Simple turbo pump controller

Every fusor and fusion system seems to need a vacuum. This area is for detailed discussion of vacuum systems, materials, gauging, etc. related to fusor or fusion research.
User avatar
Alexi Hammond
Posts: 25
Joined: Sat Dec 04, 2004 1:26 pm
Real name: Alexi Hammond

Simple turbo pump controller

Post by Alexi Hammond »

The pump:

Recently we bought a couple of cheap TPH-055 turbo pumps from eBay (Item number: 370455353773). The motor in TPH-055 is single phase, two poles, permanent magnet; therefore the pump spins at the drive frequency. The winding is center-tapped, which makes it easy to drive it with only two transistors. I designed simple controller for those pumps.

The controller:

During powering up capacitor C4 resets D-trigger D2 and the red LED turns on (see the diagram). If signal from the Hall sensor (pin G) is high then transistor T1 turns on and current flows from the 28V power supply through 10 Ohm current limiting resistor and through A-B halve of the motor winding. If signal from the Hall sensor is low then T2 turns on and current flows through A-C halve of the motor winding. At this point the pump starts rotating and spins faster and faster.

Each falling edge of the Hall sensor signal triggers the 555 timer, which generates a single pulse. Each rising edge of the Hall sensor signal locks the state of 555 timer output in the trigger D2. When the halve-period of the sensor signal becomes shorter then duration of the pulse from timer, then the trigger D2 changes its state. Green LED turns on and both transistors remain off until halve-period again becomes longer then the timer pulse and the trigger changes its state back.

Normal speed of the TPH-055 turbo pump is 90 kRPM or 1500 Hz. The resistor R1 sets pump’s final speed. R1 should be either adjusted to make 333 microsecond pulses (halve-period of 1500Hz) at pin 3 of the 555 timer or could be set to turn on the green LED when the pump reaches desired frequency.

Testing:

The pump (right out of the box: no cleaning or oil change was done) was connected to the mechanical pump and to the small chamber with an ion gauge inside. The controller was assembled on a breadboard and the wires were connected to the pump with push-on connectors. The Hall sensor signal frequency was measured with a multimeter and an oscilloscope. After power-on the pump reached 1 kHz in about 5.5 minutes, and it took about 14 minutes to reach 1.5 kHz. The potentiometer R1 was adjusted to hold 1.5 kHz frequency. Pressure in the chamber at this point dropped down to 6e-6 torr. During a few days test run the frequency remained within +/- 1 Hz of 1.5 kHz, and the pressure in the chamber dropped another order of magnitude.

Alex
Attachments
TP controller diagram 1.PNG
TPH055 controller 001.jpg
TPH055 controller 004.jpg
TPH055 controller 008.jpg
TPH055 controller 003.jpg
TPH055 controller 009.jpg
TPH055 controller 017.jpg
User avatar
Carl Willis
Posts: 2841
Joined: Thu Jul 26, 2001 7:33 pm
Real name: Carl Willis
Location: Albuquerque, New Mexico, USA
Contact:

Re: Simple turbo pump controller

Post by Carl Willis »

Brilliant addition! Thanks very much for this report. The TPH/TPU pumps are always on eBay at attractive prices, but seldom come with a controller. This kind of homebrewing might just revolutionize access to cheap turbo pumps for hobbyists.

-Carl
Carl Willis
http://carlwillis.wordpress.com/
TEL: +1-505-412-3277
John Futter
Posts: 1848
Joined: Wed Apr 21, 2004 10:29 pm
Real name: John Futter
Contact:

Re: Simple turbo pump controller

Post by John Futter »

Alexi
Very good I think you can lose the series resistor or seriously reduce it and the unit will speed up faster.


For all the rest of you out there most turbos are three phase and I have made a universal controller to make them go. Because they are three phase the rotor tracks the drive signal and when nearly in sync the current drops dramatically details here http://www.coultersmithing.com/forums/v ... ?f=12&t=83.
Note that two phase units do not have hard coupling and to know where the rotor is you have to use the inbuilt hall sensor or put one in yourself


Now there is a complete solution for all pumps out there orphaned from their controllers
DaveC
Posts: 2346
Joined: Sat Jun 30, 2001 1:13 am
Real name:

Re: Simple turbo pump controller

Post by DaveC »

Alex -

Sweet! One of those things we all think about doing, "someday"... and you did it.

The small load of the little pumps, when finally at base pressures, means the drive hardware, doesn't have to be massive.

Nicely done. Thanks for sharing with us.

Dave Cooper
David D Speck MD
Posts: 129
Joined: Sat Jan 26, 2008 2:05 pm
Real name: David D. Speck MD
Location: Auburn, NY

Re: Simple turbo pump controller

Post by David D Speck MD »

Alex,

Thanks for posting the controller schematic. Based on the ease of constructing this controller, I bought one of the pumps for myself and a friend.

Three questions:

Does the pump "know" how to always start up in the correct rotational direction? With only 2 phase drive, it seems like it could go either way, unless it had a shading coil built into the stator to insure proper rotation.

Do you have any info on proper lubrication and oil change procedures for the pump?

Do you have the designations for the pump flanges? All of my equipment is Conflat style, so I'll have to scare up a transition connector.

Thanks in advance.

Dave Speck
User avatar
Chris Bradley
Posts: 2930
Joined: Fri May 02, 2008 7:05 am
Real name:

Re: Simple turbo pump controller

Post by Chris Bradley »

Do these TPH pumps have a passive magnetic bearing (upper bearing)?
User avatar
Alexi Hammond
Posts: 25
Joined: Sat Dec 04, 2004 1:26 pm
Real name: Alexi Hammond

Re: Simple turbo pump controller

Post by Alexi Hammond »

David,

The pump does "know" how to always start up in the correct rotational direction. Sense of rotation is determined by the phase relationship between the hall sensor signal and the current through the winding. If the hall signal and the current are in phase then the motor spins in one direction, if they are 180 degrees out of phase, then the motor spins in opposite direction. No shading coil is needed. If you switch wires B and C, then the pump will try to spin in the wrong direction. You can see how this type of motor works if you draw a picture of a magnet inside a coil and 90 degrees shifted hall sensor.
Obviously, there are two dead spots from which motor will not start: where the magnet is perfectly aligned with the coil’s axis. The pump is designed to always stop between those dead spots.

There is some info on lubrication, oil change and cleaning in the manual (attached).

The pump flanges are: DN 63 ISO-KF and DN 16 ISO-KF.

Chris,

Yes, those TPH pumps do have a passive magnetic bearing (upper bearing).

Alex
Attachments
tphu060.pdf
(326.52 KiB) Downloaded 781 times
User avatar
Chris Bradley
Posts: 2930
Joined: Fri May 02, 2008 7:05 am
Real name:

Re: Simple turbo pump controller

Post by Chris Bradley »

I was wondering, though, whether that was an active magnetic bearing. I have a TPH (not sure of the number) and it has a zillion pins on it, so I figured those were some wonderfully over-complicated means for active control. I suppose if it has one fixed lower bearing, retaining it axially, then I guess that probably means the upper can be passive (I seem to recall that passive mag bearings can only retain in 2 axes).
chad ramey
Posts: 107
Joined: Tue Apr 28, 2009 10:45 pm
Real name: Chad Ramey
Location: Georgia
Contact:

Re: Simple turbo pump controller

Post by chad ramey »

To the people who ordered these and have already received one:
Did the pump come with the ISO centering ring that is pictured in the listing?

I bought one of the pumps and need to know if I need to buy a centering ring.

Thanks,
Cahd
richnormand
Posts: 307
Joined: Sat Aug 27, 2005 8:30 am
Real name: rich normand

Re: Simple turbo pump controller

Post by richnormand »

"Did the pump come with the ISO centering ring that is pictured in the listing?"

Mine did not.
chad ramey
Posts: 107
Joined: Tue Apr 28, 2009 10:45 pm
Real name: Chad Ramey
Location: Georgia
Contact:

Re: Simple turbo pump controller

Post by chad ramey »

Thanks, I'll go ahead and order one from lesker. Have you been able to test your's yet?
richnormand
Posts: 307
Joined: Sat Aug 27, 2005 8:30 am
Real name: rich normand

Re: Simple turbo pump controller

Post by richnormand »

Not really. Got it to turn with a two-phase sine signal from my audio generator into a 100W strereo amplifier as a quick test after downloading the specs sheet. I got the vacuum chamber and coupling from e-bay during the summer. Already have the backup pump. Now waiting for some SS hoses and two KF25 valves. All the stuff is in a box at this time. The "honey-do" list takes priority..... Please post cost and ## for the ring once you find it.
Cheers and thanks.
chad ramey
Posts: 107
Joined: Tue Apr 28, 2009 10:45 pm
Real name: Chad Ramey
Location: Georgia
Contact:

Re: Simple turbo pump controller

Post by chad ramey »

Thanks for the information. Sounds promising, I wish you the best of luck with getting it running. I have full intentions of having mine running on my reactor by the third weekend of september (hopefully earlier than thst).

I have the ISO63 centering rings on several sites:
Lesker-
http://www.lesker.com/newweb/flanges/ha ... fm?pgid=al
$26.00
LDS Vacuum Shopper-
http://vacuumshopper.stores.yahoo.net/lfalcenrinwv.html
$20.00
MDC-
http://www.mdcvacuum.com/urd/uniface.ur ... ay?1.a.1.m
$29.00

I'm going to talk to a friend at Quintronix tomorrow to see if they have any available, and if they do I'll let you know.

-Chad
DSchultz
Posts: 4
Joined: Tue Aug 10, 2010 2:07 am
Real name:

Re: Simple turbo pump controller

Post by DSchultz »

I found a surplus source for the RFP2N08L FETs. I need to buy 20 of them to make the minimum order. Does anybody else need these FETs or have you all found some other FET that works as well without being un-obtanium?

I am going through Amphenol and ITT Cannon catalogs to try to identify the power connector. No Luck so far.
richnormand
Posts: 307
Joined: Sat Aug 27, 2005 8:30 am
Real name: rich normand

Re: Simple turbo pump controller

Post by richnormand »

Hi Dan.

You might be interested at the NTE line of substitutes. They list the NTE2382 as a replacement for both the RFP2N08 and the RFP2N10.... The specs are identical except for a 100V rating instead of 80. I ordered 5 NTE2382 from Allied Electronics ( $3.95 CAD each) since they were in stock. No min orders either. They have been working for almost one month and no issues. The NTE line is very popular for all sorts of repairs and it has a very extensive range of "equivalent" components. Most local electronics repair stores will carry them but they are slightly more expensive than ordering the exact part (if still available) from the large outfits such as Mouser, Allied, Digikey and such.

I initially replaced the 555 with a phase locked loop for smoother transitions and went back to the original 555 as there were no real improvement. I did add a switched selection of both full speed and standby (60%), each with its own 10 turn pot. Also for about $12 on eBay I got a small four-digit display frequency module. This gives me the rotational speed, 1500 Hz full and 900Hz standby. I can now follow the pump in ramp up or down to a full stop.

For power a 24V OEM surplus supply can be internally (mis)adjusted to provide the 28V. I included a separate switch for that too. This allows keeping the 5V to the logic electronics during ramp down. The motor back emf at full speed will heat the MOSFETs too much unless you keep the 5V alive to keep switching (with the 28V off).

Finally I noticed that in very rare cases the pump will not start and one FET is full on and the other full off. Not a good scenario. A small twist of the pump or a restart will cure that. I guess the rotor can stop in a no-man-land position. The freq display solves that issue since I can immediately see the startup upon switching the 28V on. If it stays static it is just a matter of turning it off and reset the 5V and then the 28V.

As far as the connector is concerned I ended up soldering a wire to each pin and fan them to a strip junction block with a label for each function. Nice that the full specs for the pump are available online too.

By the way I am not using the fan that came with the pumps either. They both barely get warm after several hours of running. Maybe if I flow lots of gas through them.... but that is for later.

EDIT: See note by Alexi further down in the posts about the difference between the RFP2N08 and the RFP2N08L. The NTE sub does work well for me however.
DSchultz
Posts: 4
Joined: Tue Aug 10, 2010 2:07 am
Real name:

Re: Simple turbo pump controller

Post by DSchultz »

I had a negative experience several years ago with a NTE "universal replacement" FET which was not so equivalent. Since then I try to find the original part whenever possible. I have too many projects going than to try to re-engineer what others have already proven to work.

What are you using for a power connector? Do you just push the wires onto the pins one at a time?

Dan
richnormand
Posts: 307
Joined: Sat Aug 27, 2005 8:30 am
Real name: rich normand

Re: Simple turbo pump controller

Post by richnormand »

I agree with you Dan,

Note that I used "equivalent" in quotation (as with a big grain of salt) as YMMV but for this applications they seem to work for me and others may find the NTEs easier to obtain than chasing the exact part. Been using the substitutes for more than 30 years now from the time they were RCA SK series, Sylvania ECG subs and now NTE and, yes, you need some creative empathy from time to time as they are not the real McCoy. Initially made for people fixing TVs and such at the time.

I always try to find the exact match when possible too. However a friend of mine that is in the electronic component failure analysis and counterfeit or rebranded detection business has a mitigated opinion about getting components from non-authorized dealers, in particular resellers of older parts no longer available. What you see may not be what you think you have. Not much of an issue here but not funny in medical, aerospace or other safety related applications. That being said, I have used resellers of obsolete or hard-to--get components several times with great success. It all depends on the price and minimum order. Or... how desperate I was to get the part vs redesigning the whole mess with modern components!

For the connector I simply tinned the end of the wires and tinned each individual pin on the pump connector. Laid the two side by side and I then wetted each together one by one and heatshrinked each and then the lot together with an RTV silicone core. Dont overheat as I did not check how the wires are attached to the pins inside the pump body. Quite sturdy and easy to dismantle when or if I manage to find the mating connector at a decent price. I have two other pumps as spares. Did not want to invest much money until I got a feel. I will next change the oil wick but these look like they were an excellent deal.

Cheers
gpecke
Posts: 2
Joined: Fri Jan 07, 2011 4:39 pm
Real name:
Contact:

Re: Simple turbo pump controller

Post by gpecke »

Dear Alexi,

Thank you very much for your post. I have bought two pumps on the strength of your information.

I am glad you mentioned the fine black muck, which needed very careful removal.

I have already bought two Alcatel/Adixen MDP 5011pumps, also from E-Bay. These work very well. I have posted some details on how to drive them, and will try the TPH055 on the same inverter. The motor is very similar in principle to that of the TPH055, except it has no winding center-tap, so your simple inverter would not work without an external transformer.

The MDP pumps have a much lower rate , but a much higher ratio and exhaust pressure. I will try using one as a clean scavenger and fore-pump to the TPH055, and hopefully retire my noxious Hg vapor pump, cooling water and cold traps once and for all.
User avatar
Alexi Hammond
Posts: 25
Joined: Sat Dec 04, 2004 1:26 pm
Real name: Alexi Hammond

Re: Simple turbo pump controller

Post by Alexi Hammond »

Rich,

There is big difference between the RFP2N08L and RFP2N08 transistors. The L sands for "use with logic level driving sources". The RFP2N08L can provide 2-3 amps drain current at 4V gate voltage and the RFP2N08 - only 0.1-0.3 amps.

Any transistor, which can provide at least 2 amps drain current at 4V gate voltage, will work fine in my controller circuit.

Alex
richnormand
Posts: 307
Joined: Sat Aug 27, 2005 8:30 am
Real name: rich normand

Re: Simple turbo pump controller

Post by richnormand »

Hi Alexi,

first of all many thanks for posting the circuit.

As mentionned in my previous post the only mods that I did (after trying the PLL and such) was to add a full speed and a standby mode switch. I am now waiting for a cheap 4 digit frequency counter (eBay $12) to monitor the pump speed. I was using my good HP frequency meter not only the adjust the pots but to monitor the pump so I figured it would be a good addition.

Here is the spec sheet of the NTE MOSFETs I have been using for over a month. Rock solid, no issues. Looks like they meet the L specs OK. There were cheap and in stock at Allied Electronics.

Looking at your photos: does your pump need the fan? Mine stays cool in the test setup so I removed it.

Best wishes for the New Year.

Edit1: You are right of course about the difference between the two types. Interesting thing is the search software for NTE does not list the L variant in my software version.
Edit2: here is a pix of the setup. The empty space on the board is for the freq module.
Edit3: I did add a note to my original post pointing to your post about the L variant difference. Thanks for catching that.
Attachments
P1094820 (Medium).JPG
nte2382.pdf
(26.03 KiB) Downloaded 583 times
User avatar
Alexi Hammond
Posts: 25
Joined: Sat Dec 04, 2004 1:26 pm
Real name: Alexi Hammond

Re: Simple turbo pump controller

Post by Alexi Hammond »

Hi Rich,

Yes, the pump gets barely warm after many hours of pumping. However, I left the fan on the pump anyway. The fan seems to be made for 208 or 240V AC. It runs on 120V AC very slowly, but that is enough to keep the pump cold.

Alex
richnormand
Posts: 307
Joined: Sat Aug 27, 2005 8:30 am
Real name: rich normand

Re: Simple turbo pump controller

Post by richnormand »

I received the eBay frequency meter and here it is.

Just wire the +5V, grounds and signal from pin 3 of the 555 timer in Alexi's controller. You can follow the pump ramp up and slowing down from reading its frequency (1500 Hz is 90,000 rpm). Not bad for $12 and free shipping.
Attachments
P1124821 (Medium).JPG
Richard Hester
Posts: 1519
Joined: Thu Jul 05, 2001 12:07 am
Real name:

Re: Simple turbo pump controller

Post by Richard Hester »

I may try a simple controller as well, as I have an Alcatel 5010 drag pump coming in the mail, as well as one of uncertain pedigree already in house. If anyone has a copy of the manual, for this one, I'd appreciate it. Otherwise, I'll try contacting Adixen/Alcatel.
AFW
Posts: 157
Joined: Sun Dec 10, 2006 5:51 pm
Real name:

Re: Simple turbo pump controller

Post by AFW »

Alexei
Thanks for the controller circuit. I notice that on the TPH 055 diagram there is a contact "D" that doesn't have a corresponding connector on the controller. Is this contact left floating? Also, contact "E" is labelled as 15V, but is connected to the +5V line of the controller. Is this a misprint in the TPH055 diagram?
UK purchasers of these pumps should remember that they will have to pay VAT, which, with postage, brings them to about £100. Still a bargain , though.

Tony Webb
richnormand
Posts: 307
Joined: Sat Aug 27, 2005 8:30 am
Real name: rich normand

Re: Simple turbo pump controller

Post by richnormand »

I think the D to F is a variable resistance as the temperature sensor. I did not connect anything to the D terminal. I'll monitor its value next time I ramp the pump and see what happens. It's most likely used with the OEM controller to ramp down or shut down the pump in case it is overheating .

Hall effect sensor for rotation (E) works fine with 5V in my setup, as stated in Alexi's diagram.

Alexi will probably know for sure.
Post Reply

Return to “Vacuum Technology (& FAQs)”