5972 Repeller Voltage

[benhutcherson]

What do your quads look like?

The 5971s/72s are notorious for getting little "bites" where the ion entrance lens fits into the quads.

If it's really bad, you can flip the quads around(so that the former detector end is by the source) and potentially get some more life out of it, but don't undertake that task lightly.

[LALman]

I'll break vacuum and take a look. I talked to FSA and they probably don't have any top board units. However, he opined that it was likely the Gamma High Voltage Research Model MC30 12VDC to 3KVDC module. He said if the peak width was clean and constant but intensity was all over the place that would be his best guess. He said it might also be the ribbon cable but was unlikely to be the filament cable.

Its possible that I have one of the older power supplies. It looks like it has a skinny electric blue mount/heatsink (which is the older model) and he said that might also be a point of failure.

[benhutcherson]

Email me at hutchersongcservice AT icloud.com . I probably have a spare top board(board only on a chamber top cover-no source/quads/etc) that's tested/known good. I have a complete one(quads, source, analyzer plus top board) that carefully wrapped in a box in my car now, but I'd rather not part with that one since it's a handy diagnostic tool.

[James_Ball]

I'll break vacuum and take a look. I talked to FSA and they probably don't have any top board units. However, he opined that it was likely the Gamma High Voltage Research Model MC30 12VDC to 3KVDC module. He said if the peak width was clean and constant but intensity was all over the place that would be his best guess. He said it might also be the ribbon cable but was unlikely to be the filament cable.

Its possible that I have one of the older power supplies. It looks like it has a skinny electric blue mount/heatsink (which is the older model) and he said that might also be a point of failure.

I would agree. Quad problems are usually evident with peak width problems. Another place to look is the ceramic feedthrough where the top board connects to the board inside with the wires for the lenses. I had one once that when they welded it in, there was ion sputter on the ceramic and at times it would conduct between the pins. The service engineer had to sand the ceramic with the green paper and it started working so they sent a new top plate. It it was working before then maybe it got dirty on the inside under that small board. Also that board on the inside can go bad and give problems sending clean current to the lenses.

If it was the high voltage circuit you would need a special attachment for a voltage meter to test that. I saw a service engineer use one once, but I never had one of the attachments to reduce the voltage to within the range of normal meters.

[LALman]

Here is what we did this last Wednesday...
I got in a new MC30 3KV DC to DC converter. We measured all the voltages at the test points with the following results.

MAINBOARD
+5V  was 4.98V with 2.8mV of AC ripple
+15V was 14.98V with 2.8mV of AC ripple
-15V was -15.24V with 2.2mV of AC ripple
+24V was 28.2V with 1.0mV of AC ripple
Fluke meter was giving 3mV of AC ripple with leads simply crossed.

We checked each of them while running an Air and water check
+24V varied from 27.00 to 27.60V
+15V varied from 14.90 to 14.89V
-15V varied from -15.2V to -15.28V
5V varied from 4.98V to 5.02V

TOP BOARD
HV has 0.8mV AC ripple

We alternately checked both new filaments while scanning They were very close to 70V and remained steady during ramps.
FIL to FILA (2) had 1.26 VAC
FIL to FILB (1) had 1.60 VAC

We checked Ion Focus set at 85 and measured -85.4V
Ramped: repeller, ent off, and ent-lens with no voltage change

Repeller set at 24.53V we measured 24.36V
Ramped: ion focus, ent lens, and ent-off with no voltage change

Ent-Lens set at 17.32 we measured -18.33V
That actually changed depending on ion we ramped
Ramp 69 gave 13.48V
Ramp 219 gave 15.15V
Ramp 502 gave 18.33V

We standard tuned and got 1353EV (calculated 2.841V) and measured -2.821
Ramping one mass for Ion focus, Repeller, Ent lens, and Ent-off we got almost no variation in the measured voltage 69 -2.821, 219 -2.822, 502 – 2.831
Doing a continuous scan did cause varition of the high voltage input from -2.821V to -2.855V which would be maybe an EV variation of 1345EV to 1359EV

The only power supply voltage that seemed to vary during profile scanning was +24V.
At 1353EV mass 69 gave 27.50V, mass 219 gave 27.25V and mass 502 gave 27.05V. The other rails (+15V, -15V, and +5V) were steady

We calculated that if -6.3V is 3000V at the high voltage section we could guess pretty close what the voltage would be for a set EV.

For EV 1106 we calculated -3.310 V and measured -2.305 V
For EV 1206 we calculated -2.533 V and measured -2.501 V
For EV 1306 we calculated -2.743 V and measured -2.723 V
For EV 1400 we calculated -2.940 V and measured -2.920 V
For EV 1506 we calculated -3.163 V and measured -3.142 V

Next we Set EV at a value and put the MS in scan mode and measured the variation at the test point of the voltage going into the HV section
Set EV 1000: Test Point -1.93V to -2.10V calculates as 919EV to 1000EV
Set EV 1100:  Test Point -2.19V to 2.42V calculates as 1042EV to 1152EV
Set EV 1200:  Test Point -2.40V to 2.63V calculates as 1143EV to 1252EV
Set EV 1300:  Test Point -2.63V to 2.84V calculates as 1252EV to 1353EV
Set EV 1400:  Test Point -2.83V to 3.04V calculates as 1348EV to 1448EV
Set EV 1500:  Test Point -3.07V to 3.27V calculates as 1462EV to 1557EV

Now that I look at these variations above. I think this is the possibly a problem here.

While doing all this we got an HV section fault when we had no leads on it and were just asking for a profile. We decided that this happening was reason enough to swap out the HV30 module.

We did this and on inspection found one resistor lead in the low voltage section that he thought should be touched up. We put it back under vacuum and the problem is still happening.

THE PROBLEM CONTINUED:
Lens Ramps of all types, are not repeatable. Each mass has multiple humps and repeated ramps never gives the same scan twice.

SO:
I think it may be the top board. I think the top board high voltage section is changing too much during scans. But I have no idea what part of the circuit keeps that steady. What do you think?

[LALman]

I got in the new RV3 backing pump Sept 15-16 (Tues or Wed). Since I was redoing the whole shebang, I broke every joint in the system and sonic cleaned that whole 3-way port assembly that mounts the PTFBA vial and vacuum triode. I cleaned and baked all the seals and reassembled with Apezion M grease. While I was at it, I opened up the diff pump and measured the amount of fluid. I was showing 15mm deep. That is actually too much. Should be 9-12 mm. So, I sucked out about 6mL and ended up with about 10-12 mm of fluid. While I was at it, I sonicated the stack assembly in 99% isopropyl alcohol, then DIW, and then dried it thoroughly at 105C and then briefly at 180C before reassembly. I did as you suggested with the diff pump seal and used a smear of diff fluid.

When I got it buttoned back up the RV3 pump did indeed pump down very fast to about 30 mTorr and even then pretty fast further down to about 14 mTorr. I fired up the diff pump and have spent all last week testing your board with the RV3 (it works!) then with the E2M2 (also works!) and then this monday I put my old top board back on and tested it with the E2M2 (now works) and then swapped to the RV3 (it works!). I was slowed down a bit this week because of a big job that had me working a couple of 14 hour days.

So, it would appear to have been a vacuum problem.
With He flow at 0.7 mL/min the E2M2 gets down to 14.5 mTorr and the RV3 gets down to ~13.5 mTorr but it can also get there quicker. I'm supposing the excessive diff fluid was not helping but the Apezion M grease did the trick on my tune valve vial. It still leaks a bit but nothing like it was leaking. I assume that its helping at all the other seals also. While I was running all the lens ramps, the vacuum went up from 2.5E-5 Torr to 3.5E-5 Torr and was stable there which is a great improvement. Before it was bouncing around between 5E-5 to 1E-4 Torr while tune vial was open.

Every time I've broken vacuum the tune EV goes higher but I figured I should try varying as few things as possible. So, I am leaving a thorough lens cleaning for next. I have the new Apezion L just in so when I do the source cleaning, I will apply it.

[LALman]

So, now I have cleaned the source within an inch of its life and I am getting good vacuum but low sensitivity. Its tuning at 2900V. I wonder if I have borked my detector with excess diffusion pump back flow from it being too full. I'm thinking of trying to clean it. What solvent should I try and should I put it in a heater to dry or just put it right under vacuum for fear of oxidation?

[James_Ball]

So, now I have cleaned the source within an inch of its life and I am getting good vacuum but low sensitivity. Its tuning at 2900V. I wonder if I have borked my detector with excess diffusion pump back flow from it being too full. I'm thinking of trying to clean it. What solvent should I try and should I put it in a heater to dry or just put it right under vacuum for fear of oxidation?

Cleaning the EM will depend on which one you have installed. Look at the manufacturers site to see if they recommend a solvent. I would definitely avoid chlorinated solvents and water, I am pretty sure those are not good for them. The biggest problem is getting all the solvent out before applying a current to it. I would not heat it at atmospheric pressure, just put it under vacuum a few days, then start at the lowest voltage you can(usually around 800v) and scanning in manual tune before trying to tune or ramp up the EM volts. Usually when they start requiring more and more voltage with each venting they are getting near the end of their life unfortunately.

I normally replace mine with the ETP units, they seem to have less noise and operate at lower voltages and last longer.

[LALman]

This is my new 1X ETP Model 14516 AF516 Electron Multiplier bought in mid June 2020 before I realized I had a vacuum problem. I don't see any visible spotting on it from diff pump oil but I'm getting terrible signal with a strong left to right assymetry of the peaks. e.g. background is higher on left than on right.

[LALman]

Reading this source cleaning method https://www.agilent.com/cs/library/technicaloverviews/Public/5989-5974EN.pdf It says that heating the source with high background water will reactivate the source all over again.

I pumped to <30mTorr then turned on diff pump and then after high vacuum was
good I turned on source heating and left for the day. But when I checked water and air the next day...

Day 1: 4.3% Nitrogen, 1.2% Oxygen and 50% water vapor. 69 = 628544
Day 2: 2.1% Nitrogen, 0.7% Oxygen and 36% water vapor. 69 = 1043840
Day 3: 1.4% Nitrogen, 0.5% Oxygen and 20% water vapor. 69 = 724928

Today is day 3, so I am guessing maybe that water vapor ruined my cleaning job and I should re-clean the source get it under vacuum and not heat the source until water signal is low as per this pdf? Could take several days but at least I would not be undoing my lens cleaning job.

[James_Ball]

Reading this source cleaning method https://www.agilent.com/cs/library/technicaloverviews/Public/5989-5974EN.pdf It says that heating the source with high background water will reactivate the source all over again.

I pumped to <30mTorr then turned on diff pump and then after high vacuum was
good I turned on source heating and left for the day. But when I checked water and air the next day...

Day 1: 4.3% Nitrogen, 1.2% Oxygen and 50% water vapor. 69 = 628544
Day 2: 2.1% Nitrogen, 0.7% Oxygen and 36% water vapor. 69 = 1043840
Day 3: 1.4% Nitrogen, 0.5% Oxygen and 20% water vapor. 69 = 724928

Today is day 3, so I am guessing maybe that water vapor ruined my cleaning job and I should re-clean the source get it under vacuum and not heat the source until water signal is low as per this pdf? Could take several days but at least I would not be undoing my lens cleaning job.

When I clean a source, I will clean it with aluminum oxide grit mixed with citranox, then wash well with tap water to remove the grit and soap. Then I sonicate with DI water to remove any residual minerals from the tap water, then sonicate with methanol to help remove water then sonicate with methylene chloride to remove any hydrocarbon residues that might be left. After I put the source parts in a beaker and stick them into the GC oven at 50C for about 30 minutes to remove residual solvents and water. After assemble and install the source and wait a couple hours with the heater on before scanning to check for leaks. There will always be some water, usually it will be half the height of 69m/z but will go away overnight. Never had any problems with the water affecting the source cleaning.

If the peaks are still asymmetric after the water is gone, I would be looking at the quads or coils as the peak symmetry is affected most by the quad circuit. Don't forget to check the flat woven leads that connect to the side of the quads, those can go bad, I have had to change them before a few times, easy to do if you can still get them.

[benhutcherson]

To me, one of the biggest annoyances with the 5971 is just how badly the manifold sucks up...pretty much anything...any time it's open. Given a choice, I try to not even power it on until it's had a solid 72 hours to pump down.

Of course, your big issue with heating is that it's just not possible to get things that hot in the instrument. Even if you set your transfer line to, say, 300º for a bakeout, you might still only see a source temperature of ~180º or so if that high, and in normal operation with the line at 280º, 160º is a good number.

I pulled one sometime last year that was tuning at ~2200V chasing some other problems that it didn't fix, but left it since I liked getting 1400-1500V tunes and figured it was still serviceable just in case. Don't remember the p/n, but it was the ETP discrete dynode type-those are what I've always stuck in when I needed a replacement. I probably have one or two serviceable ones of the HP spec continuous dynodes that I got from Geoff(he said he usually used those). I can get one in the mail in the next few days.

[LALman]

I hope this was mostly using the 9mm draw out plate and not having tune vial sealed enough. I vented, swapped in the 3mm draw out plate and pumped down again. I'm leaving source cold but under high vacuum until water signal is down. Also trying to get that darned tune vial to seal properly. Applied Apiezon L grease emptied vial added new PTFBA and reinstalled vial. That dropped water vapor from 40% to 7%! When I left yesterday I had only ~3000 counts of mas 28 and mass 32 and around 40,000 mass 18. This with EV ~2000 and Mass 69 at 700,000.

[James_Ball]

I hope this was mostly using the 9mm draw out plate and not having tune vial sealed enough. I vented, swapped in the 3mm draw out plate and pumped down again. I'm leaving source cold but under high vacuum until water signal is down. Also trying to get that darned tune vial to seal properly. Applied Apiezon L grease emptied vial added new PTFBA and reinstalled vial. That dropped water vapor from 40% to 7%! When I left yesterday I had only ~3000 counts of mas 28 and mass 32 and around 40,000 mass 18. This with EV ~2000 and Mass 69 at 700,000.

Mass 69 at 700,000 is quite high unless you are trying to do some ultra trace work. For most analysis in low ppm on column you should only need about 300,000 for mass 69 which should bring your EM voltage down quite a bit. I can run sub ppm on column with 69 at or below 500,000 no problem.