• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Help with Hybrid Shuntregulator!

Status
Not open for further replies.
Jump to Latest
S

stigla

Member
Joined 2001
#1
HI!

I've been trying to tame this Shunt reg ripped from TubeCAD.

It imposes a cascode of the TL431 shunt reg (adjustable Zener) headed with a ECC99 all fed by a 20mA CCS.

But I won't get it to work, I only lose a couple of volts from 330V and I've tried everything! Swithced the TL431, tried without caps at all, different resistor values, everything! I'd love to get this to work! :magnify: '

btw, I picked the cap values myself...

Stigla

Reg1.jpg
 
#4
Hmmmm...:wchair:

I put in a load in the form of a resistor 127k resistor and fired it up.

It obviously started oscillating at a very low frequency becaouse my voltmeter acted very strange, bouncing up and down (the scale that is 🙂 )

So I increased C4 from 3uF to 10uF. It stoped oscillating but now it was back to where it started... No regulation at all.

I found that the cathode of the TL431 did NOT move up and down at when adjusting P1... guess Ill try to change it, but hav to order some new ones, all my other ones are used eslewhere...

Man, its almost embarrasing NOT to get a simple circuit like this to work... :cannotbe:
 
#5
Stigla,

Try deleting C3.

This cap is integrating the voltage sense, and holding constant voltage at the reference. You want it to move up and down as the voltage output varies, giving the TL431 the opportunity to make adjustments to keep the voltage stable.

Hope this helps.

Cheers,

Hugh
 
#6
Thanks all for suggestions!

I omitted C3 and C1 also, no effect.

I strapped a 0,1uF cap in parallel wit P1+R2 and then I got something like 140V with a 22mA current but still no adjustment possible with P1. Measured the Cathodes to be a 2,5 volts and the reference a 2,48V... Strange I HAVE connected the TL431 the right way... Tripplechecked that. Maybe I'll take it out and run the suggested test circuit from the datasheet on it.

I further shorted the gridresistor, so that the grid was connected to ground. I could then hear an oscillation from the tube, between 10k and 20k i belive... Voltage then was a 176V at 22mA...

man, this is frustrating, soon 'ill be :smash: 'ing the thing!

Well anyway, thanks for taking your time guys!
 
#7
REG.

Hi,

Stig,

C3 shouldn't be there as Hugh pointed out, most shunt regs don't like caps at their output.

C1 however should stay.

And of course you need to provide a load at the regs' output, if you tested without it chances are that the TL is toast already.

Never ever test regs with no load, same advice goes for most circuits anyway.

To paint a little smile on your face, when prototyping a 211/VT4C PP amp we had about 4 light bulbs attached to the output as a load...we were very happy when they all lit and nothing smelled suspicious.
BTW, probably the only 211 PP in the world with tube regulators for all stages bar the output.

Cheers,😉
 
#8
Wohoo!!

It works!

As it turns out, my (old and heavily used) experiment boards had some bad contacts! I was fiddling around when suddenly the voltage fell around 185V, and I'm like freezing my movements, "what wires did I touch right now" and so on... Pulling out my fingers and it fell back to a 327V... I could hear that oscillation varying when moving my hands about, I found it was the TL431. Well, I strapped a 180pF cap from the Reference pin to the cathode and a ferrite bead from P1 / R1 to the reference pin... It stopped oscillating!

Then I just played around a bit to find the culprit, namely the wire going from the plate of the triode to R2. It had a very bad contact, and after some inspection I saw that the metal clamps on the experiment board itself was in a bad shape. After swithcing to a newer board, everything works! I can now adjust the voltage so smoooothly! It's really amazing how little can affect the behavior bigtime!

The TL431 obviously survived my hefty treatment. But my CCS seemingly took the night off... I get a 39mA through the circuit and cant adjust the current and a lousy 0,9V over it... Oh well, I just had a TL783 strapped as a CCS anyway.

So now I have no C1, C3 and no R3 and C4 is 10uF.

So C1 must be in? Ok, Ill try that later today.

Now my regulator looks like this.

So thanks to everybody!

Reg2.jpg
 
#10
More important is to put the C2,R3 network back in IMO.
As you have it now, the grid is referenced to ground.
Click to expand...

Hmmm, well, it will be referenced to ground with the gridresistor too... The idea with this was that the triode should be taking care of the dynamic situations, while the TL431 should control the DC. Thus the intentional C3 in the first schematic. To couple the AC away from the TL431 reference... I guess that if I re-introduce the C2 / R3 I have to omitt C3 (in the new schematic)?

I have put C1 back and everything works great! I tried to take the 180pF out and it still didnt oscillate... Nice.
 
#11
RE:RE:SHUNT.

Hi,

Hmmm, well, it will be referenced to ground with the gridresistor too...
Click to expand...

The purpose of the cap and resistor network is to reduce ripple on the B+, the AC voltage (ripple) is injected at the grid of the triode and will reapear with inversed polarity at the anode, therefore the resultant signal is nulled out.
(CMR)

For C4, I've used values as high as 10 uF behind the reg.
You'll need to experiment a little but there is a point of no return where the reg will start to oscillate.

Cheers,😉
 
#12
I tried to re-introduce the C2,R3 network and the regulator behaved VERY strange. I got a 279V by just putting it in. Then, I clipped off the C3 in the last schematic, and, get this, I measured an insane 1038V, yep, a 1038V during the 3-4 seconds it took to hit the powerswitch! I don't know... I give up, It works now anyway, with the grid grounded...

Thanks guys!
 
#14
The pictures?

Hi Stigla,
How come you are not able to attach the circuit diagram ?
None of your attachments are on your posts.
Some of us would like to 'see' the circuit diagram as well as read all the comments.
Could you please try attaching it again . Is there any problem attaching it ? Someone could help out.
Thanks.
Cheers.
 
#15
Hi!

Sorry 'bout the pics, but its a fairy old tread, and the server holding the pics went dead some time a go, and I haven't been putting the pics anywhere else yet.

I will not be able to get the job done in some time either, so... Im sorry for that.


stigla
 
#17
Your TL431 may be oscillating (or the EL34). Try strapping a small cap, (pF) from the reference pin to the cathode of the IC. Also check all of your connections. If you use a experiment board then try to move all part to a different place on the board. Keep wires short and try first with no caps anywhere, exept at the output and possible the small cap I mentioned in the beginning. Try also to put a 1k gridstopper at the EL34 grid.

What amp are you going to use the reg on, and what what currents / voltages? EL34 might be a little overkill...?



stigla
 
#19
Zix;

I am very happy for your helpfulness, and I would very much like to post the pictures again.

But, it will sadly take some time before I will be able to send you the pictures, because I am currenlty situated pretty far from my computer at home the next mounth or so.

For those who cant wait that long, I would advise them to go to http://www.tubecad.com/july99/ witch is where I found the schematic and article.

Hope this helps 🙂
 
Status
Not open for further replies.
Top Bottom