Hi ,
I'm working on a curve tracer/matcher for power valves using the X/Y inputs of a scope , but just run into problems ! The circuit uses full wave rectified (unsmoothed) DC and at present I've been using a 4017 decade counter / potential divider / diode matrix staircase generator to provide the grid bias . Clocking is derived from the 100Hz HT via 470k resistor with zener clamp on the clock pin . Seems to work ok testing a 12AU7 down to -8v in 1 volt steps but those crappy 4017s always seem to go 'funny' on some output pins at times (I've run into this problem before with some 4017's in nixie tube clocks) . It draws curves nicely (when it works that is!) but I really need to find a more suitable solution to the staircase generator design . Anyone have any ideas or point me to schematics on the web ? I may require a design for generating up to -150V in 10V steps so if anyone has any ideas on any high voltage op-amps I'd appreciate it (logic/digital is not my strong point , only used to dealing with 5 legged devices usually :-triodes
) Thanks !
316a
I'm working on a curve tracer/matcher for power valves using the X/Y inputs of a scope , but just run into problems ! The circuit uses full wave rectified (unsmoothed) DC and at present I've been using a 4017 decade counter / potential divider / diode matrix staircase generator to provide the grid bias . Clocking is derived from the 100Hz HT via 470k resistor with zener clamp on the clock pin . Seems to work ok testing a 12AU7 down to -8v in 1 volt steps but those crappy 4017s always seem to go 'funny' on some output pins at times (I've run into this problem before with some 4017's in nixie tube clocks) . It draws curves nicely (when it works that is!) but I really need to find a more suitable solution to the staircase generator design . Anyone have any ideas or point me to schematics on the web ? I may require a design for generating up to -150V in 10V steps so if anyone has any ideas on any high voltage op-amps I'd appreciate it (logic/digital is not my strong point , only used to dealing with 5 legged devices usually :-triodes
) Thanks !316a
Sample & hold
It's not too difficult to make a staircase generator with valves:
1 Clip your sinewave to produce a squarewave
2 Differentiate
3 By clamping or rectifying, use the positive tips to cause a valve to conduct
4 Arrange the valve to be in series with a resistor such that it becomes a switched current source/sink
5 Use the current source to charge a capacitor (low leakage type)
6 Feed the capacitor charge directly into the grid of a cathode follower
A stairstep will appear on the CF's output.
You will need a reset circut also. This is easily done by sensing the stairstep level and dumping the cap charge. The cap is best to be in the cathode circuit, for ease of dumping.
OK so I'm mad
but there are less actual elements in my circuit that in a semiconductor equivalent. Also you can use "real" voltages...
It's not too difficult to make a staircase generator with valves:
1 Clip your sinewave to produce a squarewave
2 Differentiate
3 By clamping or rectifying, use the positive tips to cause a valve to conduct
4 Arrange the valve to be in series with a resistor such that it becomes a switched current source/sink
5 Use the current source to charge a capacitor (low leakage type)
6 Feed the capacitor charge directly into the grid of a cathode follower
A stairstep will appear on the CF's output.
You will need a reset circut also. This is easily done by sensing the stairstep level and dumping the cap charge. The cap is best to be in the cathode circuit, for ease of dumping.
OK so I'm mad
but there are less actual elements in my circuit that in a semiconductor equivalent. Also you can use "real" voltages...
Re: Sample & hold
John ,
This sounds ten times more complicated than what I had planned . How is this triggered ? I'm triggering at present using a simple zener diode clamp which works perfectly driving the clock pin of the 4017 counter . Got a suggested schematic ? I was expecting a reply regarding using a DAC driven by a binary counter to generate the staircase but I think I'll end up using a simple pot for testing power valves unless someone has a simple way of doing this . I don't really want to spend more than a few hours developing this idea : it's only for matching triode strapped PL509 which I can't fully test on the AVO
316a
dhaen said:It's not too difficult to make a staircase generator with valves:
1 Clip your sinewave to produce a squarewave
2 Differentiate
3 By clamping or rectifying, use the positive tips to cause a valve to conduct
4 Arrange the valve to be in series with a resistor such that it becomes a switched current source/sink
5 Use the current source to charge a capacitor (low leakage type)
6 Feed the capacitor charge directly into the grid of a cathode follower
A stairstep will appear on the CF's output.
You will need a reset circut also. This is easily done by sensing the stairstep level and dumping the cap charge. The cap is best to be in the cathode circuit, for ease of dumping.
OK so I'm mad
John ,
This sounds ten times more complicated than what I had planned . How is this triggered ? I'm triggering at present using a simple zener diode clamp which works perfectly driving the clock pin of the 4017 counter . Got a suggested schematic ? I was expecting a reply regarding using a DAC driven by a binary counter to generate the staircase but I think I'll end up using a simple pot for testing power valves unless someone has a simple way of doing this . I don't really want to spend more than a few hours developing this idea : it's only for matching triode strapped PL509 which I can't fully test on the AVO
316a
Generator
John ,
Thanks , I get the general idea . I may try an adaptation After plotting a curve for a pl509 at -80V 300V I get the feeling that a cuve tracer may take it's toll on the valves as they're high gm . I've now added a stack of 25-0-25V transformers and can switch between different anode sweep voltages . Very interesting , at 0V the curves really do go straight . taking a few traces and placing a ruler on the 'scope and I'm getting Ra of about approx 400 ohms and mu of 4 at 300V . Like a WWII radar operator I'm drawing the lines on the scope screen then testing another valve !(must add more sockets and a switch!) The good thing is that these valves seem quite consistant (Tungsrams) and at £3 a throw I'll easily be able to make up a matched set for the next amp
316a
dhaen said:
John ,
Thanks , I get the general idea . I may try an adaptation
After plotting a curve for a pl509 at -80V 300V I get the feeling that a cuve tracer may take it's toll on the valves as they're high gm . I've now added a stack of 25-0-25V transformers and can switch between different anode sweep voltages . Very interesting , at 0V the curves really do go straight . taking a few traces and placing a ruler on the 'scope and I'm getting Ra of about approx 400 ohms and mu of 4 at 300V . Like a WWII radar operator I'm drawing the lines on the scope screen then testing another valve !(must add more sockets and a switch!) The good thing is that these valves seem quite consistant (Tungsrams) and at £3 a throw I'll easily be able to make up a matched set for the next amp 316a
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.