Chromatography Forum

In my continuing saga of bringing back my 6890+-5973 that was allowed to regurgitate foreline fluid into the diff pump I am having 2 main issues now the biggest of which is tuning.

I had this happen once on my turbo system. I had a fan failure in the MS and the resulting shutdown pulled rough pump vapor through my purge vent valve because I had all the vents ganged together in the hood. I ended up cleaning the inlet, replacing the purge vent filter, and finally swapping the EPC. All had been contaminated with rough pump oil vapor. Once I did that, things quieted down.

I hope its not your calibration vial valve. Those darned things are super expensive (~$3000).

You can try for more quiet in the baseline by swapping in an EPC from a working system after making sure the rest of the inlet hardware is clean.

Another possibility is filament sag. If any part of the filament body gets too close to the source, it will ground and turn all signal into hash. If this were intermittant maybe that would cause some of what you are seeing.

I think that is the most likely problem at this point. I'd rather have Agilent do it because I can't confirm this and I can't buy a $2500 board, and then risk having to tell my boss nope guess that wasn't the problem.

But yea the problem appears to be it is unable to direct ions from the filament to the detector. That is why I only saw a bit of acetone when I flooded the instrument with it.

$2500 is a steal for a 5973 sideboard. I use Full Spectrum Analytics for support. A few years ago, they were willing to send me a sideboard to try out when I was getting similar transient artifacts. Said it was either the sideboard or the Electron Multiplier. If the side board fixed it then I would be out $4000 and if not I could send it back. As it turned out, I needed a new Electron Multiplier.

Agilent wants $5400 to fix it. $3k for the board and $2,400 service. If I could absolutely confirm it is the sideboard there is a place online that will repair them for $1800.

When I try the tune there are several faint noise peaks especially arround 69 that the MS plays with and turns up the detector to 2600 and usually gives up and says no consitent peak widths and such.

Plus I am going to purchase an ALS controller card kit $1050 for my 6890+ so I can use my 7683 rather than have the g1512 fixed. I opened it up and it was full of dust. I used a whole can of duster but it still cannot communicate with either the 6890 rs-232 nor 5890 gpib. I may try rinsing it out with IPA since why not. I think the production facility has a bad dust problem. though the MS itself looks ok no sign of dust. I tried hitting the electronics with duster spray anyways but no success there.

I don't think it is the calibrant valve at this point. It clicks very loudly and the acetone test makes me think it is a matter of not getting ions from the filament to the detector.

I checked for any sign of shorts. I always worry about the lens insulator as it is 2 piece and the screw presses on the crack opposite the wire leads. However I am very careful to leave all that stuff barely finger tight plus the first time arround I didn't touch the source just fixed the vacuum system and tried it.

I wonder what If I just took the entire door off with the whole analyzer assembly and sent it out for repair some place.

I wonder what If I just took the entire door off with the whole analyzer assembly and sent it out for repair some place.

I'd think there is no way you could keep that door from damage. Its not that hard to ship the whole 5973 unit.

I suggest you contact FSA. Their field support man in Denver is very good. I shipped my whole 5973 inert in to their Denver office. They sent me a special shipping box. ran me about $270 for shipping FedEx insured for $20,000.
https://fsaservice.com/locations.php
It only took him 3 hours of labor plus parts to diagnose and fix my sagging filaments problem and the source came back cleaned really nice.

I wonder what If I just took the entire door off with the whole analyzer assembly and sent it out for repair some place.

I have ordered that part before, for a 5971, and when Agilent sent it, it came attached to an analyzer chamber, which I believe were just extra ones or ones that failed QC due to nicks on the sealing flange or something. It had the vacuum pump and inlet ports capped off. This way it can be shipped and the quads are protected from damage. Of course from Agilent that sells for something like $10,000 now days, it was $5,000 back then.

When I received it I called and asked if I was supposed to install the whole analyzer! My local service engineer just laughed and told me not to because the housing could be defective and give me more problems than I already had

I would'nt assume its the whole assembly. The last time I had odd artifacts and weird peaks coming and going, it was the Electron Multiplier. If you had some backflow of vapor, it might be coated in all the wrong places.

I don't think any oil got into the analyzer chamber. It was operating fine when it got taken down 1.5 years ago. I was at the plant because they thought they had an ALS problem when in reality the inlet was so poorly insulated it wasn't holding temperature and getting to ready status. That was a easy fix with some glass wool. I happened to notice the vacuum was sucky -4 torr instead of -5 and couldn't see any diff oil in the window. When I took it down it came down and formed 2 layers of inland 45 contaminant and santovac. Quads look very clean. A bit of santovac in the baffles at the bottom of the analyzer chamber.

Do you really think shipping just the door is extremely risky? the quads look to be decently shielded by the metal enclosure something would have to poke in there. If I wrap it in foil and air cushion and put it in a big box with peanuts it looks reasonably safe.

I could send just the side board. How does it come off? I see the 5 wires on the outside how does it seal and what inside has to come off other than the source and quad wire leads.

The quads are very breakable and practically the most expensive part of the unit.

The 5973 side board does come off. You might want to think about this before deciding to do it.

Wear an antistatic strap and have an antistatic bag ready for the board.
1a) Completely power down the 5973 MS.
1b) You can keep vacuum on the chamber with the roughing pump (plugged in elsewhere) if you like.
2) remove connectors from sideboard
3) remove cooling cowl screw and cooling cowl from sideboard
4) remove mounting screws from side board
5a) The RF coils come with the board. Don't try to take any other component off the board.
5b) Cant remember, but I think the RF enclosure screws go into the vacuum chamber door and must come out but the RF coil enclosure stays on the board. Don't try to take anything (except the cooling cowel) off the board.
6) very gently with an even pressure and not bending the board; pull board away from several groups of pin sockets that penetrate the door of the vacuum chamber.

Thats it.

You might want to get an electronics tech to do this for you or have agilent walk you through this as you do it.

I was just looking at my 5973, I see the board is held on by some torx screws. I think the one in the back side of the RF box and the two on the front side are also holding the board in place. If you remove the door you can look between the board and the door and see for sure. I remember the 5971/72 had screws mounted right up against the RF box, so the probably do also.

It is pretty easy once you get the screws out, the feed through pins in the circles at the front slide off easily, I think the ones under the RF box are a little more tight but not bad.

Just an update We decided to go with Full Spectrum. They are coming in a week. They think it is the detector not the electronics board that if it was electronics I would get nothing I injected acetone again and got a good spectra but way low sensitivity. If there was any way to confirm that I could fix it myself but I can't spend another $1700 then shrug my shoulders when it still isn't working.

I am also having a bit of trouble with the triode guage/ionization controller. When I activate it it drifts up to 1E-4 before finally settling back to 5E-5 which is probably correct and stays there. The tube isn't pulsing but looks a bit dark. Is there any way to tell if it is the triode tube which I can replace for $250 or try cleaning, or the guage which I see from several sources usually needs 3 capacitors replaced to fix.

Just an update We decided to go with Full Spectrum. They are coming in a week. They think it is the detector not the electronics board that if it was electronics I would get nothing I injected acetone again and got a good spectra but way low sensitivity. If there was any way to confirm that I could fix it myself but I can't spend another $1700 then shrug my shoulders when it still isn't working.

I am also having a bit of trouble with the triode guage/ionization controller. When I activate it it drifts up to 1E-4 before finally settling back to 5E-5 which is probably correct and stays there. The tube isn't pulsing but looks a bit dark. Is there any way to tell if it is the triode tube which I can replace for $250 or try cleaning, or the guage which I see from several sources usually needs 3 capacitors replaced to fix.

I posted this to James Ball a while ago....

I got a used HP/Agilent 59864B ion gage on E-bay for <$300. If you have one that has died; I can post a link to the most likely repair that will fix it, that worked for mine.

I seem to remember a fix for the ion gauge, just didn't remember where I found it before. Is it similar to cleaning the penning gauge used in the ICPMS from Agilent?

Its the filter caps. Here is the link... HP/AGILENT HP59864B Micro-ion Ionization Gauge also Granville-Phillips 342, 343 series vacuum gauge. Also he provides a test circuit to emulate the triode tube. It worked beautifully for us.

Yep I saw the youtube video. You basically have to desolder 3 capacitors and resolder 3 new ones in. I'm not sure if it is the guage or tube though. The tube isn't pulsing and the gauge isn't drifting as much as the video and is finally settling after 3 minutes.


https://www.youtube.com/watch?v=TFAi3vP4rUE

Just an update FSA is pretty sure it is the detector not any of the boards. It often starts the tune then after the first few screens loses all the PFTBA ions. I also tried injecting an n-alkanes mix and it gave good peaks and spectra and low sensitivity.

It looks like some oil did get in the MSD but it didn't get to the analyzer components, The sides of the analyzer chamber had some coating of crud that was found when a swab or kimwipe was run over it but not the analyzer area and the quad voltages are good. Also there was some crap in the triode tube looked like brown polymer of something. I took the tube out and rinsed the heck out of it with MeCl2 and methanol as well as the analyzer chamber wiped that down with kimwipes. I also had maintenance replace the filter caps on the ion controller.

The only other issue is I am having trouble with the transferline leaking. I tried several vespel graphite ferrules and it initial makes a good seal then after a day it starts leaking and tightening doesn't seem to help. I tried a silltite nut and ultimetal ferrule and it leaked as well. Transferline looks in good condition threads are fine and surface looks good overall.

We'll see what happens when I get the new detector in.

Good to hear FSA is working out for you.

Just an update We decided to go with Full Spectrum. They are coming in a week. They think it is the detector not the electronics board that if it was electronics I would get nothing I injected acetone again and got a good spectra but way low sensitivity. If there was any way to confirm that I could fix it myself but I can't spend another $1700 then shrug my shoulders when it still isn't working.

I am also having a bit of trouble with the triode guage/ionization controller. When I activate it it drifts up to 1E-4 before finally settling back to 5E-5 which is probably correct and stays there. The tube isn't pulsing but looks a bit dark. Is there any way to tell if it is the triode tube which I can replace for $250 or try cleaning, or the guage which I see from several sources usually needs 3 capacitors replaced to fix.

If you need the triode tube you can get it from...
MKS Instruments, Inc
6450 Dry Creek Parkway
Longmont, CO 80503-9501 USA
Part for the 5973 is is (342003 mini IGF Triode IC .75KOVAR) for $170. They have the cable also for $105. I converted my 5971 system over and the same tube works on both.