- Joined
- Feb 15, 2014
- Messages
- 20,194
And I'm back with photographs and scope shots.
Let's move on to something more interesting, and keep in mind that this board is the same as the one from the scope shots upthread, after the repair - both clock lines were pulled up to 3.3. V by around 140-150 Ohm:
Yellow is the TPM LPC clock line, previously working (compare with post #164 above), Magenta is the BMC clock line, previously dead (I'll spare you the scope shot, the results are all over the place - bottom line, it was dead).
The TPM clock still looks decently nice, though it doesn't go down to 0 V. Shouldn't be a huge deal, but it shows that the hard pull-up does have an effect on healthy lines.
The BMC clock is not great, but the BMC makes it work. Probing on this one is also a bit more dodgy, since I have to use the test point helpfully identified by @RZSN, which does not have any particularly convenient grounds nearby, so antenna lead over to the VGA connector is about as good as it gets. Still, it looks like a slightly stronger pull-up might clean things up somewhat - though I'd much rather use the weakest that makes the CPU and BMC happy.
I mean, one of them doesn't even look terrible, it jut ended up upside-down. The other one is so gnarly I can't even visualize if it's okay, but buzzing out all nearby pads shows that everything is connected as desired. How well is a different matter...
Let's move on to something more interesting, and keep in mind that this board is the same as the one from the scope shots upthread, after the repair - both clock lines were pulled up to 3.3. V by around 140-150 Ohm:
Yellow is the TPM LPC clock line, previously working (compare with post #164 above), Magenta is the BMC clock line, previously dead (I'll spare you the scope shot, the results are all over the place - bottom line, it was dead).
The TPM clock still looks decently nice, though it doesn't go down to 0 V. Shouldn't be a huge deal, but it shows that the hard pull-up does have an effect on healthy lines.
The BMC clock is not great, but the BMC makes it work. Probing on this one is also a bit more dodgy, since I have to use the test point helpfully identified by @RZSN, which does not have any particularly convenient grounds nearby, so antenna lead over to the VGA connector is about as good as it gets. Still, it looks like a slightly stronger pull-up might clean things up somewhat - though I'd much rather use the weakest that makes the CPU and BMC happy.
Last edited: