I add a stronger motor ...tim allin grunt...jk
well not quite , these pumps help out the very last molecules in a high vacume like we like to see in glasswork on neon and cathode ray screens , as you noticed the multi blade increased vane rotor is the most effective broom on the market
if it can reach 100k rpm in open air I’ll salvage one and make a micro jet
1:15 Two things come to mind. They may be independent... or combined.
1: Potential for left hand threads.
2: Chemical thread locker. Heat threaded area to about 250°C
I have dismantled and cleaned the same pump like two or three years ago and it as much nastier inside and really had a terrible smell to it, the copper fins were extremely corroded and covered in green oxide, I also did not disassembled the rotor because I couldn't get it apart and I didn't want to mess up the balancing.
I watched your video about the lasers for laser shows. Great video! I always wondered why the relatively powerful lasers are safe to use. Now I know it's due to the fast scan speeds. Very informative!
Ive never seen inside a turbomolecular pump before, what fantastic tolerances to ensure the blades dont touch when spinning. This stuff is way outside my experience but fascinating none the less
As somebody who is new to working on TMPs at work, I really appreciate your vids!
While like no documentation is available regarding rebuilding specific TMPs, they're close enough that your vids are useful as references.
I had a conversation with the tech support guys over at Edwards vacuum pumps and they said that they would only work on pumps where the customer could sign a document to certify as to exactly what chemicals it had been exposed to. Otherwise, they would refuse to service the pumps. I hope you know exactly what this pump was used for. Apparently they are often times used for some pretty nasty stuff.
That's actually very good advice. I also mention it in my newest video about the large turbomolecular pump. I should have probably also pointed that out in this video.
I don't even know what a turbomolecular pump is, I got this recommend to me.
Turbomolecular pumps are specialized devices used to achieve very low pressures, also known as high vacuum and ultrahigh vacuum. These pressures wouldn't be attainable with conventional vacuum pumps like rotary vane pumps.
The rotor of the turbomolecular pump spins at very high speeds (in this model, at 60,000 rpm). The blades of the rotor impart momentum to the gas particles in the vacuum, compressing them so they can be pumped out by a rotary vane vacuum pump.
That's a somewhat simplified explanation, but I hope it gives you a rough overview :)
You will need to rebalance if you disassemble completely. Thanks for sharing :) very cool
due to high speed, the dry air will carry static charge and give it to the pump, i think the copper sheet is for that
i used to do helium leak tests in a clean room - filled some latex gloves with helium and floated them around room -
was amazing how many asked - how I did it
Those must be the hydrocoptic marzlevanes I've read about. When fitted to the ambifacient lunar waneshaft, side fumbling is effectively prevented as well as producing power from the modial interaction of magneto reluctance and capacitive duractance. Quite remarkable, really.
Well you got me. I googled hydrocoptic marzlevanes😅
Words right outta my mouth! But you failed to mention the continuum transfunctioner and its role in flatulation during the emancipation proclamation of the alopecia prone philatelists…🤔
….methinks
@@revmsj well, I was considering encabulation from purely a historical perspective, as I'm only familiar with the specifics around retro-encabulation. I hear that with modern engineering approaches, great strides have been made, increasing the repeatability and reliability of encabulation, not to mention the AI applications that have been presented recently. I'm always interested to learn more, so thanks for elucidating us further on this ever developing field. The applications seem to be ever expanding.
Old Leybolds used greased bearings and did not last long. They switched to oil-lubricated and had more success (but could no longer be mounter horizontally).
7:16 9.5e-06 mbar is more than ok. That’s insane.
The copper piece is probably a leaf spring for bearing preloading. Most electric motors have wave rings for the same purpose.
The copper piece is very thin. There is no way it acts as a spring. It bends by looking at it the wrong way :D
OK, I thought it's thicker....
I was thinking spring as well, but to align and maintain blade positioning, but I too was thinking the copper bit was thicker. Oh well…
I’ve never seen a turbomolecular pump that tiny before! 😂
It's so cute!
The cutest!
My thoughts on the mystery copper part within your pump is that it’s likely a spring to ensure the rotor/stator blade assembly stays together somewhat tightly and resists vertical movement along the shaft (lol that’s what she said…) and prevents blade/blade and blade/housing interactions which could be catastrophic…
I saw someone else mention it’s some sort of bonding mechanism to prevent any static buildup of the shaft and rotor to protect the bearings, but my understanding of this sort of motor is that the only electrical activity within the rotor is back electromotive force (BEMF) in the armature which dissipates as heat leaving nothing static, but I’m no engineer and am in no way formally educated so what do I know. Hell, perhaps it’s both a spring for turbo blade assembly stability as well as a bonding assembly all in one…🤷🏾♂️
This guy running helium leak checks in his garage like I do at work checking chambers for silicon wafers 😅 nice!
That copper piece is to dissipate static charge
very good engineering of that thing
Using a can bus or rs485 to collect data would be a great feature.
And I know that this video is pretty old by now.
0:55 Its evident its an -90 pump just by the simple fact that the vanes are just bent to get the right shape instead of being milled from a solid piece as newer are. Its also evident its older due to the fact that the edge of the vanes now after they have been bent have a small sliver of a 90 degree angle pointing at a angle upwards to the chamber so some small amount of gas will be going in to the chamber again instead of being pumped out, so these pumps are little slower than all others and also have a slightly lower ultimate vacuum.
5:50 yeah if you have a helium leak detector you end up using it for everything. Glad you're getting good use out of it. Turbomolecular pumps have a great compression ratio for oil due to the high molecular weight, so I would leave it running a while (days) but monitor the motor drive power (make sure it's either staying constant or decreasing) as that may be more effective and save you some troubleshooting for dealing with the last bit of oil.
Thanks for the advice! But I think I will take the oil as an excuse to dissasemble the rotor. I would like to know how complicated it is to change the bearings.
@@AdvancedTinkering Yeah that would be cool too, and probably more useful long term. Good luck!
dann frag halt nicht, press fit - viel spaß , turbo gibts nur in 2 zuständen - funktioniert oder explodiert @@AdvancedTinkering , ohne grund ... fürs video ok
Verstehe die Antwort nicht ganz. Wen soll ich fragen?
Dass die Lager über ein pressfit angebracht sind, ist ziemlich wahrscheinlich. Die Frage ist, wie leicht es ist sie zu tauschen. Lassen sich beispielsweise die nötigen Werkzeuge dafür herstellen und so weiter. Das wären alles wichtige Informationen für meine andere Pumpe gleichen Modells.
Thank you! I will let you know if it worked.
That looks like the thing off of Jurassic Park
Perhaps that copper sheet is an electron emitter to neutralize charge in the gas stream.
Electrostatic dust is heavy to remove for turbine and that copper plate probably is special to discharge some residual voltage, but I’m not sure
The leybold pumps are simple to fix phiffer pumos are nightmarish. This one seems very easy as well. ❤
The rotor might feel okay but on that type of equipment the fact you can hear it spinning the way it did at 1:34 to me says it indeed was discarded for a bad bearing.
such high quality content. thank you so much!
Thank you! I appreciate hearing that!
Oh my lord I haven't seen anyone disassemble or otherwise work on such a neat little machine, only other folks I know who do similar things are The Thought Emporium and NileRed!
its funny that you let the prepump running for a while before you start the TMP, at work we got a new highvacuumoven with a spektrometer and it has a Turpovac50 for its own vacuum side and we have like 2 meters of pipe to the prepump that turns on the same time as the TMP. the entire setup for it was planned by Leybold and so far works fine, needs like that 2 min to get to operation vacuum. the oven needs way more time to get under 5.0e-04 mbars that is the limit for the valve to open :D
Word on the street is you can damage your TMP if you let it run prior to establishing a good rough vacuum…🤔
@@revmsj the oven maker and the guy from leybold said its fine, apparently the volume is so small that the prepump will created enough vacuum to not be bad for the turbo pump. But if not its not my problem if it dies
The screws that appear to be super tight might have a thread sealant to lock them or are reverse thread (lefty tighty)
It's not a reverses thread and I allready tried heating the nut to loosen any thread sealant. Without luck.
I'd love to get my hands on a pfeiffer leak detector
The copper ring... they work as shims (spacer) perhaps?
nice job, i number the parts with permanent marker to assamble them correct, the vacuum grease seems a bit execessive or not necessary. Also because its aluminium, i cleaned the parts in like 25% NaOH, but with the oil you had i guess thats enough. Maybe instead of taking it appart again, just let it run for a week? (but e-6, with nothing attached, is a bit high)
also, for my application allot of P,S and metallic powders are the problem, still running at 1-e7, fresh it was -8 or lower, piriani cut off, thats why it is called the undertaker :)
Thank you! Filming during dissasembly helps me a lot. And I take a reference pictures. The permanent market would simply wash away in the solvents.
Yes, someone else also recommended letting it run for a while. Thanks for the tip! But since I have another turbomolecular pump (same model) I think I will try to dissasemble the rotor to see how a bearing change would work.
Using 25% NaOH on aluminium parts would destroy them, don't you mean less concentration?
well, it was like 5-10 seconds :), but we have like the periodic table in the pump, the blades showed dents of impacts with residue, might be a bit extreme :P but helped, i had to mechnical remove some chunks stuck in betweeen - so big@@teresashinkansen9402
i see, but the oil would not be there (radial forces), but for a video -shure, bearings are press fit - not so easy, just with solvents and re-grease, maybe, GL,Grüße aus Dresden@@AdvancedTinkering
Can you please mix lime juice and lithium?😀💙
I used to rebuild maglavs for Alcatel before Pfeiffer bought them.
Copper piece is a shim to get the correct pressure on the O ring the fins are to disperse heat
I don't believe they are intended as shims. The shims are simple steel disks, and this wouldn't account for the unusual shape of the copper piece. Additionally, it would be too thin to effectively disperse heat. Another comment suggested that its purpose is to ground the shaft, to protect the bearings.
Amazing job well done how are you 😊
1:20 dis nuts🤣
The copper fingers look like some kind of Hall effect device.
Can you make a video describing the pros and cons between a Turbomolecular and diffusion pumps? I need a vacuum pump to make art display plasma discharge tubes but don't know which is the right one. Thanks.
alcohol cleaning should always be done last to drive any water out of the cracks and prevent rusting.
Nice to see someone giving some love to these kinds of pumps.
When you spin the pump by hand is it slowing down to a halt or is there some „abrubt“ stop at the end? If so the bearings may not be the best anymore.
Also did you do a soft start for starting the pump after reassembly? The theory behind this is for the bearing that may have been sitting for some time and being somewhat stuck, meaning the grease in the bearing may increased in viscosity due to not operating the pump. What could lower the life expectancy of the pump/bearing.
Turbomolecular pumps are simply fascinating devices.
I'm not an expert on ball bearings, but I would say that they came to a stop as one would expect from a greased (permanently lubricated) bearing. It didn't slow down as slow as an oil-lubricated bearing, but that's to be expected.
Nevertheless, it's probably a good idea to replace the bearings. I'm currently waiting for the tools I've ordered and will then attempt to remove the two nuts again.
Thanks for the tip about the soft start! I assume the pump controller has a slow-start mode? I'll check the manual for that.
@@AdvancedTinkering I would love to see about your attempt! And in general about all kinds of vacuum devices.
I‘m not a hundred percent sure about Leybold controllers. I know Agilent/Varian controllers have it. If not maybe you can power the pump somehow manually and therefore control the RPM for the first start. But how you would do that I don’t know.
If i am ever get my hands onto a Turbomolecular Pump i know what to watch. But i feel like this will never happen.
Thanks, good video, very educational _)) Do you know what could be a possible reason why the HiPace 80 spins at 100% speed but only reaches as low as 0.032 Torr? (excluding leaks)
How do you always get those cute pumps :( the smallest i always get are 15kg kinetic bombs....
that piece of copper is really interresting, i have little clue as to what it is, but id expect something usual or boring like heat transfer from the motor coils to the casing, or just to get a better seal to the spacer, or a fine spacer itself to fix the bearing in a more rigid way?
yeah if anyone know, i too wanna know it now
Haha, we can swap. Tiny cute pump for a kinetic bomb :D
ill definetly consider it, if my next one turns out to also be one, for sure!@@AdvancedTinkering
Can this sheet be an abuse indicator? For example, if a user dumps a lot of gas into a fully spinning turbopump, the turbulence will bend these leaves out of shape. That can later be used by servicemen to see if the pump has catastrophically failed due to manufacturing/design defect, or abuse...
Is the copper part a shim for end float
sorry I just wondered to this video by accident. What is the purpose of such a pump?
Nice to see that you dropped by ;) Turbomolecular pumps are specialized devices used to achieve very low pressures, also known as high vacuum and ultrahigh vacuum. These pressures wouldn't be attainable with conventional vacuum pumps like rotary vane pumps.
The rotor of the turbomolecular pump spins at very high speeds (in this model, at 60,000 rpm). The blades of the rotor impart momentum to the gas particles in the vacuum, compressing them so they can be pumped out by a rotary vane vacuum pump.
That's a somewhat simplified explanation, but I hope it gives you a rough overview :)
@@AdvancedTinkering I got that, thanks. what do you use it for?
For my vacuum chamber. You can see it in my other videos if you are interested. I want to evaporate metals in a vacuum and deposit them on surfaces like glass. At least that's one of my planned projects.
I wouldn’t touch the rotor.
Why not just heat and vacuum?
hadnt the older Tvac pumps a nasty habit of exploding?
Haven't heard that before. I never had any problems with mine.
it surprises me that still a lot of people know these apparatus. very high end very niche ... and if you want to improve the bearings.... rollerscate ceramics do better
How does something with this many loose parts right next to high frequency spinny things not make air raid siren levels of noise
So how are you doing with my Mr. Fusion I dropped off a week ago?
Could the copper be functioning as a heatsync?
Interesting idea! But it's directly on the aluminum. And in a vacuum, heat transfer through the air would be practically zero. I suspect it serves a different purpose.
Is it some weird type of o ring? For super high vacuum I think copper rings are used as o rings. But also I think once they are used once they can’t really be used again.
Copper gaskets are used for CF flanges. However, these are much thicker and have knive edges on the respective flanges. It is definitely not a gasket.
What tweezers do you use?
Hey, I actually have an osaka 1800L/m turbo pump with controller. It has quite a bit of corrosion on it. would you be willing to point me in the right direction for restoring one of these?
Do you love next to a busy road? Can't work out what that noise is in your video.
Yes, it's a busy road. I hope to be able to upgrade my audio equipment in the future so all the background noise will be gone.
How you get such plausible English
Hi AT,
I´ve got a question.
Are the rotor also weighted or is it just the shaft?
Kind Regards
There are holes from balancing the rotor at the top of the rotor and at the bottom of the shaft (basically also the rotor). If that's what you mean.
Noice
Per hour? Are you sure?
I'm not sure what you are referring to.
@@AdvancedTinkering I believe you've misrepresented the performance of your vacuum pump a bit. My suggestion: m^3/minute instead m^3/hour.
@@nikolaybolobanov2634 Oh, did you possibly comment under the wrong video? I suspect you're referring to the video about my new vacuum chamber, where I discuss the pump's performance of 2.5 m^3 per hour. Or TH-cam might have placed the comment under the wrong video. However, the value is accurate. The pump has a pumping speed of 2.5 m^3 per hour, not per minute. That's a typical value for rotary vane vacuum pumps of this size.
@@AdvancedTinkering Oh, you're right, my bad
@@nikolaybolobanov2634 In the future always provide a time stamp. Let's not waste peoples time.
Never spin dry high speed bearings, EEEEK, ! NO! and when you disassembled it appears you did not assemble exactly the same? now it wont be in balance .
The bearings weren't dry (at least not more than they were before the dissasembly). That's basically what I said in the video. I only cleaned the top part of the rotor to avoid washing all the grease out of the bearings.
And the rotor is the only moving part and was not disassembled. The rest was assembled as it was before. So how would there be imbalance?
1:20 dez
You use a high amount of Helium for the leak check. You contaminate your workplace with Helium and it can be difficult to do a second Test in a shorter time period. Because Helium is everywhere. Normally you let it come out on a tip of a needle. You can set the amount be holding the needle into IPA and adjust the flow to slow bubbling.
Copper is static discharge path to prevent bearing damage
I know that grease is very expensive, I work for an NDT company, you only have to apply enough to make a light sheen. You shouldn't see any globs or texture when applying it. If you use too much, it can be sucked into your system(potentially adding to the contamination in your turbo). Also the usual flow rate for helium, especially when doing such a small test fixture, is three bubbles a second. This can be measured in room temperature water. When you place the nozzle under the water, you should see three small bubbles form in one second. Too much helium doesn't hurt, or make it easier to find leaks, but it will screw with your background reading on your MSLD.
That copper shim that you have wondering about, I believe that's for neutralizing static electricity, I'm not sure
Other comments also suggest that. Sounds reasonable! Thank you!
What are these pumps used for
Turbomolecular pumps are specialized devices used to achieve very low pressures, also known as high vacuum and ultrahigh vacuum. These pressures wouldn't be attainable with conventional vacuum pumps like rotary vane pumps.
The rotor of the turbomolecular pump spins at very high speeds (in this model, at 60,000 rpm). The blades of the rotor impart momentum to the gas particles in the vacuum, compressing them so they can be pumped out by a rotary vane vacuum pump.
That's a somewhat simplified explanation, but I hope it gives you a rough overview :)
@@AdvancedTinkering thank you, great explanation.
They gave me no end of problems when working on x ray inspection machines
I think the copper disk is for static discharge!
No audio?
I am really sorry. I don't know what went wrong. I'm looking into it.
Should ne fixed now :)
I'm not sure that fully disassembling the rotor is a good idea, as rebalancing it later will be next to impossible (this is the first video I see of yours, so maybe you have the capabilities/skill to do it)
I can tell you that I don't have the necessary equipment/skills to balance a rotor anew ;) But if I mark the exact position and orientation of the individual parts, it should be possible to disassemble and reassemble the rotor without generating too much imbalance.