I am typing this with the OSDEHA driving my Stax X9000 on my ears, now in stereo. The amp is just great. It's dead quiet, and the music is beautiful!
There's still a lot of work to do. One of the major construction sites is the PSU, which still lives in a very temporary "chassis". However, before working in this, I need to hunt down a little gremlin. If the electric heater in my workshop turns on, the DHT bias drops instantaneously. It goes back to normal when the heater turns off. I somehow suspected the heater causes a little drop in the mains voltage, which then reduces the B- or B2- voltage output from the PSU (which should not happen, since the PSU uses a voltage reference). However, the DVM says the PSU voltages are rock stable, so there must be some higher-level magic happening here. If anyone has ideas, please shoot!
There's still a lot of work to do. One of the major construction sites is the PSU, which still lives in a very temporary "chassis". However, before working in this, I need to hunt down a little gremlin. If the electric heater in my workshop turns on, the DHT bias drops instantaneously. It goes back to normal when the heater turns off. I somehow suspected the heater causes a little drop in the mains voltage, which then reduces the B- or B2- voltage output from the PSU (which should not happen, since the PSU uses a voltage reference). However, the DVM says the PSU voltages are rock stable, so there must be some higher-level magic happening here. If anyone has ideas, please shoot!
So it turns out the DHT bias issue is related to changes in the mains voltage. I looked at the wrong DVM yesterday. No high-level magic involved.
What happens is that if the mains voltage drops a bit, the B- voltage output from the PSU will also drop, while the B2- stays stable. I guess the dropout for the voltage reference in the B- PSU is too low. This will surely need some work down the road.
However, the real issue is that the DHT grid bias really shouldn't be so sensitive to small changes in the rail voltages. Looking at the current circuit (see post #63), I see that if B- drops but B2- stays stable, the DHT grid bias will not move along with the B- / cathode voltage. This throws off the grid-cathode voltage and hence the bias of the DHTs. To prevent this from happening, I revised the bias voltage circuit to use a Zener that provides a fixed voltage relative to the B- rail / cathode. See attached (yes, there were a few other small changes in the circuit, but they were not related to the DHT grid bias).
It would be great to get some feedback on this from the brains around here before I start messing with the build.
What happens is that if the mains voltage drops a bit, the B- voltage output from the PSU will also drop, while the B2- stays stable. I guess the dropout for the voltage reference in the B- PSU is too low. This will surely need some work down the road.
However, the real issue is that the DHT grid bias really shouldn't be so sensitive to small changes in the rail voltages. Looking at the current circuit (see post #63), I see that if B- drops but B2- stays stable, the DHT grid bias will not move along with the B- / cathode voltage. This throws off the grid-cathode voltage and hence the bias of the DHTs. To prevent this from happening, I revised the bias voltage circuit to use a Zener that provides a fixed voltage relative to the B- rail / cathode. See attached (yes, there were a few other small changes in the circuit, but they were not related to the DHT grid bias).
It would be great to get some feedback on this from the brains around here before I start messing with the build.
Attachments
Last edited: