Hi there,
Thank you to everyone who responded to my request for help with the SGU problem. It turns out that the problem was due to unstable output from the REF-22 board which provides reference frequencies for various components, including the SGUs. I had tested the outputs from this board and the amplitudes all seemed comparable to those from the REF board in another console and so I initially ruled it out. However, the Bruker engineer visited today and when he pulled the board from the rack he noticed it was cold to the touch. He suspects that the temperature regulation of the oscillators has failed leading to unstable output. Swapping the board fixed the problem.
It might be worth noting that the REF board we used to test was labelled as a REF600-22, but it still appeared to function O.K. in the 700 console, at least for testing purposes.
Regards,
Pat.
________________________________
From: Edwards, Pat [P.J.Edwards_at_massey.ac.nz]
Sent: Tuesday, 8 November 2016 10:44 a.m.
To: ammrl_at_ammrl.org
Subject: AMMRL: Low output from all SGUs on a Bruker Avance console
Hi there,
I am hoping someone might be able to help diagnose problem with our Bruker 700 MHz console.
We have a 12 year old 700 MHz Bruker Avance console with an AQS rack which includes four signal generation units (SGUs). The output from all four units is much lower than the expected 1 Vpp _at_ -6 dB. I am measuring ~170 – 300 mV depending on the SGU. This means that I can no longer get acceptable pulse widths even after compensating by reducing the transmitter attenuation (the console has 1 X BLAXH and 2 x BLAX500). I was wondering if anyone has experienced a similar problem and if so then how did they manage to fix it?
Some relevant info and the tests I have performed so far are listed below.
(1) All the power status LEDs on the AQS are lit, but nevertheless I have tried exchanging the AQS power supply modules from a working console, but the problem persists.
(2) I have tried exchanging SGUs from another console. The SGU which tests fine in the donor console gives a low output as soon as it is placed in the affected console. Conversely the SGU from the affected console works fine in the donor console.
(3) I have exchanged the FCU, TCU, Both routers (not at the same time), the RCU and the HADC/2 boards, all with no improvement.
(4) The AQS fans were replaced a few years ago. I have not pulled the AQS from the rack to see if they are all functioning, but there seems to be plenty of air passing through the rack. The SGU temperatures reported by UniTool seem fine (all ~35 deg C). The DDS temperatures are all 55 deg C, which I think is spec. The temperature in the lab is a steady 21 deg C.
(5) I have tried a fresh install of Topspin 2.1.8, using no cortabs. Again, this makes no difference.
(6) The PSM3 power module has blown a fuse maybe 3 times in the past few years. This has not happened recently, but hints at something drawing too much power. Perhaps this is related?
I am making my measurements on the SGUs at 100 MHz with the power level set to -6 dB, using a 300 MHz scope with a 50 ohm load taking the output directly from the J2 output of the SGU (there is no attenuator on the output). Testing using the identical setup and parameters on another console gives the expected ~1 Vpp from the SGUs.
I am now at a loss as to what to try next. Any advice would be very much appreciated.
Regards,
Pat Edwards,
Massey University.
Received on Wed Nov 09 2016 - 22:17:56 MST