Maybe you can split up the 500 kOhm potentiometer in 1 resistor and smaller value potentiometer just to minimize dissipation in that potentiometer. And a smaller output range as you don't need 0 to 300V but more like 250 to 300V.
This one is intriguing but I never tried it to be honest (but heard positive comments about them). I would drop voltage with CRC to an acceptable level (to lower dissipation) and choose 2 so 1 for each channel (again to lower dissipation per LR8). Take a minimum of 12V dropout into account. 15V is a safe choice I guess so 315V at the input for 300V at the output and 150 mW dissipation of each LR8 when the load is 10 mA. Keep the input and output caps as specified in the datasheet close to the IC and treat them ESD safe. They cost less than 1 Euro/piece 🙂
Low part count and seems to do the job.
https://www.mouser.com/datasheet/2/268/20005399A-909415.pdf
Last edited:
For the LED, if running from the filament supply isn't an option (for instance, when using a switch with integrated LED+resistor that does not light on 6.3V), using a capacitive dropper off the high voltage secondary winding or straight off AC mains voltage may be more efficient then a resistive dropper off your high voltage DC supply
http://www.designercircuits.com/DesignNote1a.pdf
http://www.designercircuits.com/DesignNote1a.pdf
Hence the diode bridge and cap. It will give 8V and no flickering and a long living LED. Todays Chinese 12V switches with LED are often already too bright on 6V. These LEDs need only 2.5 mA for normal operation. The means that an extra series resistor is nearly always needed. Running B+/mains voltage or filament AC to a LED is somewhat odd as it can be done elegantly and safe.
Last edited:
Using an LED on the B+ is a nice feature to remind you of a shock hazard from the still charged caps. I run blue LEDs on 320V with a 1 meg resistance (47k bleeder). I use 2 meg when running from 600V (200k bleeder).
You're right about blue LEDs being too bright. Even with 1 meg you can land a plane with it.
You're right about blue LEDs being too bright. Even with 1 meg you can land a plane with it.
Like many I can't stand blue LEDs and notice the horror that manufacturers choose blue LEDs as a standard for almost any device. From what I have learned this is specific to many whose eyes tend not to focus correctly with intense blue light. I experience the same with blue neon light. Unfortunately the blue LEDs also have a narrow angle beaming a blue beam straight in your face. Bah!
Tube guys in general like to repeat stuff from the past and many cherish gold old techniques like dropping many volts for simple things. Therefor, the good old bleeder resistor did and does its job just fine and that in a relatively short time so no added risk. No LED will improve that process, there is even a higher chance on interruption as it is 😱 a semiconductor. Routing B+ to a front panel for a power LED (while having way lower voltages onboard) will give any manufacturer a hard time for approval of the device.
It would be OK, like you indicate, to use the bleeder resistor and a LED on the PSU (no B+ wiring to a front panel) and have a power LED on the switch fed by the filament power. A power LED normally should indicate that the device has been switched off and such directly. When servicing the device the second LED for bleeding would then have the right function without the need to show that at the front panel.
Tube guys in general like to repeat stuff from the past and many cherish gold old techniques like dropping many volts for simple things. Therefor, the good old bleeder resistor did and does its job just fine and that in a relatively short time so no added risk. No LED will improve that process, there is even a higher chance on interruption as it is 😱 a semiconductor. Routing B+ to a front panel for a power LED (while having way lower voltages onboard) will give any manufacturer a hard time for approval of the device.
It would be OK, like you indicate, to use the bleeder resistor and a LED on the PSU (no B+ wiring to a front panel) and have a power LED on the switch fed by the filament power. A power LED normally should indicate that the device has been switched off and such directly. When servicing the device the second LED for bleeding would then have the right function without the need to show that at the front panel.
Last edited: