The black stuff inside the detector is indeed a resistive layer. It provides voltage gradient along the length of the electron multiplier horn's wall. It's basically a standard electron/photomultiplier-like construction, except there are no discrete dynodes (each having a slightly higher potential than the previous one), the dynode is distributed in the form of a continuous voltage divider. The horn shape and field gradient makes the electrons travel to the centre of the detector, knocking off more and more electrons as they collide with the walls on their way in. The detector is mounted at an angle relative to the ion stream, probably because there is a conversion dynode, from which the electrons are emitted when it's hit by a positive ion. This prolongs the life of the detector, since there are no ions reaching it (and eventually depositing on the surface).
Hi, i have experience in such high vacuum systems. In your setup are many things that are not quite right. But first yes your rotary pump has reasonable size for your system. One source of error could be the oil in the forevacuum pump which is outgasing or the grease in the prcision bearings of the turbopump is not warm and the pump needs longer to reach its operating speed (some computerbased controler have a special mode for this scenario). And you should use vacuum grease (e.g. PTFE) to get a better seal ob the o rings. It is very complicated to take a close look at all facetts in a such short time because there are far to many. If you want i could try to answer your questions and give you tips.
This thing shown around 10:35 is *_ion trap_* - arguably the most important part of this machine, and everything else is there just to make this thing do its job. :)
Okay, I know how a mass spectrometer works, it's quite easy in principle. But that you need something like THIS to actually perform the experiment is impressive!
I always find it fascinating to see how huge and macro-scale the machining and shaping of devices have to be in order to operate properly on the microscopic scale. The majority of modern state-of-the-art electronics is unbelievably tiny, so it's really interesting to see something that's so big and physical in order to get super precise measurements.
Man this thing is insanely interesting. This sort of stuff is way above my head. I can follow most of the EE/CE stuff (I would hope so at least, as that's what I'm studying), but man, this sort of engineering is just on another level. Thanks for taking this sort of stuff to bits and sharing it with us all (and testing it out and all that)!
I really love these industrial component teardowns. I've been watching some of your older videos and I wish you could do these more often. Keep up the good work!
With a big enough leak (which you may very well have), squirting solvents like isopropyl alcohol on points you think might be leaking will normally cause a spike in the pressure that you can see with the gauge. If the leak is really huge you may see the pressure go down with isopropyl alcohol, as the solvent freezes into the gap, and then you'll see it rise. Good way to detect leaks because it doesn't depend on absolute accuracy of the gauge.
I would venture to say with at least a semi-straight face that for most of your previous videos I had a decent understanding of what it is you're showing us, but in this case I am just hopelessly lost. Looks nice, though!
Yep - didn't really get any understanding of any of the kit being torn down. "Here's a thing, here's another thing". - would be cool to get a bit more system description, not just components. :)
That pump gurgling is the key Mike, tune it by ear - it sounds it's pulling a significant volume of air. When it's pumped down it will thrum, and there will be a very fine snapping sound, like when you had the meter jammed up against it. I'm unsure about the o-rings too - from experience at only 1-6 torr they can be a problem if they're not done right, properly seated and or greased etc... but if your backing pump can pull 1--4 torr and the turbo pump works then you should get that green light! +NanoCottage, service the pump too if needs be, get proper oil in there and let 'er rip :)
Not a clue what all that stuff inside the vacuum chamber was, but it was definitely cool as hell! I think you need an industrial grade vacuum pump, possibly a dual stage or so, for this to operate. Just guessing.
When did you last change the oil in the vacuum pump? To get the best performance the oil needs to be fresh as it provides the seal on the rotary vanes. Having new oil makes a huge difference, refrigeration guys change their oil often, like once a week. Some pump manufacturers recommend new oil every job for peak performance. If it's still not good enough after an oil change then you can get a rebuild kit with new seals and vanes for not much money.
er, never! I have no idea what state the pump was in when I got it maybe 15 years ago, and it;s hardly been used since. Quite likely it's not in good condition!
Great video Mike, thanks for sharing. A decent low vapor pressure rotary vane pump oil does not come cheap. The closest budget option I found for mine was Anderol 555, i think I got it shipped from UK. Not knowing the type of the oil you currently have in the pump it's definitely worth flushing it out completely and refilling with something new. A liter of new oil made a huge difference on my Edwards pump. cheers, Val
Yeah, something doesn't sound right. The gurgle sound on the exhaust of the RV pump I use goes away once its got the system pumped down. Plus the overheating or reduced vacuum when its warm is another sign pointing at the oil. A RV pump should be able to run indefinitely below a certain pressure. (10Torr on the one I use, allowed upto 10 minutes of use above 10 Torr). If the oil is contaminated with something with a high boiling point (water, other solvents etc) as it tries to expand the chamber, the contaminate evaporates out of the oil, then as it compresses it condenses and goes back into the oil. Sounds like an oil change would be a good starting point to me.
the overheating is also an indicator of massive leaks. the pump is no more than a compressor in reverse. It doesn't "suck" but it compresses gas to atmospheric pressure. So leaks = heat. That being said, that freefloating o-ring or the silicon sheet are not good enough at the level of vacuum the RV can go to. but of course, if the oil is toast it won't go very far anyway. For the turbo, my guess is it's triggering on high motor current. Usually the machine it's mounted on would stop the turbo via the interface by monitoring the backing pump pressure, or flat out not allow starting it if the vacuum behind the turbo is not low enough. But here, it's possible the pump driver doesn't like running at high current for too long.
If the inlet of the pump had been open to atmosphere the oil will accumulate moisture and the lowest base pressure will be much higher then without any moisture in the oil, and this especially noticeably when the pump starts to heats up. To get the moisture out from the oil you should block the input, preferably with a vacuum gauge, and then let the pump run and be really warmed up and let it run until the vacuum is at its lowest with should be at best 0.000 Torr or lower. The pump also have a gas ballast that can be open a little if there is a lot of moisture in the oil and/or the last vacuum take a long time to go down at the end. A fresh good 2-stage pump will go down to 0.000 Torr pretty much no matter how old the oil is, so have that as a number to aim for. I have a pump from -97 that never have had a oil change and that goes don to 0.000 Torr. never the less. An oil change is more to get fresh additives and remove dust, debris from wear and tare and what not. But if the pump rums very hot due to high load the oil could get burned and get broken down and thus loose some of its characteristics and then its a good idea to change it.
i believe what happens in the octepolar trap is that you get a standing sinusoidal electric field which is frequency variable. that ions that enter it only get to the end if the frequency of the electric field is the same as that of the natural frequency of the ions, if they don't match then they are lost out of the trap (natural frequancy is based on mass) so the system sweeps the frequency of the electric field and to separate ions by mass. please correct me if i am wrong but i believe that is how it works.
You should try using valve between your 'regular' vacuum pump and the rest of the system. Otherwise these two pumps just 'fight' each other and you get stuck with the pressure of the weaker one.
Gorgeous stuff, seems an incredible shame to scrap this sort of stuff. I really like that you actually try to get some of this gear going. I would live some high vacuum stuff but there is far more to it than just connecting a few pipes thats for sure
The thing @5:40 is the electron multiplier, not the metal part. @7:50, the rpm of turbopump is way higher than 6k rpm, it should be around 50k rpm. You will need a roughing pump which is more powerful that what you have there. I believe you need a flow rate at around 300 lpm.
Well, I might miss the "60k" part. Sorry about that. In our lab, we used two Edward 30 pumps for a similar model. Well, I guess working with two roughing pumps is a bit crazy. But I guess you may need a better pump that the ultimate pressure is in range of 1e-3 to 1e-4. Anyway, thanks for the video! It is really cool to see you holding an ion trap in person without gloves. J
That's not true. Pfeiffer specifies a fore vacuum of < 10 mbar for this type of pump. In ultra high vacuum applications we typically use dry scroll pumps (you don't want to have oil anywhere near a UHV system) which reach a final pressure of about 0.1 mbar. This is easily sufficient to get to 1E-8 mbar with a decent turbo and some baking.
I don't see a conflict between your point and mine. Specifically, the hardware manual of Thermo LCQ suggested a forepump that meets 650 L/min displacement, which can reach the pressure around 1 Torr (~1.3 mbar, I believe). However, there are something I respectfully disagree. Some mass analyzers (e.g. 3D trap, linear trap, quad, etc.) do not require ultrahigh vacuum (UHV). The pressure of ~1 mTorr (1e-3 mBar) is sufficient for such applications. In contrast, most of the high-resolution instruments require "true" UHV (e.g. Thermo Orbitrap requires 1e-10 mbar). I have seen people using dry scroll pumps (I have used a few before, e.g. SH110, Triscroll 300, etc.); but quite a lot of them are atmospheric scientists. In analytical labs, I am not quite aware of the use of dry scroll pumps, which might due to the difficulties during servicing. It would be nice if you may share your google scholar profile, which would be very helpful for me to get to know your works. :)
Scroll pumps are used pretty often. In fact i see dry pumps more often than oil sealed ones in analytical labs now (scrolls, claw/roots mainly). Scroll pumps easily reach 10-3Torr ranges or less and require less (and easier) maintenance than vane pumps.. and no oil backstreaming into the system. That's one of the main reason they are preffered in high vacuum for some applications. Downside is they don't tolerate well corrosive gases with is usually not a problem for mass spectrometers, electron microscopes and other mostly clean applications.
Very cool!! Be careful about just cutting power to that turbo, those mas spec units typically use active mag bearings. Maybe its not a big deal until youre up to full speed... and I'd change your pump oil out, sounds like youve got some disolved contaminants that have high vapor pressure at higher temperature in it.
Love your videos Mike.....but as someone who works with mass spec's for a living I can honestly say I have never cringed so much when watching a video as I did with this one :)
0:44 Cool.... it'd fun to attach a little getter inside and seal off the top of that triode and pump out the air and set off the getter and then seal the vacuum tip and make an awesome regenerative shortwave receiver or QRP transmitter or something with it. :-)
Any suggestions for a high vacuum shutoff/relief/diverter valve? Something that will work well in the milli-torr range? At least with the Spectral instruments cameras there is an electro-mechanical valve you can close before shutting off the pump. It's a bitch to pull the hose off with a vacuum remaining in the tubing once the valve is closed, though.
20:23 You might know this by now but that is very dangerous to do above a turbo pump. The o-ring over the pump can be is sucked in a little so it creates a big opening all of a sudden and that huge inrush of air would make the wings on the turbopump to go upwards and hit everything in it and at that point I would not want to be even remotely close to it since it will go off lik a bomb.
@@christopherleubner6633 You have? Dam, luckily I have not but seen many and they are absolutely totalled afterwards. One guy I know that work at a university said that he was at the corridor where all the the teams did experiments and he was at his small office with the door opened. A turbopump little bigger that the one in the video totally sized after a small nut fell into it and the guy said it sounded like bomb went off and he was still some distance away from the room. No one was injured since all of the conditions and preparations of using a turbopump had been followed.
In general (at least when using ion traps as second stage for UHV), you want to to pump down to ~10^-4 torr. There's 100% going to be a pressure transducer (probably based off a Penning gauge) in the turbo molecular out to that main board. Normally I'd tell you to probe until you find a signal that varies between 4mA and 20mA but since it looks like that whole pump assembly was designed in house (rather than just calling up Edwards and saying "hey we need $n units of X that can get us down to 10^-8 torr), I doubt they're using the ol' sensor standards. Look for ADC's?
Cleanliness is of major imoprtance when working with vacuum equipment, especially when it comes to high vacuum and mass spectrometers. ALWAYS use clean gloves, and touch as little as possible inside, yes the outgassing of your fingerprints is a thing in these machines. Also, be very careful with sharp tools like screwdrivers, it's very easy to scratch a sealing surface. And regarding the backing pump, i'd pull it apart and clean it and give it some new oil. It only needs a litre or so of oil, and a good mineral based vacuum oil shouldn't cost you more than 20'ish dollars.
Maybe I missed something but it looks like there is a considerable amount of rough "as cast" aluminum surface in the vacuum area. Why was is OK to leave that un-machined but then everything else in there is super polished - even other parts that look like they would be at ground potential?
You need turbo or diffusion pump. Look at Ron Soyland's videos - he makes vacuum tubes in his home and he describes how and what you need. His website: tubecrafter.com
Thanks for the link. I have went through the webpage, and it seems that making tubes is a really expensive hobby. Plus I do not even want to think of the cost of the cathode and the gather materials which are not even mentioned in the pages, but I definitely know for the cathode they use special doped materials with low work function.
That is what I was afraid of, that one also needs a fancy pump in addition to an oil pump. It would be still nice though to be able to make some good vacuum tubes at home. Btw, regarding these great scores, do you buy them from Ebay? Here in German speaking countries there is never something nice on Ebay. Even basic ICs one need to order from the UK. No wonder the UK is leaving the EU, since the EU is decades behind when it comes to trading.
If you ever go to Mars, be sure to pack your mass spectrometer. Should you become stranded without hope of rescue, just start to disassemble it and in no time, some bloke will appear to tell you you're doin it wrong.
You're pumping against too many leaks given the roughing pump you have. To have pumped it down I would have made sure to seal all the possible inlets, He, N2 and the sample inlet. If the vacuum hose was the age of the mass spec, I'd be surprised if it did not have some cracks. Any replacement hose needs to be selected carefully since most commonly available vacuum hose may look the same as what would have been on the instrument, but is most likely too porous. Using good connecting hardware such as the KF style metal ring supported o-ring and clamp is best (that is what the flange s on the Edward's gauge). I would not suggest any sort of grease PTFE or otherwise, it will out-gas. Also to that point, unless you gave it a real good cleaning after the initial teardown and inspection, it looks like nearly every surface in the chamber will have been contaminated after handling with unloved hands, touching with tools that have oil or grease residue, etc. If your Edwards E2M5 was in good health you may still have had difficulty pumping it down without excessive load on the turbo. an E2M30 or E2M28 is what you want for that. Truly impressive how you popped the top and basically correctly identified and described all the components and connections. Most all mass spec experts and even those how use instruments like that one would not be able to get close. I think if you hadn't completely disassembled it you could have gotten it operational and had some fun making ions out of anything that struck your fancy, or sold it for a good profit. The main board is very large on that one and unwieldy for sure but it is common practice on these type of instruments. I think maybe the idea is to make field service easier. On this and similar instruments there is a lot of information you can get from the LEDs and from measuring voltage and resistance at one or more of the dozens of test points. Plus the various subsystems are grouped together in a logical order.
Your piping seems to be the dodgiest part. I'd try to test the vacuum just with the pipes between pump and sensor. I can't believe there's anything more leaky than that.
I have never seen sputtering in mass spectrometers. Since they come with a glass top nowadays that would be easy to see. All ions are confined and can only interact with anything at very specific points.
The craftsmanship on that thing is incredible. I could look at it all day.
The black stuff inside the detector is indeed a resistive layer. It provides voltage gradient along the length of the electron multiplier horn's wall. It's basically a standard electron/photomultiplier-like construction, except there are no discrete dynodes (each having a slightly higher potential than the previous one), the dynode is distributed in the form of a continuous voltage divider. The horn shape and field gradient makes the electrons travel to the centre of the detector, knocking off more and more electrons as they collide with the walls on their way in. The detector is mounted at an angle relative to the ion stream, probably because there is a conversion dynode, from which the electrons are emitted when it's hit by a positive ion. This prolongs the life of the detector, since there are no ions reaching it (and eventually depositing on the surface).
I was gonna explain this too...…….but you did a great job of it...…...Great Stuff !
Hi, i have experience in such high vacuum systems. In your setup are many things that are not quite right. But first yes your rotary pump has reasonable size for your system. One source of error could be the oil in the forevacuum pump which is outgasing or the grease in the prcision bearings of the turbopump is not warm and the pump needs longer to reach its operating speed (some computerbased controler have a special mode for this scenario). And you should use vacuum grease (e.g. PTFE) to get a better seal ob the o rings.
It is very complicated to take a close look at all facetts in a such short time because there are far to many. If you want i could try to answer your questions and give you tips.
This thing shown around 10:35 is *_ion trap_* - arguably the most important part of this machine, and everything else is there just to make this thing do its job. :)
Okay, I know how a mass spectrometer works, it's quite easy in principle. But that you need something like THIS to actually perform the experiment is impressive!
I always find it fascinating to see how huge and macro-scale the machining and shaping of devices have to be in order to operate properly on the microscopic scale. The majority of modern state-of-the-art electronics is unbelievably tiny, so it's really interesting to see something that's so big and physical in order to get super precise measurements.
you might want to get with the guy from applied science. he has experience with rigging turbo pumps to things they weren't designed for.
Man this thing is insanely interesting. This sort of stuff is way above my head. I can follow most of the EE/CE stuff (I would hope so at least, as that's what I'm studying), but man, this sort of engineering is just on another level.
Thanks for taking this sort of stuff to bits and sharing it with us all (and testing it out and all that)!
I really love these industrial component teardowns. I've been watching some of your older videos and I wish you could do these more often. Keep up the good work!
With a big enough leak (which you may very well have), squirting solvents like isopropyl alcohol on points you think might be leaking will normally cause a spike in the pressure that you can see with the gauge. If the leak is really huge you may see the pressure go down with isopropyl alcohol, as the solvent freezes into the gap, and then you'll see it rise. Good way to detect leaks because it doesn't depend on absolute accuracy of the gauge.
Yeah isopropanol leak detection …………………..Im an EX...…...vg scientific engineer...….I would love to be doing this vid,it takes me back
Jeez, that thing is beautiful inside!
I would venture to say with at least a semi-straight face that for most of your previous videos I had a decent understanding of what it is you're showing us, but in this case I am just hopelessly lost. Looks nice, though!
Yep - didn't really get any understanding of any of the kit being torn down. "Here's a thing, here's another thing". - would be cool to get a bit more system description, not just components. :)
If you want more, read the tech manual linked in the description.
Thanks! Just as a mention on the side, you and Ben are far and away my favorite content creators on youtube. So, thanks for that, also :P
Holy crap, this is the non-RF black magic! Thanks for the peek inside of such an unusual instrument!
That pump gurgling is the key Mike, tune it by ear - it sounds it's pulling a significant volume of air. When it's pumped down it will thrum, and there will be a very fine snapping sound, like when you had the meter jammed up against it. I'm unsure about the o-rings too - from experience at only 1-6 torr they can be a problem if they're not done right, properly seated and or greased etc... but if your backing pump can pull 1--4 torr and the turbo pump works then you should get that green light! +NanoCottage, service the pump too if needs be, get proper oil in there and let 'er rip :)
Some incredible engineering and machining in this thing. Now i feel compelled to research turbo pumps
No kidding, those domed polished reflector parts had to be incredibly expensive.
Incredible really.
Two vids out in two days!
Two great too.
Cheers Mike ya legend!
Not a clue what all that stuff inside the vacuum chamber was, but it was definitely cool as hell! I think you need an industrial grade vacuum pump, possibly a dual stage or so, for this to operate. Just guessing.
When did you last change the oil in the vacuum pump? To get the best performance the oil needs to be fresh as it provides the seal on the rotary vanes. Having new oil makes a huge difference, refrigeration guys change their oil often, like once a week. Some pump manufacturers recommend new oil every job for peak performance. If it's still not good enough after an oil change then you can get a rebuild kit with new seals and vanes for not much money.
er, never!
I have no idea what state the pump was in when I got it maybe 15 years ago, and it;s hardly been used since. Quite likely it's not in good condition!
Great video Mike, thanks for sharing.
A decent low vapor pressure rotary vane pump oil does not come cheap. The closest budget option I found for mine was Anderol 555, i think I got it shipped from UK.
Not knowing the type of the oil you currently have in the pump it's definitely worth flushing it out completely and refilling with something new. A liter of new oil made a huge difference on my Edwards pump.
cheers,
Val
Yeah, something doesn't sound right. The gurgle sound on the exhaust of the RV pump I use goes away once its got the system pumped down. Plus the overheating or reduced vacuum when its warm is another sign pointing at the oil. A RV pump should be able to run indefinitely below a certain pressure. (10Torr on the one I use, allowed upto 10 minutes of use above 10 Torr). If the oil is contaminated with something with a high boiling point (water, other solvents etc) as it tries to expand the chamber, the contaminate evaporates out of the oil, then as it compresses it condenses and goes back into the oil.
Sounds like an oil change would be a good starting point to me.
the overheating is also an indicator of massive leaks. the pump is no more than a compressor in reverse. It doesn't "suck" but it compresses gas to atmospheric pressure. So leaks = heat.
That being said, that freefloating o-ring or the silicon sheet are not good enough at the level of vacuum the RV can go to. but of course, if the oil is toast it won't go very far anyway.
For the turbo, my guess is it's triggering on high motor current. Usually the machine it's mounted on would stop the turbo via the interface by monitoring the backing pump pressure, or flat out not allow starting it if the vacuum behind the turbo is not low enough.
But here, it's possible the pump driver doesn't like running at high current for too long.
If the inlet of the pump had been open to atmosphere the oil will accumulate moisture and the lowest base pressure will be much higher then without any moisture in the oil, and this especially noticeably when the pump starts to heats up. To get the moisture out from the oil you should block the input, preferably with a vacuum gauge, and then let the pump run and be really warmed up and let it run until the vacuum is at its lowest with should be at best 0.000 Torr or lower.
The pump also have a gas ballast that can be open a little if there is a lot of moisture in the oil and/or the last vacuum take a long time to go down at the end. A fresh good 2-stage pump will go down to 0.000 Torr pretty much no matter how old the oil is, so have that as a number to aim for. I have a pump from -97 that never have had a oil change and that goes don to 0.000 Torr. never the less.
An oil change is more to get fresh additives and remove dust, debris from wear and tare and what not. But if the pump rums very hot due to high load the oil could get burned and get broken down and thus loose some of its characteristics and then its a good idea to change it.
16:00 I spot a bigclive safe box as the main switch
i have no idea what this does or how it works, but it looks interesting! Also i can really apreciate all the machining that went into that thing.
Machining on this is just a joy to watch.
i believe what happens in the octepolar trap is that you get a standing sinusoidal electric field which is frequency variable. that ions that enter it only get to the end if the frequency of the electric field is the same as that of the natural frequency of the ions, if they don't match then they are lost out of the trap (natural frequancy is based on mass) so the system sweeps the frequency of the electric field and to separate ions by mass. please correct me if i am wrong but i believe that is how it works.
9:51 reminds me of that General Electric turboencabulator design.
You should try using valve between your 'regular' vacuum pump and the rest of the system. Otherwise these two pumps just 'fight' each other and you get stuck with the pressure of the weaker one.
That spike thing in the of the sample input, looks a lot like a Field Emission Electron gun.
Neat bit of kit! Love these tear downs.
look at that physics package
i came
man that whole thing just oozes expensive
cool! i was waiting for part 2, great bedtime vid for today, thanks!
Gorgeous stuff, seems an incredible shame to scrap this sort of stuff. I really like that you actually try to get some of this gear going. I would live some high vacuum stuff but there is far more to it than just connecting a few pipes thats for sure
The thing @5:40 is the electron multiplier, not the metal part.
@7:50, the rpm of turbopump is way higher than 6k rpm, it should be around 50k rpm.
You will need a roughing pump which is more powerful that what you have there. I believe you need a flow rate at around 300 lpm.
I said 60Krpm. I think the E2M5 should be OK, but could be mine is under-performing.
Well, I might miss the "60k" part. Sorry about that.
In our lab, we used two Edward 30 pumps for a similar model. Well, I guess working with two roughing pumps is a bit crazy. But I guess you may need a better pump that the ultimate pressure is in range of 1e-3 to 1e-4.
Anyway, thanks for the video! It is really cool to see you holding an ion trap in person without gloves. J
That's not true. Pfeiffer specifies a fore vacuum of < 10 mbar for this type of pump.
In ultra high vacuum applications we typically use dry scroll pumps (you don't want to have oil anywhere near a UHV system) which reach a final pressure of about 0.1 mbar. This is easily sufficient to get to 1E-8 mbar with a decent turbo and some baking.
I don't see a conflict between your point and mine. Specifically, the hardware manual of Thermo LCQ suggested a forepump that meets 650 L/min displacement, which can reach the pressure around 1 Torr (~1.3 mbar, I believe).
However, there are something I respectfully disagree. Some mass analyzers (e.g. 3D trap, linear trap, quad, etc.) do not require ultrahigh vacuum (UHV). The pressure of ~1 mTorr (1e-3 mBar) is sufficient for such applications. In contrast, most of the high-resolution instruments require "true" UHV (e.g. Thermo Orbitrap requires 1e-10 mbar). I have seen people using dry scroll pumps (I have used a few before, e.g. SH110, Triscroll 300, etc.); but quite a lot of them are atmospheric scientists. In analytical labs, I am not quite aware of the use of dry scroll pumps, which might due to the difficulties during servicing.
It would be nice if you may share your google scholar profile, which would be very helpful for me to get to know your works. :)
Scroll pumps are used pretty often. In fact i see dry pumps more often than oil sealed ones in analytical labs now (scrolls, claw/roots mainly).
Scroll pumps easily reach 10-3Torr ranges or less and require less (and easier) maintenance than vane pumps.. and no oil backstreaming into the system. That's one of the main reason they are preffered in high vacuum for some applications.
Downside is they don't tolerate well corrosive gases with is usually not a problem for mass spectrometers, electron microscopes and other mostly clean applications.
Very cool!! Be careful about just cutting power to that turbo, those mas spec units typically use active mag bearings. Maybe its not a big deal until youre up to full speed... and I'd change your pump oil out, sounds like youve got some disolved contaminants that have high vapor pressure at higher temperature in it.
Love your videos Mike.....but as someone who works with mass spec's for a living I can honestly say I have never cringed so much when watching a video as I did with this one :)
"Ohh look at this nice and shiny, probably electropolished super clean surface" *Touch*
those parts would make great background props for a sci fi film.
hell that's one way to vent VG Scientific will be spinning in its grave " RIP best firm I ever worked for
The inside of this is like some alien technology.
0:44 Cool.... it'd fun to attach a little getter inside and seal off the top of that triode and pump out the air and set off the getter and then seal the vacuum tip and make an awesome regenerative shortwave receiver or QRP transmitter or something with it. :-)
I deleted my other comment, teslaTrooper.... it was meant for the genome sequencer video. :-)
A fine instrument. To be sure. To be sure.
Any suggestions for a high vacuum shutoff/relief/diverter valve? Something that will work well in the milli-torr range? At least with the Spectral instruments cameras there is an electro-mechanical valve you can close before shutting off the pump. It's a bitch to pull the hose off with a vacuum remaining in the tubing once the valve is closed, though.
20:23 You might know this by now but that is very dangerous to do above a turbo pump. The o-ring over the pump can be is sucked in a little so it creates a big opening all of a sudden and that huge inrush of air would make the wings on the turbopump to go upwards and hit everything in it and at that point I would not want to be even remotely close to it since it will go off lik a bomb.
Seen this happen. They blow spectacularly ☹😲💩
@@christopherleubner6633 You have? Dam, luckily I have not but seen many and they are absolutely totalled afterwards. One guy I know that work at a university said that he was at the corridor where all the the teams did experiments and he was at his small office with the door opened. A turbopump little bigger that the one in the video totally sized after a small nut fell into it and the guy said it sounded like bomb went off and he was still some distance away from the room. No one was injured since all of the conditions and preparations of using a turbopump had been followed.
In general (at least when using ion traps as second stage for UHV), you want to to pump down to ~10^-4 torr. There's 100% going to be a pressure transducer (probably based off a Penning gauge) in the turbo molecular out to that main board. Normally I'd tell you to probe until you find a signal that varies between 4mA and 20mA but since it looks like that whole pump assembly was designed in house (rather than just calling up Edwards and saying "hey we need $n units of X that can get us down to 10^-8 torr), I doubt they're using the ol' sensor standards. Look for ADC's?
1:25 aren't those three grub screws in the periphery for active locking?
I a sure that part at 9:45 is in the centre of the tardis 🙂
You have probably fixed it by now but your rotary pump is leaking somewhere I can hear it, if the pump is happy it won't gurgle like that.
Cleanliness is of major imoprtance when working with vacuum equipment, especially when it comes to high vacuum and mass spectrometers. ALWAYS use clean gloves, and touch as little as possible inside, yes the outgassing of your fingerprints is a thing in these machines. Also, be very careful with sharp tools like screwdrivers, it's very easy to scratch a sealing surface.
And regarding the backing pump, i'd pull it apart and clean it and give it some new oil. It only needs a litre or so of oil, and a good mineral based vacuum oil shouldn't cost you more than 20'ish dollars.
Maybe I missed something but it looks like there is a considerable amount of rough "as cast" aluminum surface in the vacuum area. Why was is OK to leave that un-machined but then everything else in there is super polished - even other parts that look like they would be at ground potential?
Don't see a need to machine it - I expect they probably acid-clean it though.
For making electron tubes (audio triodes at home) does one need such a turbo pump, or the normal oil pump is good enough?
You need turbo or diffusion pump. Look at Ron Soyland's videos - he makes vacuum tubes in his home and he describes how and what you need. His website: tubecrafter.com
You need a turbo pump for that sort of thing.
Thanks for the link. I have went through the webpage, and it seems that making tubes is a really expensive hobby. Plus I do not even want to think of the cost of the cathode and the gather materials which are not even mentioned in the pages, but I definitely know for the cathode they use special doped materials with low work function.
That is what I was afraid of, that one also needs a fancy pump in addition to an oil pump.
It would be still nice though to be able to make some good vacuum tubes at home.
Btw, regarding these great scores, do you buy them from Ebay? Here in German speaking countries there is never something nice on Ebay. Even basic ICs one need to order from the UK. No wonder the UK is leaving the EU, since the EU is decades behind when it comes to trading.
Donkey Learning IT That problem with eBay... never seen some good stuff.
Is that normal IDC ribbon cable in the RJ45? Never thought to do that. Neat.
No it's flat sheathed cable. However you can crimp 1mm pitch IDC ribbon cable into RJ connectors, which can be very useful.
Interesting, thanks, useful to know.
Huh. Today I learned. That's a neat idea.
If you ever go to Mars, be sure to pack your mass spectrometer. Should you become stranded without hope of rescue, just start to disassemble it and in no time, some bloke will appear to tell you you're doin it wrong.
OMG that's a lot of "black magic" happening inside that machine :)
You're pumping against too many leaks given the roughing pump you have. To have pumped it down I would have made sure to seal all the possible inlets, He, N2 and the sample inlet. If the vacuum hose was the age of the mass spec, I'd be surprised if it did not have some cracks. Any replacement hose needs to be selected carefully since most commonly available vacuum hose may look the same as what would have been on the instrument, but is most likely too porous. Using good connecting hardware such as the KF style metal ring supported o-ring and clamp is best (that is what the flange s on the Edward's gauge). I would not suggest any sort of grease PTFE or otherwise, it will out-gas. Also to that point, unless you gave it a real good cleaning after the initial teardown and inspection, it looks like nearly every surface in the chamber will have been contaminated after handling with unloved hands, touching with tools that have oil or grease residue, etc. If your Edwards E2M5 was in good health you may still have had difficulty pumping it down without excessive load on the turbo. an E2M30 or E2M28 is what you want for that. Truly impressive how you popped the top and basically correctly identified and described all the components and connections. Most all mass spec experts and even those how use instruments like that one would not be able to get close. I think if you hadn't completely disassembled it you could have gotten it operational and had some fun making ions out of anything that struck your fancy, or sold it for a good profit. The main board is very large on that one and unwieldy for sure but it is common practice on these type of instruments. I think maybe the idea is to make field service easier. On this and similar instruments there is a lot of information you can get from the LEDs and from measuring voltage and resistance at one or more of the dozens of test points. Plus the various subsystems are grouped together in a logical order.
Your piping seems to be the dodgiest part. I'd try to test the vacuum just with the pipes between pump and sensor. I can't believe there's anything more leaky than that.
Yeah, if you let air in while turbo pumps are running they will fail.
Wear some gloves if your gonna touch something that's going to be in high vacuum. You don't want your skin oil to gas out in high vacuum.
A spinning fan with no air?
warm tube video )
How much did it cost???
Has this thing ever been used? i see zero indication of sputtering and that's really something that's unavoidable in something like this
I have never seen sputtering in mass spectrometers. Since they come with a glass top nowadays that would be easy to see. All ions are confined and can only interact with anything at very specific points.
If sputtering is seen then something is terrible wrong.
No sputtering takes place in mass spectrometers. However there are visible ion burns on the inner ring of the trap which denote heavy use
Perhaps I should be a bit ashamed to say this, but to me, the name "Finnigan Mass Spectrometer" sounds a bit oxymoronic.