Hi all,
I would like to use a Wilson Current Mirror as the plate load of an output PP pair. I need the average DC to be constant in both legs and the AC signal will be filtered out at the output of the Current Mirror. However i also need to set the DC bias current at the anodes as well at 60mA.
I can find plenty of information on the Wilson Current Mirror - but only setup as a current sink rather than an active load.
Can anyone advise me how you can configure a Wilson Current Mirror as an active load with Constant Current properties. For this purpose ignore the complication of the AC signal and only consider it's DC behavior.
Any help is appreciated.
Shoog
I would like to use a Wilson Current Mirror as the plate load of an output PP pair. I need the average DC to be constant in both legs and the AC signal will be filtered out at the output of the Current Mirror. However i also need to set the DC bias current at the anodes as well at 60mA.
I can find plenty of information on the Wilson Current Mirror - but only setup as a current sink rather than an active load.
Can anyone advise me how you can configure a Wilson Current Mirror as an active load with Constant Current properties. For this purpose ignore the complication of the AC signal and only consider it's DC behavior.
Any help is appreciated.
Shoog
Are you sure you want a current mirror? It looks like you're asking for a current source with two equal outputs. Or, are you looking for a current mirror like what is used in op-amps? In that case the current mirror is a textbook design (with buffering for the high voltage).
I am looking for a current source with two equal outputs. My understanding is that a Wilson might be able to do this.
I have used individual CCS (IXYS chips) to do this job and they work well - but I have to go to some trouble to get them and they have a tendency to fry under fault conditions in this application.
I want something which guarantees current balance and is a bit more robust.
Shoog
I have used individual CCS (IXYS chips) to do this job and they work well - but I have to go to some trouble to get them and they have a tendency to fry under fault conditions in this application.
I want something which guarantees current balance and is a bit more robust.
Shoog
Last edited:
If you are forcing exact current balance in a P-P pair at the anode then you had better have a means of ensuring voltage balance too. If not, one anode could end up 'grounded' or sky-high if the valves are not perfectly matched.
Bear in mind too that a current mirror is only a high impedance at the output terminal. It is not symmetric like a transformer. The input terminal is low impedance: that means that a triode output would lose triode feedback that side, while a pentode output is more likely to suffer from anode voltage problems on the other side.
Bear in mind too that a current mirror is only a high impedance at the output terminal. It is not symmetric like a transformer. The input terminal is low impedance: that means that a triode output would lose triode feedback that side, while a pentode output is more likely to suffer from anode voltage problems on the other side.
Voltages find their own level without issue - which maybe somewhat different for each side - this is not a major issue though.
The application has a split primary with each side having its own effective power supply. the application I am trying to create is to tie those anodes together for DC standing current.
The application works, as I have an amp working stably for over a month using IXYs chips to do the job.
--------
However, the more I read about it, the more I realize that it may not be possible since the Rset of the Wilson is dependent on the Vcc(B+) to ground which in this case will be a variable quantity.
It may very well be that individual floating CSS is the only way to go.
Shoog
The application has a split primary with each side having its own effective power supply. the application I am trying to create is to tie those anodes together for DC standing current.
The application works, as I have an amp working stably for over a month using IXYs chips to do the job.
--------
However, the more I read about it, the more I realize that it may not be possible since the Rset of the Wilson is dependent on the Vcc(B+) to ground which in this case will be a variable quantity.
It may very well be that individual floating CSS is the only way to go.
Shoog
Not really... It can still be done with a modified mirror. If someone else hasn't drawn something up by the time I'm done mowing, I'll draw up a quick circuit.
However, everything that DF96 pointed out needs to be considered. Just because a matching current source is possible, doesn't mean it's a good idea.
That would be the szort of thing, but I was hoping for a floating implementation.
Shoog
Yah need a complete DC path for the current loop, not sure a floating CCS is possible? Choke? But u already know about chokes...
- Status
- Not open for further replies.