MKS 901P Loadlock Transducer Fault

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.
Post Reply
User avatar
Emma Black
Posts: 198
Joined: Wed May 11, 2022 9:42 am
Real name: Emma Black

MKS 901P Loadlock Transducer Fault

Post by Emma Black »

I've been running the fusor the past week, after a couple of months of it sitting dormant. Today my 901p vacuum gauge stopped working. It was left switched on the entire time the fusor was not in use but was kept under vacuum. Side note: in about 6 weeks it only crept up to just over 100 microns. The gauge has been functioning fine all week until today when it died during a run.

Quick video of the gauges lights and the transducer clicking. Yes that noise is from the gauge. It does not read a pressure nor does the serial work. The power supply and display work fine with another gauge. I previously had it programmed to "pretend" to be one the Edwards gauges and has worked fine for several months.

https://www.youtube.com/watch?v=NvXN2Ckcr-s

I know a fair few people are using these 901p's and wonder if anybody has had one expire in the same way. Any prospect for a repair or is this one dead?
Jerry Biehler
Posts: 975
Joined: Tue Nov 24, 2009 3:08 am
Real name:
Location: Beaverton, OR

Re: MKS 901P Loadlock Transducer Fault

Post by Jerry Biehler »

That's looks like switching power supply overload to me. Possibly something inside the sensor shorted out and is overloading the internal or external power supply. I'd put a volt and ammeter on the supply and see of it is being overloaded. If not it's something internal. Maybe a shorted capacitor in one of the internal supplies. It might be repairable but without schematics that is going to be a pain. Its going to depend on what your electronics skill level is and what you time is worth to you.
User avatar
Emma Black
Posts: 198
Joined: Wed May 11, 2022 9:42 am
Real name: Emma Black

Re: MKS 901P Loadlock Transducer Fault

Post by Emma Black »

Thanks Jerry. As an update this turned out to be completely correct. The power supply works just fine with another gauge but cuts out as the current spikes, something internal in the gauge must have shorted. I have a few of these so doing a repair is likely a little project for a rainy day when there is nothing else to do.
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: MKS 901P Loadlock Transducer Fault

Post by Matt_Gibson »

I just had mine fail last night. I had some popping (shorts) from within the chamber so assume this contributed. The 901p flashed a red led a few times and then the output voltage went to near max despite being in the low microns.

-Matt
User avatar
Richard Hull
Moderator
Posts: 14976
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: MKS 901P Loadlock Transducer Fault

Post by Richard Hull »

Yes, severe external arcing, after the 5th time, killed my turbo controller a couple of years ago, as many will remember. Modern electronics just can't handle near mega watt peak pulses. Grounds in electronics become power supplies and inductors become little Tesla coils.

I ultimately repaired the electronics in my turbo controller. (couple of CMOS logic ICs). The key was ending the possibility of nasty external arcing. That fixed everything after I fixed the controller....

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
Andrew Seltzman
Posts: 814
Joined: Sun Feb 01, 2004 8:02 pm
Real name: Andrew Seltzman
Contact:

Re: MKS 901P Loadlock Transducer Fault

Post by Andrew Seltzman »

The MKS 901P gauges are sensitive to voltage spikes on the analog output line is connected. I've had this exact problem on my fusor where grid arcing kills the gauge, and output goes to full scale. The board with the CPU/sensors is the one that gets killed. Replacing this board fixes the gauge. The MKS 901P is an excellent gauge, arguably one of the best for fusors, but care needs to be taken with EMI protection.

Part of the problem on my fusor was using a single round copper ground wire instead of a low inductance grounding strap, the other problem is not having shielding between the vacuum vessel and the electronics (there is isolation with a ceramic break, but little shielding on my fusor) and no signal isolation. During a grid arc the output filter capacitors in the power supply form an LC circuit with the ground wire inductance causing the vacuum vessel to "ring" at around 5MHz at about 800v amplitude; the radiated EMI then couples into the surrounding conductors.

The power/relay/serial boards and display boards on display equipped 901P models seems unaffected by the arcs, it's always the sensor board that is killed. The analog output line is passed directly to the sensor board so likely has no protection from the power board. disconnecting the analog out line seams to protect the gauge from arcs. Scope captures of voltages during arcs below:


Fusor shell voltage during arc, longer ground path with round copper wire
arc EMI fusor shell.png

Fusor shell voltage during arc, shorter ground path with round copper wire, shortening the ground path helps greatly
arc EMI fusor surface 2.png

Voltage at vacuum hub surface (common ground with gauge), isolated from fusor shell with ceramic break
arc EMI vacuum hub surface 2.png

Voltage on MKS901P analog out ling
arc EMI vacuum gauge analog out.png

Voltage on MKS901P rs485 line
arc EMI vacuum gauge rs485+.png

Voltage on MKS901P dc power line
arc EMI vacuum gauge power 2.png
Andrew Seltzman
www.rtftechnologies.org
User avatar
Emma Black
Posts: 198
Joined: Wed May 11, 2022 9:42 am
Real name: Emma Black

Re: MKS 901P Loadlock Transducer Fault

Post by Emma Black »

Thanks for the helpful replies, EMI seems to be indeed what killed my 901P as well as it did die during some light arcing. Some of the digital panel meters I was using before died very easily before I added a MOV.

Andrew your comment about using a single copper wire for the main fusor ground is interesting. I thought that just using a large gauge wire to ground would be sufficiently low impedance, but perhaps I should upgrade to using an earth strap. The fusor has a dedicated earth spike, hammered 6 feet into the ground right next to the power supply.
User avatar
Richard Hull
Moderator
Posts: 14976
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: MKS 901P Loadlock Transducer Fault

Post by Richard Hull »

"Ground bounce" is a common thing creating EMP around any high voltage, High energy arcing or discharges. It will bounce right into instruments of a modern digital nature where CMOS is the norm. It loves to screwup modern micro-processors and the ubiquitous micro-controllers that are at the core of all modern instrumentation. Typically, such issues are not an issue with old analog meters especially if bypassed by back-to back reversed diodes if the meters are of a sensitive movement like 50ua to 200ua where the movement will not read more than the junction voltage as a norm.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: MKS 901P Loadlock Transducer Fault

Post by Matt_Gibson »

Would a gauss meter be useful to have running nearby to alert us to potentially damaging conditions?

What I know about grounding is that it’s much more complex than most seem to understand and that whenever you think that you’ve nailed it, you haven’t…

-Matt
User avatar
Richard Hull
Moderator
Posts: 14976
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: MKS 901P Loadlock Transducer Fault

Post by Richard Hull »

As more and more complex, modern systems and components are added to the fusor that self regulate and automate the fusor, the more the grounding issue comes to the fore. Many people who have automated their systems with no real problems tend to be those with money to buy the best and proper "everything" doing the fusor up right and proud. As such, those of us who cobble the fusor up and wind up with "teething problems", (arcs and sparks), tend to blow up the more complex components where micro-controllers we either buy or self-manufacture using same.

The solution is get the system 100% arc free with good crude rather bullet proof older gauges, etc.. Once happy and stable feel free to load the system up with the most modern and interesting electronic goodie you can afford or create from your own design work.

Star grounding is the typical solution mentioned in FAQs in the past. From HV forum FAQ way back in 2015

viewtopic.php?t=9968

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: MKS 901P Loadlock Transducer Fault

Post by Matt_Gibson »

So how do we ground these type of gauges/transducers where there’s an analog output? If it’s the analog out that’s the cause, can we put some metal braid/mesh sleeve around the analog out wire and ground that?

-Matt
Andrew Seltzman
Posts: 814
Joined: Sun Feb 01, 2004 8:02 pm
Real name: Andrew Seltzman
Contact:

Re: MKS 901P Loadlock Transducer Fault

Post by Andrew Seltzman »

For measurement of EMI, a gauss-meter would basically be useless; you are looking for very high speed EMI transients.

I used a tektronix p6015a with the probe tip touching the fusor shell. You can find these used on ebay. A single turn loop antenna would also be reasonably good.

In terms of the gauge, I would isolate it from the fusor shell using a dielectric break, ceramic if you can find one cheap, or an insulating KF o-ring with a plastic centering ring and a plastic KF clamp. I believe Edwards makes these.

Edwards C10512393 oring/centering ring
https://us.my.edwardsvacuum.com/en/USD/ ... /C10512393

Clamp, remove metal grounding bridge inside clamp
https://www.ajvs.com/new/product_info.p ... s_id=23876
Andrew Seltzman
www.rtftechnologies.org
User avatar
Emma Black
Posts: 198
Joined: Wed May 11, 2022 9:42 am
Real name: Emma Black

Re: MKS 901P Loadlock Transducer Fault

Post by Emma Black »

Thanks Andrew - I was considering forking out for a ceramic break but the plastic KF fitting and clamp is a much more budget friendly option.

Especially as I've just spend some cash on a gamma setup, I can't wait to start some experiments.
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: MKS 901P Loadlock Transducer Fault

Post by Matt_Gibson »

Thanks! I didn’t even know these existed (learned some knew). While waiting for these bits to arrive, can the system be pumped down to operating point and then just unplug the gauge when running high voltage?

-Matt
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: MKS 901P Loadlock Transducer Fault

Post by Matt_Gibson »

Emma,

Have you found a source for the polymer centering ring? Looks like it’s out of stock and with very long lead times (20weeks).

I might try removing the SS and using just the o-ring to see if that’ll seal up.

-Matt
User avatar
Emma Black
Posts: 198
Joined: Wed May 11, 2022 9:42 am
Real name: Emma Black

Re: MKS 901P Loadlock Transducer Fault

Post by Emma Black »

For once Edwards UK have them in stock at £5 for a KF16, happy to post you a couple.
User avatar
Richard Hull
Moderator
Posts: 14976
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: MKS 901P Loadlock Transducer Fault

Post by Richard Hull »

Keep in mind you can't use a normal KF16 clamping ring. However you hold the polymer spacer ring and the two fittings tight, the clamp must be non-conductive.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
User avatar
Emma Black
Posts: 198
Joined: Wed May 11, 2022 9:42 am
Real name: Emma Black

Re: MKS 901P Loadlock Transducer Fault

Post by Emma Black »

Of course, they sell the polymer clamps as well. I was just looking at the prices for ceramic breaks for the new fusor I am working on. Ouch.

This is off topic and new thread of course when construction begins but as a quick preview. The viewport is on the bottom and I think I can suffer its restricted view for the simplicity of cooling that the "pipe" provides. I.e. wrapped in copper pipe.
NewFusor.JPG
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: MKS 901P Loadlock Transducer Fault

Post by Matt_Gibson »

Thanks for the offer, Emma. I decided to go a different route and use some kf25 polymer rings and clamps (found a few on eBay) to (attempt) isolate the chamber from the rest of everything.

-Matt
Post Reply

Return to “Vacuum Technology (& FAQs)”