Hello,
In the user guide Figure 5-1, you can see the location of R5 and R8 in the schematic. These components are not populated in the EVM when it is shipped.
Please see the snippet of schematic below where I have added marks to show where the trace cuts should be done on the schematic. The cut described in Figure 1-1 is shown with a blue mark below, and the cut described in Figure 1-2 is shown with a green mark. and don't forget to replace the ferrite bead with a 0 Ohm resistor for the most accurate stability measurements.
These changes allow you to inject a signal across the 49kOhm resistor as described in the user guide.
Hi Steve,
Thank you for your reply on this + on going support on this thread.
I've also run over the trace cuts myself - looking more closely at the SLVUBX2A EVM board datasheet and referring to the BSR077B EVM gerbers, I also agree we are cutting the trace between VO and L1 as you suggested previously. However, my read is that the other cut is actually disconnecting R8/C13 node from VOUT_FILT. The screenshot below shows that R5 is still connected to R8, R1 and C13 - I think this means we are actually cutting the R8/R5 node from VOUT_FILT as this is the long trace that connects after FB1 to the C7 output bulk.
I think this means we're doing something like this:
I suspect I'm still missing something as populating R8=0R would short out the insertion resistor R5 and the feedback network doesn't appear to now be connected to VOUT? It would be great if you could review and confirm if you agree with this.
If so, please could you review the correct test procedure and make sure we understand what's going on as I'm losing a lot of time on this and I feel this should be properly documented.
As a minor note, for benefit of others, could you also confirm you mean 49Ohm for the insertion resistor rather than 49kOhm?
Many thanks
Hello,
To answer you last question first, yes, it is a 49Ohm resistor, not 49kOhm.
As to the VOUT sense, you can see on the bottom layer that installing the 0Ohm resistor for R8, it connects the VO pin feedback to VOUT through R5 (the 49 Ohm resistor). This moves the sense point to near the primary LC filter output caps C4 and C5. I see where my previous comment was confusing by stating VOUT_FILT, when the sense point is now closer to the primary LC filter and the VOUT_FILT sense point is not used, as you pointed out.
I hope this clears up the connections.
Hi Steve,
I can now see that the first trace cut is not only disconnecting VO from VOUT but also R8 from VOUT which isn't obvious from the schematic alone given the number of nets that connect to this node. I've tried updating the SCM to more accurately reflect the routing topology - I think the below makes it much clearer what is going on here but I'm open to your opinion here:
Please could you put in a request to have the EVM documentation updated to make this clearer. I'd also strongly encourage the EVM boards to have population options to prevent the need to do trace cuts if possible.
Many thanks
