FET Questions=Martin?

[martin manning]

I don't think you'll make much headway there. The FET needs about 1 mA, and roughly the same amount flows to ground through the lower leg of the divider, so maybe you could reduce the draw by half a mA? Can you reduce the first dropping resistor after the screen node, or maybe the 22k after the PI?

PS I'm interested to hear what your impressions are after adjusting the FET voltage and bias.

[ijedouglas]

Martin,

I adjusted my board to replicate the readings Charlie posted. Like Charlie, I never really got the FET sounding good but I think what I have now works really well. It is pretty transparent and my strat sounds really good whereas before it was sounding a little too brittle.

I used a decade box to dial in the voltages and ended up with a 2K2 instead of the 3K3

Hope this helps

[martin manning]

Which FET, and what do you have for voltages at either end of the drain resistor?

[ijedouglas]

NTE452

10.5V and 3.8V with 302V B+

[Charlie Wilson]

Martin, what do you think about the idea that the 150k 1/2w and possibly the lower voltages also protect the FET in case of a surge. Like every time your hit the standby switch. I just have to assume that Dumble knew that he was running that 150k almost at the limit yet he uses it in every FET I have seen. I know having that resistor open up is not great but just a thought.
CW

[martin manning]

Martin, what do you think about the idea that the 150k 1/2w and possibly the lower voltages also protect the FET in case of a surge. Like every time your hit the standby switch. :D I just have to assume that Dumble knew that he was running that 150k almost at the limit yet he uses it in every FET I have seen. I know having that resistor open up is not great but just a thought.

I don't think there's an issue with a surge coming off standby. The filters and tubes will be sinking current right away. The bigger danger would be powering up without tubes, where B+ is say 30-40V higher and there is little voltage dropped down the rail (just a couple of mA flowing for the 150k/8k2 divider and the FET). The FET won't see all of that since it will be divided down, and it should stay within the 30V Vds rating even if Vdd is set up at 20V. For sure a 1/2W 150k won't like being run without tubes installed. Resistors go down fighting, where semiconductors seem to opt for sudden death, so I don't think the idea of an under-sized resistor protecting the FET is valid. Somewhere I suggested putting a Zener across the lower resistor in the divider, which would do a much better job.

[talbany]

Martin, what do you think about the idea that the 150k 1/2w and possibly the lower voltages also protect the FET in case of a surge. Like every time your hit the standby switch. I just have to assume that Dumble knew that he was running that 150k almost at the limit yet he uses it in every FET I have seen. I know having that resistor open up is not great but just a thought.

I don't think there's an issue with a surge coming off standby. The filters and tubes will be sinking current right away. The bigger danger would be powering up without tubes, where B+ is say 30-40V higher and there is little voltage dropped down the rail (just a couple of mA flowing for the 150k/8k2 divider and the FET). The FET won't see all of that since it will be divided down, and it should stay within the 30V Vds rating even if Vdd is set up at 20V. For sure a 1/2W 150k won't like being run without tubes installed. Resistors go down fighting, where semiconductors seem to opt for sudden death, so I don't think the idea of an under-sized resistor protecting the FET is valid. Somewhere I suggested putting a Zener across the lower resistor in the divider, which would do a much better job.

I blew out a FET once coming off STBY after a week or so of use.. Added a Zener there and so far so good going on a year.

Tony

[martin manning]

I blew out a FET once coming off STBY after a week or so of use.. Added a Zener there and so far so good going on a year.

I made some measurements on my ODS. Using the standby in the normal manner (main power on first, then standby to play after 5-10 seconds, or cycling the standby switch with main power left on), it takes well over 10 seconds for the FET supply voltage (Vdd) to come up, so no risk there. If you don't use the standby (turning main power on with standby set to play) Vdd overshoots significantly, about 50% at 8 seconds, returning to normal at around 25 seconds after the heaters have come up to temperature and the tubes are drawing normal current. This is similar to the "no tubes installed" case mentioned above, but it only lasts until the heaters warm up, and the FET Vds stays within limits due to the voltage drop across the drain resistor.