This is why I don't like working on subwoofer amplifiers: I was working on a John Bowers Active 1 today and was not aware that the output neutral speaker line sits at -40VDC.
Usually with some McIntosh amplifiers, I have to ground the analyzer to stop the relays from clicking periodically, and I thought this was the case with this subwoofer amplifier.
It blew out something, possibly a fusible link inside my UPV audio analyzer. Now that input no longer grounds when I set analyze to 'grounded' in the menu.
I'm looking for a schematic of the input circuitry to find out where this fuse is located so I can replace as needed. At least I hope it is just a fuse and nothing more involved.
Usually with some McIntosh amplifiers, I have to ground the analyzer to stop the relays from clicking periodically, and I thought this was the case with this subwoofer amplifier.
It blew out something, possibly a fusible link inside my UPV audio analyzer. Now that input no longer grounds when I set analyze to 'grounded' in the menu.
I'm looking for a schematic of the input circuitry to find out where this fuse is located so I can replace as needed. At least I hope it is just a fuse and nothing more involved.
I traced pin 1 on the XLR input to the analyzer and it goes to the point on this relay that I marked with yellow ink.
The arrow points to the output of the relay contact. In channel 2, this makes continuity when the analyzer is set to "ground" and is open when set to "float". On channel 1, it is always open. So I think the relay contact must have melted.
It's SMD and in a real tough spot to get at due to the partition metal separating the channels.
The arrow points to the output of the relay contact. In channel 2, this makes continuity when the analyzer is set to "ground" and is open when set to "float". On channel 1, it is always open. So I think the relay contact must have melted.
It's SMD and in a real tough spot to get at due to the partition metal separating the channels.
Do you need service manual?
If you have already found that the relay is broken, it is down to a practical problem of how to replace it.
Is it possible to (easily) remove the metal shield?
If not, perhaps you can remove the relay with a hot air gun.
Mounting a new relay may be easier if you place it on its side and add 4 small pieces of wire to connect the terminals that are "in the air". Perhaps easier if a through hole version of the relay is used.
Not the prettiest solution, but it will probably work.
If you have already found that the relay is broken, it is down to a practical problem of how to replace it.
Is it possible to (easily) remove the metal shield?
If not, perhaps you can remove the relay with a hot air gun.
Mounting a new relay may be easier if you place it on its side and add 4 small pieces of wire to connect the terminals that are "in the air". Perhaps easier if a through hole version of the relay is used.
Not the prettiest solution, but it will probably work.
The service manual is always good to have, considering R&S charge a flat $5170 + shipping to Munich, Germany for the smallest repair.
The shield may not be removable without removing the entire board, which is the footprint of the chassis and has much connected to it.
I'm leaning toward the hot air gun to R&R the relay. Just need to get the IM03 part, not the IM03 with suffixes (different specs).
Can't jury rig because this is RFI-compliant layout and would affect the performance of the unit.
The shield may not be removable without removing the entire board, which is the footprint of the chassis and has much connected to it.
I'm leaning toward the hot air gun to R&R the relay. Just need to get the IM03 part, not the IM03 with suffixes (different specs).
Can't jury rig because this is RFI-compliant layout and would affect the performance of the unit.
I bought a hot air workstation, solder paste and practiced for two days on some old VGA card.
I replaced the relay on the damaged channel tonight. That gets that channel working again.
But before I replaced the channel 1 grounding relay, I re tested both channels. Last week, channel 2 was still able to ground its input.
Tonight, it did not. I found that the voltage across the field coil was stuck at 4.87 volts. On channel 1, it goes from 4.92 to .187 when input is grounded.
I went ahead and replaced relay for channel 1 and confirmed that the input pin 1 does in fact ground when it did not before.
Back to channel 2, I have no grounding ability there now, where it WAS working last week when I first started troubleshooting.
I decided to replace channel 2 relay just in case. However, it still does not work now. So this is a failure upstream in the logic somewhere, that happened in the week that passed since the original incident that blew out channel 1's relay contacts.
So ironically, the "damaged" channel is now working and the "good" (formerly) is now bad. I have no idea why channel 2 relay control logic went bad DAYS after the incident that damaged relay in channel 1.
Without a service manual at this point, I can't do anything but start shotgunning ICs and that's not a good idea. Other than that, it's off to Germany and $6K out of pocket to have this thing fixed properly.
I replaced the relay on the damaged channel tonight. That gets that channel working again.
But before I replaced the channel 1 grounding relay, I re tested both channels. Last week, channel 2 was still able to ground its input.
Tonight, it did not. I found that the voltage across the field coil was stuck at 4.87 volts. On channel 1, it goes from 4.92 to .187 when input is grounded.
I went ahead and replaced relay for channel 1 and confirmed that the input pin 1 does in fact ground when it did not before.
Back to channel 2, I have no grounding ability there now, where it WAS working last week when I first started troubleshooting.
I decided to replace channel 2 relay just in case. However, it still does not work now. So this is a failure upstream in the logic somewhere, that happened in the week that passed since the original incident that blew out channel 1's relay contacts.
So ironically, the "damaged" channel is now working and the "good" (formerly) is now bad. I have no idea why channel 2 relay control logic went bad DAYS after the incident that damaged relay in channel 1.
Without a service manual at this point, I can't do anything but start shotgunning ICs and that's not a good idea. Other than that, it's off to Germany and $6K out of pocket to have this thing fixed properly.
UPDATE: Good News!
The channel that was not switching was not doing so because the last edit to my setup had it set for analyzer channel 1 only as I was working on a mono amp last.
I re loaded the setup and now both channels ground. All affected relays are replaced, so hopefully I've eliminated any partly damaged contacts in channel 2 as well.
My only screw up with getting gull wing leads instead of J leaded parts. Everything else in this PCB is gull wing and the slight protrusion of the leads on the IM03 relay made me think it was gull wing. I found out it was J lead after removing it. Shouldn't make any difference as the pads seemed long enough to cover the leads anyway.
I've just got done putting 10Ω resistors in series with the ground lead to the UPV, as insurance in case anything goes "hot chassis" in the future.
The channel that was not switching was not doing so because the last edit to my setup had it set for analyzer channel 1 only as I was working on a mono amp last.
I re loaded the setup and now both channels ground. All affected relays are replaced, so hopefully I've eliminated any partly damaged contacts in channel 2 as well.
My only screw up with getting gull wing leads instead of J leaded parts. Everything else in this PCB is gull wing and the slight protrusion of the leads on the IM03 relay made me think it was gull wing. I found out it was J lead after removing it. Shouldn't make any difference as the pads seemed long enough to cover the leads anyway.
I've just got done putting 10Ω resistors in series with the ground lead to the UPV, as insurance in case anything goes "hot chassis" in the future.
A UPV service manual doesn't exist outside of a secure server @ R+S in Germany. If you know otherwise, I would be sooo happy to find out! We have 3 UPVs here. 1 of them has a minor issue. It had major issues such as shorted MLCCs, blown VRegs, blown fuses etc (most likely triggered by shorted MLCCs). Not wanting the £3.7K 'fixed price repair' at R+S, we managed to get these faults fixed ourselves. But now, the software is garbled... We could transfer an HD from a working machine and see if that cured it, but seeing as both are in use daily, it feels like a risk. I am wondering that, considering the UPV is no longer made, we might have to pony up. It's about £800 to have it assessed, and then £150 p/hour + parts to repair. Or you go for the £3.7K fixed repair, which has a year guarantee. Not knowing how longer they will support it for, makes me tempted to go for the FPR. I will cry when I get the bill, though!
NB - R+S have a full office in the UK, but they still send UPVs to Germany. Good luck with that manual!
NB - R+S have a full office in the UK, but they still send UPVs to Germany. Good luck with that manual!
You can probably also make a copy of a working HD. Perhaps even to an SSD version. I have seen that done on an R&S spectrum analyzer.
I powered my UPV last night for likely the first time in 6-7 years. This process generated a "CMOS checksum error - Defaults loaded"
screen message. Any insights for a gnu-B to effect repair? I had the UPV repaired and cal'd at R&S in Columbia, MD in 2014.
It's great to know that the UPV is still known to be in the wild!
screen message. Any insights for a gnu-B to effect repair? I had the UPV repaired and cal'd at R&S in Columbia, MD in 2014.
It's great to know that the UPV is still known to be in the wild!
A friend of this has tried this for me several times, using several 'ghosting' programs. He fixed an AP a while back and has a background in doing IT for military firms. I don't remember the specifics, but he couldn't do it - and he came over and tried several times. We tried putting a working disk into a caddy and cloning it that way, but when we put the cloned disk into a machine it wouldn't boot. The original disk is tiny, so whether your machine needs fixing or not, it would be good to make it larger.
There is a separate thread: https://www.diyaudio.com/community/...pp-audio-analyzer-thread.393563/#post-7541353
Above thread suggests that 2015+ machines can have an SSD, but not before. Why is that? I recall Amp Experts' machine was rebuilt by Jonathan Zepp and it has an SSD. But he went to extreme lengths, even taking the main CPU board out of a different piece of R+S gear and then adapting it.
What is the size of the original disk?
As far as I remember, the spectrum analyzer would not recognize "large" disks (> 8GB). An 8GB SSD was used to repair the spectrum analyzer, which is probably around 20 years old. And yes, it needed a new battery as well.
The data was copied using Acronis.
As far as I remember, the spectrum analyzer would not recognize "large" disks (> 8GB). An 8GB SSD was used to repair the spectrum analyzer, which is probably around 20 years old. And yes, it needed a new battery as well.
The data was copied using Acronis.
Original disk is tiny. From memory, it's something like 40GB (thread linked above suggests so).
Do you remember which CPU board was in your spectrum analyser? FMR6?
If someone were to take time to learn how to service the old UPD / UPL / UPV, there would be a good little business in it for them. I have never heard of a service manual for any R+S AA. And if you get a repair quote from them, you will know why!!!!!
Do you remember which CPU board was in your spectrum analyser? FMR6?
If someone were to take time to learn how to service the old UPD / UPL / UPV, there would be a good little business in it for them. I have never heard of a service manual for any R+S AA. And if you get a repair quote from them, you will know why!!!!!
The best thread I've found on the UPV is here: https://www.eevblog.com/forum/repair/rohde-schwarz-upv-audio-analyzer-teardown-and-repair/
From what I've seen, they start out at FMR6 and go up to 9.
I love my UPV. As others have written, it's a workhorse. It gets about 100x more use than the SYS-2722 or the DScope. It is used day in, day out. I've tried feeding it with the oscillator from the AP and it gets the same THD+N of -115dBV (the AP osc is about 2dB better than the B1). That is good enough for 99.9% of my work. If I decide to get into the DAC arms race, I might need something better, but that's the only thing I can think of that I'd need a 555B for.
From what I've seen, they start out at FMR6 and go up to 9.
I love my UPV. As others have written, it's a workhorse. It gets about 100x more use than the SYS-2722 or the DScope. It is used day in, day out. I've tried feeding it with the oscillator from the AP and it gets the same THD+N of -115dBV (the AP osc is about 2dB better than the B1). That is good enough for 99.9% of my work. If I decide to get into the DAC arms race, I might need something better, but that's the only thing I can think of that I'd need a 555B for.
The UPL service manual, and other information, is actually available:
Rohde Schwarz R&S UPL Audio Analyzer Renovation
It seems like the UPD information was once available as well, but the links are dead by now:
Rohde & Schwarz UPD - HELP! Need docs, files & repair advice
Rohde Schwarz R&S UPL Audio Analyzer Renovation
It seems like the UPD information was once available as well, but the links are dead by now:
Rohde & Schwarz UPD - HELP! Need docs, files & repair advice
Classic case of dead CMOS backup battery. There's a small Lithium battery on the PC motherboard that stores the configuration info. Usually last about 10-12 years.
Agreed ... lithium battery failure is the likely culprit. I'll open my UPV and replace the CR2032 battery. Good/excellent probability at resolution.
Thank you for the reply!
Thank you for the reply!