Erik, Jonathan, EJ, Dave:

On Monday 21jun2026 we did some more investigation of the General Standards 20bit-DACs in h1iscex and h1susb2h34.

h1iscex:

Following my successful loop-back testing of this 20bit-DAC to the PEM ADC, EJ confirmed that this 20bit-DAC passed the Monday tests. This card is still a bit of mystery as to why it is working with a MPS=256B. It is a unique 20bit-DAC, older than all the others (it is 2015, all the others 2019 or later). It has a 4th region (Region1) on the lspci -vv scan missing in the other DACs.

h1susb2h34:

Monday's testing showed that both of the 20bit-DACs on this frontend are non-operational. The first because it shares an Adnaco with the Timing-Card, the second because it shares an Adnaco with the LIGO-DACs. In both cases the LIGO card is forcing MPS=256B which causes the 20bit-DACs to drive no output voltage (constant FIFO=empty).

My original SUSAUX testing was faulty, I was looking at the stop stages of PR2, SR2 but the 20bit-DAC drives M2,M3 for these suspensions. I was able to verify that these lower stages have not been driven since the January 2026 upgrade.

h1sush12:

As an aside, I took a look at the MPS for h1sush12. This has no 20bit-DACs, it is all LIGO-DAC. This is a W3323 computer, the additional cores were needed for the large number of models. Strangely this machine is running with MPS=128B on all cards, even though it could run at 256B, presumably resulting is a loss of DAC performance. We will investigate this on the test stand's W3323.