We tested a replacement in-vac WFS for 9MHz/45MHz (ICS ASSY load D1102002-001, S1301248) in preparation for the upcoming HAM1 vent (E2200228).
Rich will write about the details, but in short we have tested that the DC power supply current was good, DC responded as expected, and RF outputs made sense for 9MHz and 45MHz. We used screw-on type ISC AM laser head for this test.
Attached show the LOW (9MHz) and HIGH (45MHz) results measured using the network analyzer. In these pictures, reference trace (green) is Q3 and the live (yellow) is either Q2, Q1 or Q4. All in all these look good in that there's no reason to suspect that they aren't working as intended.
Minor detail is that there are differences not just in overall level because the beam is not centered on the diode and the beam is not circular either, but sometimes in the shape of the traces (for example look at the Q2 HIGH at around the shoulder of 45MHz peak). We've confirmed that both the level and the shape changed when Rich rotated the AM laser head. Don't know why the shape changes but Rich mentioned that the modulation might not uniform across the beam, or maybe we're somehow saturating something (but it didn't change much when we reduced the modulation current by 10dB w/o rotating the head).
Adding to what Keita and Jordan put in yesterday:
1. We looked at the data from the S1301248 WFS that's going to be the replacement device to be installed into HAM1. This was the original manufacturing data from when the head was built in 2014.
2. We checked that the quiescent current draw was the same as that described in the data sheet for this part. Saw 170mA and 240mA on positive and negative 18V rails respectively
3. We verified that each DC path responded both to ambient light (in terms of magnitude and polarity), and the photocurrent produced by the laser calibrator head we use to check the response of ISC photodetectors. The magnitude and polarity of the observed DC response was reasonable and confirms what Keita mentioned that the beam pattern coming out of the calibrator head is elliptic and offset slightly from the center of the four quadrants of the photodiode. Rotating the calibrator head allows the user to verify all quadrants behave similarly.
4. Using the calibrator head and a network analyzer, we were able to take the transfer function of each quadrant both 9.1 and 45.5MHz. Each quadrant's response was reasonable when compared to identical functions of other quadrants. We took the first quadrant we happened to measure and stored this to a reference trace on the network analyzer. The remaining quadrants were then superimposed onto the reference trace to make a comparison. While there are differences in the absolute amplitude of each trace (due to the elliptic nature of the calibrator head beam pattern), the overall shape of each trace was reasonable when compared to the other quadrants.
For the new WFSA noted above (D1102004-V6 S/N 005 S1301248), I wanted to update the ICS Assy Load to make more sense. So the ICS Assembly Load link above no longer works, but the new one is now this:
Just a note on these in-vac WFS: The Body for these assemblies is D1102004 and it has its own s/n stamped on it from when it was machined. However when a WFS Assy has its electronics installed, thus determining its operating frequency, and it is locked at this frequency because the WFS is then welded shut for in-vacuum install. When this happens, the WFS (assembly really) is then assigned a new s/n (such as the DCC-generated serial number for the electrical innards of this WFS---S1301248 in this case). I did not know this when I made this Assy Load, so I just picked a random s/n of 001---using the DCC s/n is a better and cleaner way to identify an installed-WFS assy.