AMD UCM-UCSI Device Error: Modules on left side dead no charge, right side no video

December 27, 2024 Error in the device manager: AMD UCM-UCSI Device Error 43

  • Charging on the left side not possible
  • Modules on the left side without function
  • on the right side no HDMI, no display port, no video signal via USB-C, no video signal via USB docking station, SD card is intermittently logged on and off, USB-A is intermittently logged on and off

The error probably occurred after connection to a mobile dock with charging current via an Anker USB-C charger.

  • PD-Contoller 1 version in BIOS setup 0.0.1E
  • PD-Contoller 2 version in BIOS setup 0.0.00

Support suggestions:

  • Install driver package
  • Repeat BIOS update
  • Perform mainboard reset, film LED codes and send to us.

unfortunately no improvement

When updating the BIOS, the framework wheel runs slowly during the update, stops briefly at PD controller 2 and then runs very quickly. The green tick in the bottom right corner is missing. Nevertheless, the message appears at the end: “Update completed successfully”.
Afterwards, version 0.0.00 is displayed again in the BIOS setup for PD controller 2.

I have documented this and sent the video to framework support

I tried to update the BIOS via EFI-USB, but the same behavior is shown: again version 0.0.00 on PD-Controller 2

  • I think that a BIOS downgrade or update with another version could help, but downgrade is not possible and there is no update higher than 3.05 available.

I have documented the suggestions in the video and sent them to support

Support last contacted me on 15.01.2025 and requested a video again, in which it can be seen that charging is not possible on the left side. Nothing happened after that.

I urgently need a working connection for the color-calibrated monitor (image and video editing as a freelance graphic designer).

BIOS mismatch would certainly cause half of it to not function correctly when it is expecting matching versions for both sides.

It is not always widely documented, though some of the BIOS updates only work for ONE half of the board; then it has to be repeated for the other half that was not updated and I thought that the power had to be supplied to the opposite side of the board for it to flash the side that did not get flashed the first time.

If I am remembering correctly these were the updates for people using Linux and not all the Linux bios updates could update the same way as the Windows updates because that is what Insyde provides first to its clients as a BIOS update path.

Hopefully support can get you straightened out and possibly downgrade the other side. Though as I am typing this it just occurred to me PD2 might have failed or have a corrupted BIOS. Very rare to see a NULL or 0.00.00 version for anything other than a device that is unresponsive or nonexistant. This might explain why the update runs quickly after it gets to PD2.

Maybe support or the dev team can package a BIOS update just for PD Controller 2 with some verbose logging to see what is going on. Though this might be a fairly big ask and more trouble vs. replacing the mainboard.

Support would probably have already suggested it, but you don’t mention it.
Try bios update with power plugged into the left side of the laptop. Repeat the bios update with the power plugged into the right side of the laptop.
Do the bios update with about 80% battery.

You describe the procedure for standalone mode.

I tried that, but two conditions make it impossible:

  1. BIOS update can only be performed with the charger connected

  2. the USB-C port on the left side must accept the charging current, which is not the case.

OK. If you have tried it already, my guess is that you need a new mainboard.

Support has contacted me and has come to the conclusion that a new mainboard will be sent.

I have accepted this and I would also accept a refurbished mainboard (my suggestion). Hopefully my defective mainboard can be repaired and returned to the cycle.

Charging via USB-C is very convenient. However, the boundary conditions are complicated. The power supply communicates with the cable and learns what the cable can do. It then communicates with the laptop and agrees a charging current. If the power supply does not adhere to the specification, the laptop can be damaged. In my case, I think I know that the damage occurred when a second device (smartphone) was connected to a second port of the USB-C power supply The charging current went from the Anker power supply via the Hyperdock to the laptop 16 and I had tested this extensively and it worked well.
If this is the case, Framework is not at fault (if Framework strictly adheres to the PD specification). Then it would be either the power supply or wrong information of the cable or both.
In my professional environment, there are a lot of problems with USB-C. The cables are usually not labeled correctly or not labeled at all and only transmit power or USB4, the power supplies sometimes do crazy things (there are some that have a USB-C port but are only suitable for one device, with a fixed ratio between voltage and current and immediate damage to other connected devices). Therefore, professionals usually have only one type of cable in their equipment and with a special color (orange) so that they can immediately see that everything is connected correctly. and they only use one type of power supply. It would be very desirable if all manufacturers took the PD specifications seriously and also labeled the cables correctly…

In my case, the damage occurred at home and I probably didn’t do anything wrong (I hope). I will continue to use the Hyperdock, but now the original Framework power supply is connected to the dock and no other device (the Framework power supply only has a USB-C port)…

I was unable to do this because a BIOS update can only be carried out with the charger connected on this side. I could imagine that this procedure (standalone mode for BIOS update) could work. However, no video signal is output on the left side of the mainboard (and none on the right side either) and charging on the left side isn´t possible, so I could not check this update either.

However, it could also be an error in the PD controller (hardware) because version 0.0.00 is not known.

Hi. Where did you read about the USB-C charging protocol?
My understanding is a little different.
As far as I understand it, the negotiation is regarding the limits and profiles. I.e. it negotiates a voltage, and a current limit for each profile. The laptop then, chooses a profile, draws as little or much of the current as it wishes.
Profiles might be something like this:

  1. 5V, 5Amp max.
  2. 20V, 5Amp max
  3. 36V, 4Amp max.
  4. 48V, 3Amp max.

For example, my FW16 plugged into a FW 180W charger generally sits at about 36V, and then draws as much or as little current as it wishes.
The FW 180W charger has these profiles:

  1. 5V, 3A
  2. 9V, 3A
  3. 15V, 3A
  4. 20V, 5A
  5. 5V, 3A
    I don’t know where the 36V , 5A profile is.

But thinking of the cause of your problems.
The FW16 can handle 48V at 5V or 240W, so I don’t see how a dock could damage a FW16.
I can understand how a laptop might draw too much current from the PSU, but I cannot see how a PSU can inject too much current into the laptop.
The only way I can think of is if the PSU outputs more that 48V, but I think that is unlikely.

We are of the same opinion and have the same level of information, I think. It is certainly a bad thing if the charging current exceeds 5 amps and thus possibly burns up the PD controller.

The maximum current must not exceed 5 amps.
A maximum of 240 watts is possible (i.e. 48 volts at 5 amps). This is the specification for PD 3.2. As far as I know, the permitted voltages are 5, 9, 15, 20, 28, 36, 48 volts

However, there are exceptions that could damage a device:

  • Some manufacturers (Dell, for example) have already proprietarily extended the existing protocols (prior to the adoption of version 3.2), in the case of docking stations via TB3. This includes the power of 90 watts at 17 volts which exceeds the 5 amps, if I am not mistaken, and 17 volts does not exist with PD (except with Dell). However, only Dell devices have demanded and received this power, I believe the limit for third-party devices was 60 watts and Dell adheres to the specification and third-party devices cannot be damaged in this way. It is only with its own devices that Dell does not adhere to this. Dell’s customers are then looking for power banks with a similar charging capacity of 90 watts “Dell protocoll” and these power banks are not allowed to exist. Dell has not done itself any favors with this.

  • Supplied power supply units with USB-C plug for a special device and a fixed current of 8 amps at 12 volts. This can damage a device immediately and such power supply units should not be imported.

  • Plug-in chargers that initially negotiate the correct protocol, but no longer adhere to it when another device is charged at another socket. Unfortunately, these power supply units do exist. You should also never use a combination cable with USB-C, mini-USB and Lightning.

All I’m saying is that laptop manufacturers, for example, must be able to rely on manufacturers adhering to PD 3.2. It would be bad business if the manufacturers of expensive laptops had to supply a new mainboard under warranty because a manufacturer of a charger does not comply with PD 3.2.

It may be a coincidence, but my laptop has been faulty ever since a charger from another manufacturer supplied the laptop and then a smartphone was connected to the second USB-C port of the charger.
This probably has nothing to do with the charging current passing through the Hyperdock, as I have been using this dock for several months (but without a second device connected to the power supply). But i don´t want to test this again…

The charger in question is now allowed to supply my battery charger. This charger also had a very loud coil beeping and the Schuko adapter (EU) is only loosely attached. The charger falls out of the Schuko adapter when it is plugged into a wall socket. A faulty design, I fixed the adapter with adhesive tape…

1 Like

Replaced the mainboard and everything is running again.

The instructions for replacing the mainboard are helpful and the comments below are important.

1 Like

Hey @Michael_Graf,

Good info to know on the Framework 16. Happy to hear you got your mainboard replacement and it went smooth to replace. I suspected a hardware error when the one side reported a version of 0.00. These days, a “refurbished” mainboard is as good as a new one; maybe better since it has been tested probably more than once before it was sent out. About the only components that might not be ideal would be WiFi cards (radios can degrade over time), and video cards if run improperly (i.e. overclocked). :grinning:

This topic was automatically closed 2 days after the last reply. New replies are no longer allowed.