Hi Igor,
Okay--I checked and your amplifiers do have input routers--if you look at
the front panel you'll see that there are 3 input connectors on each amplifier
and just the one output. You can try connecting the input on the amplifier
to one of the other ports--they're all the same--but I suspect that's not
going to make any difference. I think it's unlikely that 3 input routers
will have failed the same way on 3 separate amplifiers.
I would have a look at two things--if the REF signals are missing to the SGUs
then the output will be missing (or at least at the wrong frequency) and the
timing connectors on the PSD on the back of the console. You can
disconnect the REF cables from the SGUs or you can check one of the unused
outputs on the reference board. Those should have a mix of frequencies
and be in the range of 1Vpp or so. Those are the signals that the SGUs
use to mix up to the right frequencies, so if one or more are missing then
the SGUs won't put out the right frequencies. I believe the blanking
signals for the amplifiers come from the PSD on the back of the console.
Have a look at the cables coming from the A (and B if it's there) connector
and make sure they're in good shape and correctly seated. Also check
the lights on the PSD and make sure all are lit--I believe there are a couple
of fuses inside that board.
If the REF signal is there, when you disconnect it you'll put the SGU into
an error state so when you get things put back together you'll have to reset
the rack.
Cheers,Mike
On Tuesday, January 27, 2026 at 09:34:11 AM MST, Igor Moudrakovski wrote:
Hi Mike,
Thank you for your comments.
The service webpage for each amp in the the Channel Information has a a column
Input Num that says Via router, but each of 3 SGUs is wired directly to one
of the three amps (SGU(1) RF OUT J2 ->> BLAX1000 IN 1, SGU(2) RF OUT J2 ->> BLAH300 IN 1, SGU(3) RF OUT J2 ->> BLAX300 IN 1). I am a bit lost on this routing.
We did already try a different Ethernet switch without success. We also
checked one by one all the RF cables between the SGUs and the amps, as
well as the LAN cables between the Ethernet switch and the amps.
Regards,
Igor
On 26/01/2026 18:06, Michael Groves via groups.io wrote:
Hi Igor,
Most frequently when we see this error it's just because something didn't
get powered on in the right order so the amplifier just needs to get a new
IP address assigned so that they can communicate again. Normally you
can resolve this by simply power cycling the amplifiers, waiting for them
to boot up, and running cf again. This might not be the cause of your errors.
There are a couple of other things that you can check. When you open up ha
and see the amplifiers listed, can you open the service webpage?
And do these amplifiers have input routers or do they just have one input?
You might check the ethernet switch in the console just to see if you've got
a couple of bad ports or some other communication failure. You might
try a different input to the amplifier--we have seen cases where one or more
of the router inputs fails so when the SGU pulses during that part of cf,
no pulse gets to the amplifier and you get the "no connection detected" message.
I assume there's not a router in the AQS rack that routes the SGU signals to
different amplifiers. If there was one of these I might try bypassing
that as well. How do the cables look between the SGUs and amplifiers?
Normally those don't fail but it might be worth disconnecting them at both
ends and seeing if they're in good shape, and installed correctly. (note--I
would expect that to be the case but it doesn't hurt to be thorough.)
Since you don't have a valid configuration you can't pulse on the amplifiers
but if you were to use nmr_restore you might be able to get back a valid
configuration and just try pulsing on the amps and see what inputs and outputs
are present. If you've got an oscilloscope you can have a look at the
input and output pulses as well as the blanking signal.
Good luck, Mike
On Monday, January 26, 2026 at 09:08:52 AM MST, Igor Moudrakovski wrote:
Hello Everyone,
I have a Bruker Avance III 400 MHz solid-state system circa 2010
(Cento5.11, Topspin 3.6.2), which developed over time a communication
problem between the SGUs and the amplifiers. The CF proceeds seemingly
without any problem, only to end up with an error message
‘No connection between SGUs and amplifiers found! Acquisition will not
be possible. The wiring may be incorrect, or the emergency stop signal may be
active. Please check the hardware and do ‘cf’ again.’
No visible problems with the hardware; all LEDs in front and back are
green, no warning/flashing lights, and no connections have changed since
the instrument was functioning. The ‘ha’ reveals no obvious
problems either.
Last year, Bruker responded to the problem by replacing all the fans and
filters in AQS/3, as well as the amplifiers. That resulted in about two
months of stable work, but apparently didn’t solve the root problem.
Now the instrument is back to the state of ‘No connection between SGUs and
amplifiers…’ Numerous reboots didn’t move the problem.
I would be very grateful to hear all your suggestions for resolving the
issue.
Thank you and best regards,
Igor
--
Igor Moudrakovski, PhD
Max-Planck Institute for Solid State Research
Heisenbergstr. 1, 70569 Stuttgart, Germany
Office 711-689-1963
400 NMR 711-689-1410
600 NMR 711-689-1054
-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.
View/Reply Online (#3164): https://urldefense.com/v3/__https://ammrl.groups=
.io/g/main/message/3164__;!!PvDODwlR4mBZyAb0!X1cwjUV8yj0igIqEUUH6H3xfe3pCnF=
0HXcC_TsPx22aql_-i_7dI4I9brG-tbVdRY4L6IyPtXFJE_K9ikrSEEvU0F63W$
Mute This Topic: https://urldefense.com/v3/__https://groups.io/mt/117471018=
/7559972__;!!PvDODwlR4mBZyAb0!X1cwjUV8yj0igIqEUUH6H3xfe3pCnF0HXcC_TsPx22aql=
_-i_7dI4I9brG-tbVdRY4L6IyPtXFJE_K9ikrSEEmr1BqHU$
Group Owner: main+owner_at_ammrl.groups.io
-=-=-=-=-=-=-=-=-=-=-=-
Received on Tue Jan 27 2026 - 14:27:29 MST