You are dissipating 0.5 watts, so you're okay. 55 volts across the CCS x 9.1mA. Don't increase dissipation any more. Make sure air is relatively cool and flowing freely around the MOSFETs.
As SY said, to reduce gain change tubes. 6CG7/6FQ7 is 9-pin equivalent to 6SN7, mu of 20.5, cuts your gain in half. With LED bias and a CCS your stage gain will equal mu.
The 6CG7 likes 150V and 7mA, which can be achieved with 3.4V on the cathode; two red LEDs or four SiC Schottkies.
When you increase your 200R to 220, you'll possibly come pretty close to 7mA, but your B+ will rise as your plate voltage falls. Dissipation will be around 0.6 watts, so get those TO220 cases in the loop.
You're doing well!
Stuart
Can I use an LED across the CCS DN2540 to set the bias instead of the resistor or would it just blow due to the 120V?
From the source to the gate?
Or is the LED not seeing the HV, just setting the constant current?
If you're talking about the current-setting resistor in the CCS, use a resistor, not a LED. If you're talking about the cathode resistor that sets bias for the voltage amp, use an LED (chosen to get the right bias voltage).
Looks right to me. 18mA is pushing things a bit for many LEDs- check the max current specs for the ones you're planning to use. 4V is about two green in series.
Just a suggestion: why not try to use CCS on PNP bipolar transistors when dealing with voltages below 200V?
The parameters for Bi CCS are better then that of FET CSS, especially compliane range.
The parameters for Bi CCS are better then that of FET CSS, especially compliane range.
Remember, thermal management is as important as electrical parameters.
I see that 2N6520 is T0-92 package.
now if you sketch the schematics and calculate power dissipation levels, you will be in the range of several hundred milliwatts. I prefer cases where you can use heatsinks easily.
Did you see the performance data for the DN2540 cascode CCS that Walt Jung published?
yes I did. very impressive. I do not say that FET CCS is bad. What I believe is that them to shine you neet to apply quite a voltage across. With Pimm CCS I needed at least 35-40V for the unit to work as CCS. Similar with Cascoded IXCS.
With cascoded bipolar CCS you are down to several volts (1.8-2V in an extreme).
That's true, but the simple cascode DN2540 only need 10V or so. In the case of a plate load resistor, 4 or 5 volts one way or the other is pretty negligible.
My favorite thermal management trick is a series resistor.
My favorite thermal management trick is a series resistor.
This is what I did as of last week for my DAC. I did not need huge voltage swing and I had limited supply voltage.
It works fine but I have not measured the section yet. That is the project for next week. 🙂
It works fine but I have not measured the section yet. That is the project for next week. 🙂
Cascode CCS with depletion mode MOSFETs vs. C4S with PNPs was no contest in a modified Bottlehead Seduction. The MOSFET current source sounded better. YMMV.
Brit01: as SY said, use a resistor to set the MOSFET CCS current. These things vary a bit, and to set the current you want/need you'll need to tweak a bit. To measure the current through the tube with a SS bias device (or grounded cathode and battery bias) simply measure the voltage across your resistor in the CCS. You must measure the resistance of the resistor itself of course, when you're using a pot. Ohm's law.
Use a 500R pot, see if you can adjust to the current you want with that. Otherwise, you may have to play around a bit to get the value you want. I'd use a good multi-turn pot and leave it in. When you change tubes, you can adjust the pot to experiment with operating points a bit.
Hope this helps.
Stuart
Brit01: as SY said, use a resistor to set the MOSFET CCS current. These things vary a bit, and to set the current you want/need you'll need to tweak a bit. To measure the current through the tube with a SS bias device (or grounded cathode and battery bias) simply measure the voltage across your resistor in the CCS. You must measure the resistance of the resistor itself of course, when you're using a pot. Ohm's law.
Use a 500R pot, see if you can adjust to the current you want with that. Otherwise, you may have to play around a bit to get the value you want. I'd use a good multi-turn pot and leave it in. When you change tubes, you can adjust the pot to experiment with operating points a bit.
Hope this helps.
Stuart
post 76
Wavebourn,
can you pls educate me showing an worked out example how to get R/C/Zener values including critical component's intrinsic (e.g.capacitance) requirements?
Cheers.
Wavebourn,
can you pls educate me showing an worked out example how to get R/C/Zener values including critical component's intrinsic (e.g.capacitance) requirements?
Cheers.
- Status
- Not open for further replies.