5972 Repeller Voltage

[benhutcherson]

1. Fiilament current AFAIK is entirely in the 5972 mainboard/PSU, not the smartcard or top board

2. The 73 filaments(G7005-60061) fits perfectly into the stock 71/72 source. The alignment is perfect, or at least on the EI source. I didn't try on a CI source-I have one but have no plan to do CI on a 71/72.

3. I THINK G1701AA should be fine for the upgrade I did. I went with BA because the cost is similar and it's a bit newer. If you want to check, try adding a new instrument and see if 5972 is one of the options, but I think the whole G1701 series is new enough to support a 72.

4. Yes, you need the mainboard and I think also the PSU. Also note that the voltage is still at 70eV-all you can do is change the current from the 71 default 35µA to values lower and higher than that. You need a 73 to change voltages.

5. I've only run the standard diameter draw out plate. You may well know this, but I've had long conversations about source cleaning with Geoff Wilson, who was the one who talked me through this upgrade and sold me the parts to do the upgrade. He mention one of the most neglected parts(not 71/72 specific, but all HP/Agilents with the same general source design) being the "step" where the draw out plate sits in the source body, and if it's dirty that the lens can end up floating. He said that he uses "pointy" Q-Tips and a very thick slurry of alumina to polish this area(what he told me was to wet the Q-tip and then touch it to the alumina to get the right consistency). Doing that-along making sure all the lens apertures were clean-made my 71 behave a lot better and give cleaner ramps than I got before.

[LALman]

This sounds like a do-able upgrade. I've found a main board and a power supply board. I'm going to try this swap. I guess I'll swap back in the smaller ID draw-out plate and some 5973 filaments while I'm at it.

[LALman]

(6) I should mention, the 6 mm ID draw out plate I was using is the ~$90 "inert" version rather than the $300 version. I wonder if the alloy for the "inert" disk is different and causing the odd ion focus ramping problem.

(7) How do you keep the draw out plate (or indeed any of the source parts) from getting a wear pattern from touching the bottom of the beaker when sonicating? Should I be using plastic beakers?

(8) Does the power distribution board also need to be changed along with the power supply board?

[benhutcherson]

Can't comment on 6 as I don't know how the version you used differs. I've read on Restek's blog with them talking about machining their own from the stock part.

For 7, I don't worry too much about it. As long as the aperture in the draw out plate is in good shape, it shouldn't affect ions traveling through it. FWIW, I don't know how other folks polish, but I cut off a small piece of the cloth, put alumina on it, wet it, and then rub the faces of all the flat surfaces(repeller, all the ion optics) over it and I can usually get a mirror polish without much effort. They come out clean when I sonicate.

8. I think you'd be fine with the 71 power distribution board, but if you can find one from a 72, you might change it. I swapped the complete eMod, which is also a bolt in. The 72 eMod eliminates the silly EI/CI switch completely(you'll remove it or just leave it disconnected if dropping a 72 mainboard into a 71 card cage).

[James_Ball]

1. Fiilament current AFAIK is entirely in the 5972 mainboard/PSU, not the smartcard or top board

2. The 73 filaments(G7005-60061) fits perfectly into the stock 71/72 source. The alignment is perfect, or at least on the EI source. I didn't try on a CI source-I have one but have no plan to do CI on a 71/72.

3. I THINK G1701AA should be fine for the upgrade I did. I went with BA because the cost is similar and it's a bit newer. If you want to check, try adding a new instrument and see if 5972 is one of the options, but I think the whole G1701 series is new enough to support a 72.

4. Yes, you need the mainboard and I think also the PSU. Also note that the voltage is still at 70eV-all you can do is change the current from the 71 default 35µA to values lower and higher than that. You need a 73 to change voltages.

5. I've only run the standard diameter draw out plate. You may well know this, but I've had long conversations about source cleaning with Geoff Wilson, who was the one who talked me through this upgrade and sold me the parts to do the upgrade. He mention one of the most neglected parts(not 71/72 specific, but all HP/Agilents with the same general source design) being the "step" where the draw out plate sits in the source body, and if it's dirty that the lens can end up floating. He said that he uses "pointy" Q-Tips and a very thick slurry of alumina to polish this area(what he told me was to wet the Q-tip and then touch it to the alumina to get the right consistency). Doing that-along making sure all the lens apertures were clean-made my 71 behave a lot better and give cleaner ramps than I got before.

I have gone through many different cleaning procedures in the past, including Dremmel with brushes and buffing wheels with motorcycle chrome polish(takes forever to wash off) to jewler's rouge, to Sharkskin sanding paper down to 20,000 grit, and now just use the alumina powder diluted into citranox until is it a little thinner than toothpaste and hand rub with the swabs. It can take a while and you might get hand cramps before finished but it takes everything off the source and washes away with warm water before sonicating.

I normally sonicate with citranox/water, then clean water, then methanol, then methylene chloride, then throw the parts in a beaker and put in the GC oven at 50c for about 10 minutes, let cool and reassemble.

I only clean the metal parts, the vespel takes too long to get rid of water if you clean it. On the 71 if I remember the repeller doesn't come out of the holder so it just gets polished with ultra fine polishing paper since the vespel insulator doesn't need to get wet, but for 73 and up I disassemble it and clean with the rest of the parts, but only the flat part.

[LALman]

I also use the alumina powder. I bought the fiber free cloth from SIS along with plenty of cotton swabs and I do just like you both say. I make a paste of methanol and alumina and dribble it on a spot on the cloth and then do a "t" with all the flat surfaces till they shine. Then I polish all the hard to get places with the swabs dipped in paste. I also like to take a piece of the cloth wet with the paste and use that to rub the outside and inside surfaces of the lenses. But I am now also going to show special attention to that lip the draw-out lens touches. and to the edges of the lenses.

I'm expecting my new 5972 mainboard this Thursday. I hope to get it all working this coming weekend.

[LALman]

I put in the 5972-60020 mainboard. I blew out all the dust while I was at it and thoroughly cleaned the source. I put in new 5973 filaments. I put a yttria coated Re-Y filament in the lower (as hanging below the column inlet when fully assembled) filament position because I did not have two new regular filaments.

The manual seems to have a typo because it shows the source with the upper and lower filaments labeled as Filament 1 and Filament 2 relative to the source when it is upside down on the table for assembly. So, would Filament 2 then be the one actually above the column when it enters when the whole thing is assembled? I will be checking both filaments to see which is which in the software.

After a day water has dropped to 2-4%. Nitrogen 1.5% Oxygen 0.46%. I am starting to get good lens ramps. Although they have multiple maxima compared to my 5973. I see desirable new menu items such as 'purge calibration valve'! and I now have control of filament current.

[benhutcherson]

One of the downsides of this upgrade rather than just swapping with a complete 5972 is you're still dealing with that rough cast vac manifold and not the nicely machined/polished 72 manifold. That can still make it a bear to pump down-I was advised once to "Start it on Friday, give it a hug, and say 'See you Monday.'"

It seems to me that the 71 manifold takes a week to really stabilize even if it's just been open a few minutes at least in my experience. Upping the rough pump size from an E2M2 helps a bit(I run a Varian DS102-the guy who helped me with this said that he puts Edwards RV3s on everything from 71s on up to 77s) since it gives you a lower final pressure, but water in particular can still hang around for a while.

I think you'll be happy once this has settled down.

As for the new menu options-I wasn't sure for me if it was the new mainboard or the fact that I jumped to a totally new version of MSD Chemstation.

One small tip-at least on B.01.00(I still need to upgrade to B.01.02) I found it would lock up the software when I tried to ramp filament current, even though that was given as a rampable parameter.

[James_Ball]

I never ramp filament current since it goes up really high and it never needs to be that much, and I always worry if it will burn one out.

The manifold is one of the problems with pumping down the 71, but the other problem inherent in both the 71 and 72 is the lack of a heated source and quads. The source is heated by contact of the gold plated inlet piece with the end of the transfer line. If that doesn't make solid contact then the heating is even less efficient. When I put one back together I would always slide the top plate side to side a little and watch the column end of the transfer line, once it moved a little out from the MS I knew there was good contact and that is when I hit the power switch and let the vacuum pull the top plate down hard.

If when you run the repeller ramp, it maxes out at the right hand side of the graph, then you probably have the emission current too high, you can drop it a click at a time until you see a maxima of the repeller ramp just before the maximum voltage of the repeller and that is where you really want it. This is for m/z 69 and 219, 502 usually maximizes much higher so don't worry about it too much.

The other trick is lower current gives higher abundance to the higher masses, since there is less energy to break down the molecules.

[LALman]

My detector is coming out around 2000EV so I decided to put in a new ETP Model AF516 ion detector. While I was at it I got in a new diff pump fan installed it all and started it up.

There is one odd thing. This is supposed to be an official 5971 cooling fan. The flow is pulling in from the back over the diff pump fins and flowing out from the front. This fan has no markings on either end of the fan.

The previous fan was flowing from inside the electronics drawer and then out the back over the diff pump fins. It has a less finished looking wire harness and a 115V single phase 7W fan made by NMB-MAT Model 3115ES-12W-B30.

I could make arguments for either direction. Pulling fresh air from the back and then out from the front would warm the vacuum tank and maybe help drive moisture out. But is it going to warm the electronics too much?

Pulling air through the front of the machine and out the back over the fins give cooler electronics which might be desirable but also keeps the vacuum tank closer to room temperature.

Edit: I found this link from Agilent about the 5971A which suggests the flow of hot air will be out the back...
https://www.agilent.com/cs/library/Supp ... A20344.pdf

I found a couple more mentions that say the cooling fan should vent to the back. So, I vented and swapped out the new fan for the old one and its pumping down now.

[benhutcherson]

When I was talking with Geoff Wilson, who initially advised me on this upgrade and supplied the parts, one of the things he use to always do when he was regularly servicing 5971s/5972s was replace the fan on a somewhat regular basis. I've not replaced mine, but I do regularly blow the dust out and make sure it's spinning freely. Unfortunately, these type fans taren't really serviceable outside of that, so I do think it's worthwhile to have a replacement on-hand(although I've not followed through with that).

Geoff also made an interesting observation to me. He said that he would occasionally get calls about 71s/72s that would tend to have varying baselines throughout the day and also seasonally. Almost invariably, he'd find an HVAC vent somewhere in such a place that its airflow would directly hit the diffusion pump on the instrument. As the HVAC would cycle on and off, the baseline in the instrument would change.

Of course, the 5973 either eliminate that problem completely with a turbopump, or at least locate the diffusion pump such that it's shielded from the effects of ambient air. I dread servicing the one on my 75 because I find it really awkward to remove from the instrument compared to the 71/72. It's SOMEWHAT compensated by the fact, though, that I can check the oil level without removing. Although I've never done it, I've also been advised that the oil level can be topped up by adding oil directly through the "vents" in the vac chamber.

BTW, don't forget that the default EM voltage on the 71/72 is 1400V, which is a fair bit higher than later instruments start out. Even with a fairly new EM, mine tunes to ~1600V. I don't know that I'd discard one at 2000V, although of course if you see an improvement after replacing it's hard to argue with the result.

[LALman]

Ben, I thought I needed a new detector because I was getting very bumpy lens ramps. With the new one it, I'm getting a standard tune at 1247-1296EV. But if I run the 'repeat profile' stability test The 69 peak (for example) ranges from 200000 to 600000 up and down and back again. The repeller ramps are lumpy but fairly reproducible but the Ion lens ramp vary widely each time I run them.

I was having the same problem (except I did not try until now doing the 'repeat profile' test) with the 5971 mainboard.

Any ideas what might be causing this variability? Could this be a top board problem?

[James_Ball]

Ben, I thought I needed a new detector because I was getting very bumpy lens ramps. With the new one it, I'm getting a standard tune at 1247-1296EV. But if I run the 'repeat profile' stability test The 69 peak (for example) ranges from 200000 to 600000 up and down and back again. The repeller ramps are lumpy but fairly reproducible but the Ion lens ramp vary widely each time I run them.

I was having the same problem (except I did not try until now doing the 'repeat profile' test) with the 5971 mainboard.

Any ideas what might be causing this variability? Could this be a top board problem?

Are the peak widths changing also or just the peak heights?

If the widths are jumping around it can be a top board problem. I had that once and found a resister that was really swollen in the middle just behind the box over the quad coils.

You should be able to reverse the fan direction just by flipping it, just have to be sure the blades clear the fins. With the more finished wiring harness it sounds like a 5973 fan which does blow out the front of the instrument.

If you get a 10F room temperature change over night you will see the sensitivity on a 5971 increase as the temperature drops and decrease as the temperature rises. I used to fight with that a lot until I found out someone turned on the "night setback" setting on the HVAC in our volatiles lab years ago. The Energy Saver mode was costing us more in reruns than it was saving in electricity lol.

[LALman]

Widths are staying constant but peak height. I reduced smoothing from 21 to 0 on the lens ramps and it looks very ragged. I think with all the venting, the source is dirty again. I'll vent and clean and see it that gets me more stable CPS.

[LALman]

I've put in a new detector, new fan, and recleaned the lens set and used the newer single piece lens insulator because it keeps the lenses from moving. But no joy. It manages to tune but pick any peak and it is variably by a factor of 100% to either side of the median. say set for 400K it will range from 200K to 800K with occasional worse excursions. Ramping lenses is completely wacky, it cannot give the same ramp twice. Particularly on the ion focus and entrance lens ramps. The entrance offset ramp is steady. The repeller ramp is somewhat noisy with double and triple shoulders on each masses ramp.

Took off top board covers and the C5 cap on the high voltage end looks fine.

So, it might be the signal cable? Or more likely the top board needs to be reworked. Tht high voltage block on the end that takes 12VDC to 3KVDC is often what gets changed out in my experience. Arrrrgggg!