Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

Help with B+ supply mod (gyrator?)
Help with B+ supply mod (gyrator?)
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 13th January 2019, 01:37 AM   #11
Breezio is offline Breezio  United States
diyAudio Member
 
Join Date: Jan 2019
Location: Berkeley, CA
Default Regulator tuning

Dear all,

The regulator has worked very fine for a while and now I see some strange things happening. It might be related to the tuning of the tubes bias which I did after putting it in, but I am not completely sure.

I am attaching the schematic and a few pics (AC) from the oscilloscope. As for the DC values here is the list:
Vin=396V
VR9-R4-D1-C4= 385V
Vd=384V
Vg=384.6V
Vout=381V

There might be something related to the RC bridges. In fact I had previously observed a similar behaviour where something like the charge and discharge of a capacitor affects the Mosfet gate and therefore the Vout. In that case I had succesfully bypassed R4.

At this point I miss some practical understanding on how to remediate. I hope we can solve this together - thanks a lot in advance.

Fabrizio
Attached Images
File Type: png Regulator v0.1.png (10.5 KB, 176 views)
File Type: jpg New Board.JPG (299.1 KB, 162 views)
File Type: jpg Vin.JPG (266.8 KB, 153 views)
File Type: jpg Vin (2).JPG (289.2 KB, 123 views)
File Type: jpg VR9-R4-D1-C4.JPG (268.8 KB, 127 views)
File Type: jpg Vout.JPG (256.5 KB, 29 views)
  Reply With Quote
Old 13th January 2019, 06:31 PM   #12
GoatGuy is offline GoatGuy  United States
diyAudio Member
 
GoatGuy's Avatar
 
Join Date: Dec 2012
Location: SF Bay Area
Quote:
Originally Posted by Breezio View Post
Dear all,

The regulator has worked very fine for a while and now I see some strange things happening. It might be related to the tuning of the tubes bias which I did after putting it in, but I am not completely sure.

Fabrizio
Yah… From the (rough) equation:
VP-P ≈ I / (C 2F) (amps / farads × 2 × Hertz)
VP-P ≈ 0.250 / (0.00088 F × 2 × 60 Hz)
VP-P ≈ 2.4 V
I see that the ripple on VIN ought to be around 2.4 volts. (The 250 mA = 0.250 A came from 12 volt drop across the 47 Ω resistor.).

Thing is, that you're somewhat ironically working "too close to the margins" for the MOSFET, I think.

FIRST, ask yourself, “is 250 mA flowing out of this regulator, what I expect to be flowing out?” If "yes", then it is one heck of a power amplifier, taking about 100 input watts at near 400 volts. Bravo!

In any case though, to get the regulator MOSFET to actually regulate, you need MORE voltage on VD. The 47 Ω resistor is getting in the way. Its good protection, I suppose, but it is dropping too much. (And yes, this very well could be the result of you changing the BIAS of the output finals.)

So, as a simple experiment, simply bypass the 47 Ω resistor. You don't need to solder a wire-short-bypass, just use a well-insulated wire with alligator clips (insulated) on the ends. Bypassing the resistor will deliver ALL of the 396 VIN volts to the MOSFET, giving it some "margin" to work with.

I predict that as soon as you do that, unless some component is broken, that the output will stabilize to nice, flat DC, at around 381 volts.

The moral of that story would be … to use a smaller (like 15 Ω) safety resistor instead of 47 Ω.

Anyway, I've attached an annotated diagram.

GoatGuy
Attached Images
File Type: jpg 190113-schematic.jpg (31.7 KB, 57 views)
__________________
John Curl's Golden Rule…: 100 kHz bandwidth, 3 μs risetime, 100 W mean output, 100 V/μs slew rate, 2 Ω dynamic load, 20 amp min current source/sink

Last edited by GoatGuy; 13th January 2019 at 06:36 PM.
  Reply With Quote
Old 14th January 2019, 12:06 AM   #13
GoatGuy is offline GoatGuy  United States
diyAudio Member
 
GoatGuy's Avatar
 
Join Date: Dec 2012
Location: SF Bay Area
Alternately… (if you want to keep the 47 Ω resistor), then choose different Zeners for the Zener chain. To be around 360 V total. Or, if you remember that the gate of the MOSFET consumes no power (at all), then you can go for a simple resistive voltage divider with the existing Zeners, to get a lower reference voltage.

You know, adding a 2 resistor divider between top of C₅ and ground, the "upper" resistor being 270 kΩ, the bottom one being 5.6 MΩ. This would give about 368 V to the gate of MOSFET, and represent only a 0.066 mA load to VIN

Just saying,
GoatGuy
__________________
John Curl's Golden Rule…: 100 kHz bandwidth, 3 μs risetime, 100 W mean output, 100 V/μs slew rate, 2 Ω dynamic load, 20 amp min current source/sink
  Reply With Quote
Old 14th January 2019, 01:12 AM   #14
Breezio is offline Breezio  United States
diyAudio Member
 
Join Date: Jan 2019
Location: Berkeley, CA
Thanks a lot GoatGuy,

I should have realized that the total current drawn is just too high. Guess I am still rather rusty... I tested the regulator with a couple resistors (8k and 15k) as a load and everything works fine. So there should be no issue on the regulator alone.

Then, with the regulator connected, I measured how much current is drawn by the two tube modules (measured on R520) and by the preamp (measured on R332). According to the schematic attached, this should account for the total, which is around 28mA!

So now I am scratching my head on where the remaining current (~220mA) is going. Let me add a couple side problems I was going to check later on and which might be related or not:

- There s a low hum on the left channel only (independent of volume level) - The Direct/Phono/Normal knob makes a lot of noise everytime I switch it

Thanks
Fabrizio

PS: I forgot to add that bypassing the 47ohm resistor does not change how the regulator behaves
Attached Files
File Type: pdf AMC CVT3030.pdf (563.4 KB, 4 views)
  Reply With Quote
Old 14th January 2019, 05:26 PM   #15
GoatGuy is offline GoatGuy  United States
diyAudio Member
 
GoatGuy's Avatar
 
Join Date: Dec 2012
Location: SF Bay Area
I hope you find the source of "too much B+ current draw". It is important that you find the problem, since if it is being drawn by the output final tubes, well … a short life for them! ( 250 ma ÷ 4 ea ) × 380 V ⇒ 23,700 mW/tube. While you definitely have strong tubes, dissipating quiescent¹ 24 watts a bulb is quite a bit.

PS: In my annotated drawing, did I get the voltage values right? Here is another copy of it, with "ABCDEFG" annotations so that we can talk about test-points without having to refer to more cumbersome designations.

GoatGuy
________________________________________

¹ quiescent → the 'resting' current, power or voltage when no signal is passing thru circuit.
Attached Images
File Type: jpg 190113-schematic-2.jpg (44.5 KB, 57 views)
__________________
John Curl's Golden Rule…: 100 kHz bandwidth, 3 μs risetime, 100 W mean output, 100 V/μs slew rate, 2 Ω dynamic load, 20 amp min current source/sink
  Reply With Quote
Old 14th January 2019, 05:30 PM   #16
Breezio is offline Breezio  United States
diyAudio Member
 
Join Date: Jan 2019
Location: Berkeley, CA
Thanks!

The tubes should be fine because they draw about 11mA each.
Ill keep you posted on further findings.

Fabrizio
  Reply With Quote
Old 17th January 2019, 06:38 AM   #17
Breezio is offline Breezio  United States
diyAudio Member
 
Join Date: Jan 2019
Location: Berkeley, CA
Dear GoatGuy,

After several measuremens and talking with the AMC support I can confirm that the total current flowing through B+ is around 260mA and that the nominal B+ voltage is 375V.

So I modified the regulator to lower the output voltage to 375V by adding the voltage divider as you suggested. Unfortunately I observe the same behaviour described previously (see pics from oscilloscope) whether keeping R3 (47ohm) or bypassing it,

Referring to the new schematic attached here are the DC voltages

A= 400V
B= 387V
C= 389V (or C= 400V)
D= 376.4V
E= 376V

Now I have enough room for the regulator to work, however the problem persists.

Any idea on what to do next?
Thanks again,
Fabrizio
Attached Images
File Type: png Regulator v0.2.png (11.3 KB, 39 views)
File Type: jpg Vin (3).JPG (260.4 KB, 30 views)
File Type: jpg Vout (2).JPG (258.3 KB, 19 views)
  Reply With Quote
Old 17th January 2019, 05:00 PM   #18
GoatGuy is offline GoatGuy  United States
diyAudio Member
 
GoatGuy's Avatar
 
Join Date: Dec 2012
Location: SF Bay Area
OK… reading ( http:\\www.vishay.com\docs\91070\sihf840.pdf (repl \ with slash) ), it isn't too hard to figure out that the "turn on" for the IFR840 is about 3.5 to 4.0 volts (for ID of 250 ma). Something like that.

But your measurements of {D} and {E} are within a few tenths of a volt of each other. This implies that either the MOSFET is damaged, and its gate is broken, or, that the D5 diode has gone bad (near-short). Because "soldering is a pain", I would replace both at the same time, myself.

Leave the (not shown) voltage divider in place. It is a good idea.

The voltage you SHOULD be reading at {D} is as you say … around 275 V. The voltage at {E} though should be 4 volts or so below {D}. If the pair of MOSFET + D5 are working.

Also in the future — I cannot tell from your oscilloscope pictures at what points {ABCDE} they are being captured. Consider opening in a simple photo editor and "marking" with the test-point position from the schematic.

Yours, GoatGuy

(one last thought ... are you sure you have the D5 "pointing the right way"? Something also to check.)
__________________
John Curl's Golden Rule…: 100 kHz bandwidth, 3 μs risetime, 100 W mean output, 100 V/μs slew rate, 2 Ω dynamic load, 20 amp min current source/sink
  Reply With Quote
Old 18th January 2019, 12:05 AM   #19
egellings is offline egellings  United States
diyAudio Member
 
Join Date: Aug 2010
Location: CA
Looks like a source follower buffered Zener supply to me.
  Reply With Quote
Old 18th January 2019, 12:57 AM   #20
GoatGuy is offline GoatGuy  United States
diyAudio Member
 
GoatGuy's Avatar
 
Join Date: Dec 2012
Location: SF Bay Area
Quote:
Originally Posted by egellings View Post
Looks like a source follower buffered Zener supply to me.
Which is exactly what it is. However, there's a lot of extraneous stuff going on. With the filtering of the Zener stack before the MOSFET gate It should nominally be a dead silent supply, if nothing else.

That it is being tasked with supplying 250+ mA at 375 V quiescent is quite a thing. It must be for both channels, so 125 mA a channel. Yet that's still over 46 watts-a-channel in humming-along dissipation. And, granted, some of that 375 ought to be going to prior stages, the VAS and compliment inversion and so on, but still … lots'o'watts.

Just saying, GoatGuy
__________________
John Curl's Golden Rule…: 100 kHz bandwidth, 3 μs risetime, 100 W mean output, 100 V/μs slew rate, 2 Ω dynamic load, 20 amp min current source/sink
  Reply With Quote

Reply


Help with B+ supply mod (gyrator?)Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Gyrator or CCS MiiB Tubes / Valves 61 20th December 2018 08:38 PM
Help with gyrator mogliaa Tubes / Valves 61 9th April 2012 06:01 PM
EQ with BJT as Gyrator wleediy Solid State 0 10th February 2008 09:00 PM


New To Site? Need Help?

All times are GMT. The time now is 08:16 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio
Wiki