I have a Pfeiffer Balzers TPH 050, just the pump itself. No controller or cables. Do we have the equipment to test this device?
TM pumps without the right cables and controller are likely not much more than paperweights. You might be able to put them under vacuum and spin them with a 3 phase motor driver, but it is pretty easy to damage them without the controller logic that is designed to protect them.
You need the controller and a roughing pump ( to get you down to 10^-3 Torr so you can start it up. )
You’ll also have to close off the inlet end with a valve or chamber of some kind. Running one in open air is a Bad Thing as I recall. ( Not that that briefing was 35 years ago, so it might be a bit fuzzy… )
Yeah, a backing pump is definitely needed. As for the inlet, it does already have a proper cover and clamp.
1 Like
I understand this. My question was more around if any of the proper cabling existed at the space.
I’ve seen similar pumps on mass spectrometers. Where did this one come from? Might give us a clue on how to hook it up. We have a few vacuum pumps in the Science area that might be suitable as a roughing pump.
Sorry guys, not to derail this thread, but this is what my mind read when I first read the title. Still good for a laugh.
Turboencabulator.
Now I will flag my own post as off topic.
5 Likes
So from Kobin’s reply, it sounds like there are no TM pumps at the space, so unless a member has access to one of their own, this looks like a dead end.
From a practical matter, about a decade ago I had a conversation with a relative who had worked at a national lab and major university. His experience includes simple bench top helium leak detection systems with separate roughing and backing pumps and a mass spec, up to 2 MeV vandegraph positive ion accelerators with multiple ion sources, and hundreds of feet of beam line. My recollection is that each ion source had its own TM pump. The accelerator main line had two, and generally each experimental line at a different angle off the first magnet had at least one, and more commonly two pumps. So a fair number of them around the lab, even before considering other types of vacuum experiments. In his experience, at the time, the only time they had any interoperability between pumps and controllers was when they had ordered several at the same time, they could later swap controllers and pumps within that one lot. But even a pump ordered a year later, to the same spec and manufacturer would not be cross compatible. And more often than not, unless the pump died under warranty, the vendors couldn’t even source pumps or controllers to match existing equipment, or at least not for less than the cost of a new pump and controller.
I wish you luck on your quest, but unless the range of compatibility has improved, I fear it is a difficult task.
Here is the manual from Fusor.net
If I am reading this correctly, that pump has magnetic bearings, a 4 phase motor, and can turn up to 90,000 RPM.
The power supply looks like pretty old tech, but if you don’t know the voltage of the phases, you could really whack this thing. My company used to make an air bearing spinner that turned at 54,000. We started with 36Vdc, and commutated the phases, with a Laser and reflector on the shaft to feedback RPM. The power supply had to “ramp up” the spinner with ever increasing frequency pulses, until it reached 54K. They sounded like a tiny jet engine spinning up.
Yes, this would be an fairly involved endeavor, @festercluck and there is a fair chance that the 1st pump would get fried. If you can get the Power Supply, we can hook it up and test it in Science.
TCP40_TPH40_TSH40.pdf (2.9 MB)
Even 10 years ago, the TM pump controllers used a feedback loop on motor current. They would not accelerate past a configured motor power, so with a poor roughing pump down, or a weaker backing pump, they would spin up more slowly due to the extra load. I think the little 3 inch pump on the helium leak detector usually took over 5 min to get up to full speed, and that was with very oversized roughing/backing pumps.
Reading, thank you for this. I’m going to take some pictures to share. I’m not objected to reversing this thing, I bought it at auction for next to nothing. I have no assurances it even works
I get this needs a roughing pump, but are you getting at the idea it might be possible to safely test this thing by running a roughing pump at half power (not bringing down atmospheric as far as needed)?
Thanks for the pics.
The set-up would require a blank plate and O-ring to seal the inlet at the top. We would connect our Pirani gauge with a Klein flange lock on one side, and the other side goes to our roughing pump.
Before powering up the turbo pump, we vacuum down the system with the roughing pump, and confirm the pressure under 10 microns with our vacuum gauge. Then it is safe to spin-up the turbo pump. Those pumps are capable of vacuum approaching 10^-10 Torr. We have a Bayard Alpert gauge we can measure this with, in the event the motor spins-up, we can then consider connecting the BA detector.
Forgive me, but this is not my specialty. I believe I have a comparable piece of steel for the backing plate & an o-ring, but I’m not positive. What would be the specs on these to handle this kind of pressure?
Also, I just got a load of more toys in, including some laser equipment I believe mere mortals rarely get. Watch for a post soon if you’re willing to help me identify.
With vacuum systems, you are never dealing with more than 15 psi. But with TM pumps, you can get into ranges where the porosity in castings, or in welds can be a huge issue. Where what type of lubricant, or even the use of a lubricant can prevent getting the level of vacuum you want. Where simply the water hydrogen bonded to the inside of your vacuum line is a multi week headache until enough time passes, or you heat enough of your beam line enough to drive it off. Where accidentally letting oil from your backing pump flood back through your TM pump will ruin your day, or even week. Where some of the best gaskets are often single use copper disks indented by knife edges on the ends of your line sections.
I was rather expecting the magnetic bearing to be an active bearing, requiring active feedback from the controller. However, the manual is not as clear as I would like, but it implies that it is a permanent magnet bearing, as if it is a passive design, using really carful design of the eddy currents to be the feedback. So you potentially have dodged a huge bullet there. And the specs look like it is a lot more tolerant of atmospheric air than older pumps. I remember a lot of waiting for sufficient vacuum before turning them on, and knowledge that accidentally opening the wrong valve and venting atmospheric air into the system while they were still spinning was pretty much a death sentence to them. It looks like these were actually able to safely shut down without mechanical crashes if that happened.
The way the manual is set reinforces my opinion that these are low enough volume products, that are advancing in tech and design fast enough that there are still more models and versions in operation than the total operational count of the most common model in use.
And unfortunately, I only see descriptions of the pinouts to hook up the controller to the outside world. Nothing about pinouts between the motor and the controller. Or anything about why there are so many pins when a 3 phase motor only takes 3 wires, and a couple more for speed/rotational angle sensing if needed.
So I won’t disassemble to find the pinout. Gotcha.
Also: I have cleanroom sealed stainless piping now.
2 Likes
I can’t get explain all 19 pins, but looking at some similar drive units (TCP 015, 035, and 120) I think the following might be true.
A 4 phase motor, with a neutral, accounts for 5 wires.
Two isolated hall effect sensors account for 4-6 wires.
Two PTC resistors account for 2-4 wires.
1 or more wires may be used to identify the specific pump model.
The TCP 035 pin arrangement shows the use of a 3 wire hall effect sensor and a fixed resistor based on the pump ratings. It also establishes the idea of a wye connected motor.
The TCP 120 manual, which I think is the most similar to what this pump uses, mentions two hall effect sensors that don’t share a ground and can’t be measured together. The manual and Russell agree on a 4 phase motor, though the manual also calls out the 4 motor pins and the neutral they share in common. Also mentions a 90v operating voltage.
The THP 060 manual mentioned two PTC resistors as thermal protection, that were used by the controller to slowly stop the pump when it overheats.
1 Like
I would need to check, but I might have an identical model turbopump with controller. You’d still need a roughing pump and something to pull a vacuum on, did you have something in mind?
1 Like
I have found documentation of a DIY controller for this which handles some, but not all, functionality. While browsing that it became obvious the controller used with this pump is the TCP 040. I’ve yet to find documentation on that controller, but it’s the one most seem to use.
1 Like