# Oscilloscope LEDs flickering on/off, changing colors - possible overheating

#### Anna Swinemar

Joined Oct 31, 2017
14
This may be a long shot but I have a 4CH 350MHz Rhode & Schwarz Osc. The LEDs on the right side of the scope seem to be overheating and flickering either ON/OFF or changing colors. Each Channel button is supposed to be a different color which matches to the color of the trace on the screen. When I spray cold spray or compressed air, they seem to come on for a bit and start working again.

So far, I have replaced the power supply because I found it to be faulty as I noticed the fans weren't running. I also replaced the SMD LED drivers. The fans now turn on but there is no change in the LEDs. In the video, I sprayed compressed air and all of the LEDs turned off momentarily. (freaked me out!)

The scope still functions fine but the LEDs on this scope was an incredible feature. Every time you press a Channel button all of the measurement buttons light up the same color so you always know what channel you're adjusting.

Anyone have any advice of what may be causing this?

#### MrChips

Joined Oct 2, 2009
19,382
Sounds like a power supply problem or a heavy drain on the power supply and is pulling the power supply down.

#### Wuerstchenhund

Joined Aug 31, 2017
187
I guess this is a RTM2034? In general it helps tremendously to mention the exact model of your scope to get help as there's more than one 350Mhz 4Ch scope made by Rohde & Schwarz.

It's been a while but I remember we had a RTM2054 acting up and if I remember right the mainboard had to be replaced - at a pricely sum of course.

Since the rest of the scope seems to work fine I'd probably try to follow the LED signals back to the controller they originate from, which is probably on the PCB that holds the knobs and buttons.

Have you tried asking R&S for the schematics? Sometimes they can be quite forthcoming to hobbyists.

#### Anna Swinemar

Joined Oct 31, 2017
14
I guess this is a RTM2034? In general it helps tremendously to mention the exact model of your scope to get help as there's more than one 350Mhz 4Ch scope made by Rohde & Schwarz.

It's been a while but I remember we had a RTM2054 acting up and if I remember right the mainboard had to be replaced - at a pricely sum of course.

Since the rest of the scope seems to work fine I'd probably try to follow the LED signals back to the controller they originate from, which is probably on the PCB that holds the knobs and buttons.

Have you tried asking R&S for the schematics? Sometimes they can be quite forthcoming to hobbyists.
I have not asked for the schematics yet, but I will now. And yes, I was trying to avoid sending it back to R&S because I know it's going to be \$. It's and RTM2034.

The board I am spraying with the cold spray is the LED control board. Unfortunately, it's not a permanent fix. After about 5 mins, the LEDs start flickering again.

#### ebp

Joined Feb 8, 2018
2,332
It sounds like it might be a flakey connection - anything from a wire/cable connection to a solder joint on a component.

It can be quite tedious, but since cold spray changes things I would suggest spraying very small areas one by one to see if you can localize the problem. You might try a binary search method - spray half the board, then half of the bad area, then half of ... Changing colors suggests to me that it likely has something to do with digital data that sets the state of the actual LED drivers. With so many things done serially these days it might be something like a register IC - or even a small microcontroller. Sometimes you'll get situations where temperature changes the threshold of a digital input just enough to change behavior when the real fault is somewhere else.

The usual first step in any troubleshooting is to scope power supplies near the destination - but that's a bit of an issue if you only have one scope

#### Wuerstchenhund

Joined Aug 31, 2017
187
I have to concur with ebp here, the first thing to check are the PSU voltage lines with a scope, and if they're fine then I'd have a look at the IC that controls the LEDs. If that's fine, too, then check any flat cables and connectors. If that turns out without fault then it's very likely something like a cracked solder joint, which should be identifyable through sectional cooling of the PCB.

If the front panel PCB is separate from the mainboard, and if the fault is with the front panel PCB then I'd just ask R&S for a price for the replacement. R&S offers refurbished parts for many of its instruments so there might be a cheaper alternative to a new part.