Valve Shunt Regulator Design Walk-Through

[Mike H]

Yep that sounds reasonable. (Previous post)

[IslandPink]

Cheers
Is Dave a bit slow at responding to email ?
He hasn't replied to the last two I sent yet, last night's one was effectively giving him an order.
If you could have a word - just need to confirm the spec and any adjustment to the cost if we want to add a few more lams .

[Mike H]

Yes he can be slow. Depends what else is going on.

I'll drop a reminder anyway.

[IslandPink]

Spoke to Dave today . He's seen the transformer spec, but is very busy . He'll get onto it when he can , that's OK by me.

Just downloaded LTSpice so I can try & understand a few things without Nick or Andrew having to do it all .

[IslandPink]

Nick, when you're back from Bristol, can you pass me the Spice model of the supply that you put together ?
Cheers.

I found this post from Dave Cigna about how to model Zout , which looks very useful .
http://www.intactaudio.com/forum/viewtopic.php?t=587

[IslandPink]

Mike - I got a reply from Dave, and he can do what I want and give me 650mA on the HT winding . So, this one's a go-er .
Extra current is always useful

[Mike H]

Super!

[IslandPink]

Got the new mains transformer from Dave this morning .
Looking good !

[IslandPink]

Got a model which looks correct , on LTSpice now .
See attached .
Not getting much useful info from it yet. I ran a 'DC op pt' analysis , hoping to get out some voltages , but the most I can see in the little table it produces is about 53V .
What I'd like is to be able to probe the circuit and see the voltages and currents in various places, with it just 'idling' with a steady-state 300mA load on the end . I think I got the load, although it seems to call it a 'current source' for some reason, when you edit the boxes .

Any tips appreciated - let me know if anything looks wrong on the circuit .

[IslandPink]

There are some parameters that look OK, like the current through the inductors, but I notice some things are not right - for instance no current passing through the KT88 cathode resistor .

I'm also searching for info on how to identify the numbering of the nodes that are listed in the dc analysis , and where there are on the circuit .

[IslandPink]

Ah, once I've done a simulation run, I can hover the cursor over the various nodes and the data comes up. Excellent .
It would appear one big problem with this circuit at present is that any voltages I can find are all negative - the B+ line is at -50V or so - and none of the shunt valves are conducting .
How queer ...
I think I've connected up the rectifiers correctly and the voltage source is correct .. ?

[Mike H]

Put cursor over a 'wire' in schematic, node number shows in status bar at bottom, 'This is node Nxxx'

You can make your own names & labels for any with F4 key.


Can't see anything obviously wrong in that diagram but then I'm not familiar with that circuit.

[Mike H]

Ah, once I've done a simulation run, I can hover the cursor over the various nodes and the data comes up. Excellent .

You figured it out before I could post

It would appear one big problem with this circuit at present is that any voltages I can find are all negative - the B+ line is at -50V or so - and none of the shunt valves are conducting .
How queer ...

Ah!! It's the current source I1, it will force a Voltage regardless until its set current is achieved.

Ctrl + right mouse click on it to get the bigger attributes window, somewhere on there should be 'Make this a load', tick that check box. It will then work like a CCS.

[Mike H]

PS: personally I prefer a resistor as a load, value as appropriate Vreq.d / Ireq.d.

Have you got series DC resistances for L1 & L2?

[IslandPink]

Ah!! It's the current source I1, it will force a Voltage regardless until its set current is achieved.

Ctrl + right mouse click on it to get the bigger attributes window, somewhere on there should be 'Make this a load', tick that check box. It will then work like a CCS.

Aaargg !
So, I ticked 'Parasitic properties > This is an active load' .
Now all of the voltages are tiny, only fV .
Hmmm....

If you want to the have the circuit ( .asc ) , Mike , send me your email via a PM . Which is not to say I'm implying 'fix it for me!' at all . Just if you're interested .