I'm looking to start playing with some power tubes that require relatively high B+(700+ volts) and high currents(150+mA). I would also like to stick with a solid state design to keep the design simple.
Would any of the following designs work ?
- A typical transistor & LED/Zener reference CCS. The transistor will obviously be sized according to the voltage and dissipation.
- A whole bunch of IXYS 10M90 in parallel to deliver the required current.
Any input would be appreciated!
Would any of the following designs work ?
- A typical transistor & LED/Zener reference CCS. The transistor will obviously be sized according to the voltage and dissipation.
- A whole bunch of IXYS 10M90 in parallel to deliver the required current.
Any input would be appreciated!
How about a simple MOSFET CCS as plate-load, a mosfet Mu -follower...?
No need for complication, ref the gate to say, half +B coupled to gate by around 4.7M and you're done....
Just make sure the MOSFET chosen has a relatively low reverse-transfer capacitance, and use a Gate-Stopper of around 400-200 ohm in Carbon Compo, and keep the Gate connections to gate-stopper and protection zener good and short...
--Seeing you're running a fairly high current, a Plate-load resistor of only a few hundred ohms is needed--Figure dropping around 15-20V across it....
Ive had good results with this idea.....
No need for complication, ref the gate to say, half +B coupled to gate by around 4.7M and you're done....
Just make sure the MOSFET chosen has a relatively low reverse-transfer capacitance, and use a Gate-Stopper of around 400-200 ohm in Carbon Compo, and keep the Gate connections to gate-stopper and protection zener good and short...
--Seeing you're running a fairly high current, a Plate-load resistor of only a few hundred ohms is needed--Figure dropping around 15-20V across it....
Ive had good results with this idea.....
IXYS 10M45 with a cascode (assuming you have the headroom) would work as well.
I see no reason a normal LED-based CCS wouldn't work as long as the devices don't see over-voltage, over-current, or SOA overload conditions. I'd probably run some sims to ensure that none of these silicon killers are present during start-up as well as normal operation. Or at least think it through...
~Tom
I see no reason a normal LED-based CCS wouldn't work as long as the devices don't see over-voltage, over-current, or SOA overload conditions. I'd probably run some sims to ensure that none of these silicon killers are present during start-up as well as normal operation. Or at least think it through...
~Tom
Unless there is no reasonable way to do this with solid state parts, I'd like to avoid tube/hybrid CCS.
Hello DF, can you think of any with both suitable gain and current capability of 150mA?
Last edited:
The tubes I am trying to use are already large transmitter tubes, it makes no sense to use two of them when I can use a few power transistors to do the same thing...
Then paralleled IXYS IXTP01N100D will do the job. Rated 1000V. Cascoded ofcourse for best performance. If I'm not mistaken the "all" 100D cascade don't go much higher than 35mA (Think Kevin Carter posted that) But you can use the DN2540 (bottom) if you want a cascade with higher current rating. (Not sure why that is because the datasheet says the 100D can handle 100mA)
Last edited:
Something doesn't quite add up. if B+ is 750V, what is the 100th anode voltage? The 100TH will only swing down to about 200V at high current, and then only with super beefy grid drive.
If you plan to run at 750V anode voltage, which seems to make more sense for the 100TH, you'll need much higher supply voltage and a 1500V+ transistor (like the IXYS 6N170) to handle the swing, and still a couple in parallel to handle the ~70 watts CCS dissipation.
If you are planning to run the 100TH at low anode voltage after all, then I don't see any problem with just paralleling a bunch of the 01N100 parts with separate current setting resistors.
The cascode thing is nice but it does set an upper limit to the current because the bottom device Vds is limited by the top device Vgs. The bottom device current is limited at low Vds. DN2540 can source more current at low Vds than the 01N100 can, which makes it a good choice for bottom device.
Doh, I keep forgetting that CCS requires a higher B+; 750v is what I'm looking to run on the anode. I'm use to designing OPT loaded amps so all this CCS stuff is new to me.
As for the grid, the plan is to use a gainclone driving an OPT in reverse to supply the grid current.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.