Hello Everyone-
Finally the proximate cause of the originally described problem was found, and a solution has been crafted. Unfortunately, the actual cause of the problem has still not been discovered, but this is not needed as the crafted solution solves the issue for all purposes we are interested in. I describe in a bit more detail the solution below for those that are interested. Even if no one reads this, at least it will be in the archives in case the issue comes up for someone else with one of these older Bruker consoles.
The solution: as described in previous messages, we were having little luck getting ideas from service personnel, and had replaced multiple boards in the instrument - several of them multiple times. One snowy day in Southern Oregon (not at all common) I decided I would read through multiple Bruker manuals to see if I could determine how the preamp gating is supposed to work in this AV console. After an hour or two, I determined that the EP pulse controlled the preamp gating, and this is generated by the ‘master’ SGU (according to the manuals anyway). We physically swapped the SGUs (the master SGU is located physically in the left hand slot) and also swapped out the ACB-S, which comes between the SGU and HPPR. These actions had no effect. We removed the jumper in the HPPR that determines the polarity of the EP pulse (JU2) and found that, if the EP pulse were actually given correctly, the preamp gating would indeed work - as with the jumper removed, the ‘high’ level present gated the preamp off and allowed full power output of the pulses. We disconnected the line at J0G in the HPPR and replaced the EP pulse normally present there with a pulse from one of the ‘customer specified’ lines out of the TCU3 (we used bit 28 of NMR word 3 set with the ’setnmr3’ pulse program command). And, yes, this does result in a fully functioning spectrometer once the pulse programs are appropriately edited (Hooray!).
The question remains: what is the ultimate (actual) cause of the EP pulse not functioning correctly? We don’t have an answer for that one…….
I want to thank - in addition to others who responded to previous messages - Arnold Itkin, who responded to the last message and had some useful ideas.
Wishing everyone a Happy Spring from Southern Oregon.
Tom Pratum, Southern Oregon Univ
> On Feb 25, 2019, at 7:15 PM, Tom Pratum <tkp9551_at_gmail.com> wrote:
>
> Hello Everyone-
> It has been awhile since I last posted an update and it might have been thought that the problem was solved, but, unfortunately, that is not the case. I won’t go into details as to why this has continued for so long, but instead would like to throw out a few observations to the group to see if there are any more comments, as previous comments have been very helpful:
>
> 1. On further investigation, it is found that the ability to get a reasonable transmitter pulse out of the HPPR depends greatly on the power level: with default cabling, a power level of -4 is required; below that (less negative) the low power problems indicated in my initial message are observed. The explanation from Bruker (my apologies to any Bruker folks reading this) is ’that is just how it is for this old console’ and nothing more can be done about it. Note that a whole array of things have been tried including but not limited to replacement of the HPPR multiple times, replacement of the TCU multiple times, swapping SGUs and replacement of the BLA2BB, and probe (multiple times). By trying various changes in cabling between the BLA2BB and HPPR, I have been able to get the ‘good’ power level down to -1 (so powers from -1 to -6 work OK), but have not been able to get it below that. The Bruker explanation does not seem reasonable to me, as I too am old but am not limited in the same way this console is given the parameters they are saying it must be operated under. Has anyone out there ever seen this sort of behavior? Note that this is an Advance Microbay console dating from the early 2000’s with a HPPR1 and TCU3.
>
> 2. When I look at the power going between the BLA2BB and the HPPR using a through line tap, I see that, at the point the transmitter output from the HPPR becomes ‘bad’, the reflected power from the HPPR back to the BLA2BB increases markedly, while the forward power is still reasonable. So, the issue is not the BLA2BB (which is also the conclusion I previously came to via other means), but something in the HPPR that, when the power is low, is causing a large impedance mismatch (explaining the effect of changing the cable length). The most likely cause would seem to be the blanking. Here is the real crux of the matter: the blanking occurs in the HPPR, which has been replaced multiple times. The cover module (containing some of the logic for the blanking) has also been replaced once. We would like to be able to say that we see the blanking pulses going into the HPPR, but do not see any (for either channel) - these appear on coax lines soldered into the cover module and are easy to measure. Can anyone say whether we should indeed be able to measure the blanking pulses SPPAH and SPPAX? Unfortunately, Bruker is unclear on this, and while these show up in the HPPR manual I do not see them in the wiring diagram for the console. From the logic presented in the HPPR manual, it appears that the pulse gates, along with either SPPAH/SPPAX or EP are required for blanking - according to the HPPR manual, EP is used to gate the decoupler which just happens to work fine (pulses coming out as decoupler pulses on F2 look fine at all power levels). Can anyone comment on the presence of these SPPAH/SPPAX pulses in a console of this vintage?
>
> 3. One other observation, which I won’t comment on further: when I do see ‘reasonable’ transmitter pulses out of the HPPR, they are approximately 2 usec shorter than they should be - the initial 2 usec of the pulse comes through at low amplitude so it looks like the pulse is 2 usec shorter than what appears out of the BLA2BB.
>
> I know this is a long message, and appreciate very much anyone who has taken the time to read it!
>
> Thank you as always for your help.
>
>
> Tom Pratum
>
>> On Jan 17, 2019, at 11:51 AM, Tom Pratum <tkp9551_at_gmail.com> wrote:
>>
>> An update to my previous post: I first want to thank Robin, Martha, Yong Hua, Dee-Hua, Jerry, Ulrike, Johanna, Bob, and Clemens for responding and making suggestions - what a great group this is! The problem is not solved yet, but we did try quite a few things: checking power supplies and cables, preforming some other suggested rf tests, as well as replacing the HPPR. We really thought the HPPR would do the job, but it did not. We have narrowed our rf tests so that now we are just looking at the behavior at various power levels directly out of the BLA2BB - which has eliminated the BLA2BB - and the same tests at the probe output of the HPPR, which have showed various levels of disturbing behavior.
>>
>> The thought now is that the problem is that the HPPR blanking is not happening; I think this would explain a lot of our observations since, if the HPPR is not being blanked then it won’t look to the transmitter as it should (the lambda/4 in the hot switch won’t look like an open circuit, etc). This is obviously not due to the HPPR itself, since we just replaced it. However, we do not see the blanking pulses SPPAH coming in at J0R in the HPPR (SPPAX is also not appearing, and the X channel also shows some level of disturbing behavior, although not nearly as bad as for 1H). It would seem that this indicates that the TCU is not putting out some of its pulses, while it clearly is putting out some others. We are in the process of getting a TCU to swap in and will see what happens. I will provide another update.
>>
>> Thanks for keeping up this group!
>>
>>
>> Tom Pratum
>>
>>
>>
>>> On Jan 7, 2019, at 6:48 PM, Tom Pratum <tkp9551_at_gmail.com> wrote:
>>>
>>> Hello Everyone-
>>> I have been working with some folks at Southern Oregon University with an older 400 MHz Bruker Avance console, circa 2004. Recently it was observed that the proton (but not X) 90 was increasing about 10% per week, and this - having gone on for months - had increased the proton 90 to a nearly unusable range. In attempting to repair the issue, at one point Bruker recommended replacing the HPPR and/or BLA2BB. It appeared that replacement of both was required to get the 90 back up to its previous value, even though that didn’t make a whole lot of sense to me, so both were swapped out. At that time I observed that, when I put a Bird thru-line rf tap in line with the probe, the 90 was drastically decreased and the pulses that were produced were of very low amplitude, with significant drooping occurring even for short pulses of 10 used. In other words, now placing the tap in line had a large effect on the 90, which was not observed previously, as I had previously used the rf tap to monitor and confirm the power level decrease.
>>>
>>> The instrument then sat for 2 weeks. Upon return the proton 90 was immeasurably long (> 100 uses at pl1 = -3 db). Upon further investigation, I found that I could, by inserting lines of varying lengths between the BLA2BB and the HPPR, get the proton 90 back down to its previous level. I should mention that throughout all of these measurements I have been able to measure the amplitude out of the BLA2BB (with the thru line rf tap) and it appears to be robust - in other words, a lot of power comes out of the BLA2BB when it should, but when the proton 90 is long, it is obviously not transferred into the probe. Throughout these observations the wobb curve looks good - the probe appears to be tuned.
>>>
>>> So, it appears to me that the ‘new’ HPPR has some sort of impedance matching issue; the nature of this may have changed over the break due to temperature changes around the instrument (it is not in a temperature controlled environment). Has anyone else seen something like this? Or does anyone have any other ideas as to what the issue is?
>>>
>>> Thank you in advance for any helpful advice.
>>>
>>>
>>> Tom Pratum
>>> Southern Oregon University ‘Affiliate Faculty'
>>
>
Received on Thu Mar 21 2019 - 08:38:31 MST