Tube DIY Asylum

Do It Yourself (DIY) paradise for tube and SET project builders.

Return to Tube DIY Asylum


Message Sort: Post Order or Asylum Reverse Threaded

Phono Stage Motorboating

205.188.116.16

Posted on April 5, 2009 at 13:06:05
danlaudionut
Audiophile

Posts: 5480
Location: Schenectady
Joined: June 6, 2002




Got this phono stage up and running
but after about 30 minutes it goes
into low frequency oscillation.
I turn off the subs that do below 200Hz
and it doesn't go away so not feedback.
Turn the unit off and all is fine
for another 30 minutes or so.
Thinking a PS problem maybe upping the caps.
BTW sounds great when is is working.

DanL


 

Hide full thread outline!
    ...
Beef Up that Power supply..., posted on April 5, 2009 at 15:25:40
Interstage Tranny
Audiophile

Posts: 3063
Location: Eastern
Joined: October 4, 2006
With 5 stages being powered from the same capacitor bank, I'd say it's no surprise there is motorboating. More isolation/decoupling between stages will cure the runaway bass...interesting ckt...

 

RE: Beef Up that Power supply..., posted on April 5, 2009 at 16:13:46
danlaudionut
Audiophile

Posts: 5480
Location: Schenectady
Joined: June 6, 2002

IT

1) With cascoded DN2540s, I would think
the PSRR would be large enough to stop it.
2) Since the whole circuit is constant current,
there is no change in current then
how would motorboating occur?
I thought it was from changing current
setting up feedback oscillation that
created motorboating.

DanL


 

RE: Phono Stage Motorboating, posted on April 5, 2009 at 17:39:28
Caucasian Blackplate
Industry Professional

Posts: 8313
Location: Seattle
Joined: June 18, 2004
What is the purpose of the resistors between B+ and the CCS's? I have a 3 stage phono pre with CCS's running off one voltage source and no issues. You might also consider a large plate load resistor from plate to ground on the plate follower stages (up your ccs current accordingly)

 

RE: Phono Stage Motorboating, posted on April 5, 2009 at 17:48:17
danlaudionut
Audiophile

Posts: 5480
Location: Schenectady
Joined: June 6, 2002
CB

The resistor is a drain stopper
for reduction of oscillation.
Idea from Bas' website.
He suggests 1.5K minimum.

I just added more capacitance to the last stage
of the LCLC filter and no motorboating yet.
I have been playing it for about an hour so far.
My fingers and toes are crossed 8^D

DanL


 

decoupling, posted on April 5, 2009 at 17:56:17
Me thinks adding decoupling caps to the B+ lines powering your plates may help break the oscillations.

 

RE: decoupling, posted on April 5, 2009 at 17:58:37
danlaudionut
Audiophile

Posts: 5480
Location: Schenectady
Joined: June 6, 2002
FL

I added capacitance to my last stage of
the LCLC filter and all is well so far.
I hour and no motorboating yet.
Added plus is better bass.
I don't understand that but ...

DanL


 

Strange Thing Happening, posted on April 5, 2009 at 19:12:24
danlaudionut
Audiophile

Posts: 5480
Location: Schenectady
Joined: June 6, 2002
I added a Solen 100uFd to the final stage
of the LCLC filter and no motorboating.
Then I tried a Solen 60uFd and got motorboating.
Then I tried a CeraFine 330uFd and got motorboating.
Then back to the 100uFd and no motorboating.
Is it a small window where it works properly?
I never had this problem on any of my designs.

DanL


 

RE: Phono Stage Motorboating, posted on April 5, 2009 at 20:26:08
dave slagle
Manufacturer

Posts: 5448
Location: NYC
Joined: April 27, 2001
Can you provide a direct link?

his main page from his asylum info is forbidden.

dave

 

RE: Phono Stage Motorboating, posted on April 5, 2009 at 20:34:09
danlaudionut
Audiophile

Posts: 5480
Location: Schenectady
Joined: June 6, 2002
Dave

Link provided below.
Click on CCS for the CCS page.
From site -
"Rev 4.d. Settled for 2.4k Gate stopper
and 1.5k Drain stopper."
I assumed that that was the minimum values
for the best operation and larger values
would not be worse.

DanL


 

Power supply, posted on April 5, 2009 at 22:45:34
Michael V
Audiophile

Posts: 295
Location: Washington, DC
Joined: February 9, 2007
Dan,

I'm not sure what your power supply looks like, but I had this exact set of symptoms with a 12SL7 phono stage. Turns out, it was all due to a 0A3 VR tube that liked to oscillate after being warmed up.

Hope it's somethingn this easy...

Regards

 

RE: Strange Thing Happening, posted on April 6, 2009 at 00:56:07
The lead lengths off the sockets, combined with idle current of the tubes, the plate voltages, the grounding scheme, layout of components, phase of the moon, color of your sox, etc has made it that a 100 mfd cap cancels out that motorboating oscillations---prolly via a time-constant resonanace frequency thing.

If you are happy with the 100 mfd solution, let sleeping dogs lie (IMHO). Otherwise, I'd visit the wiring, layout, and grounding scheme.

Good luck!

 

RE: Phono Stage Motorboating, posted on April 6, 2009 at 01:55:38
Michael Samra
Dealer

Posts: 36118
Location: saginaw michigan
Joined: January 30, 2005
Dan
what is the silk RIAA LCR there for? Im not sure whats contained in that? It seems like it maybe an opto coupler. The design is pretty conventional like the marantz 7c only with hi mu octals going into a low mu octal only with added current sources.Also I see you have a transformer output..The biggest problem on this is probably a single sourced PS to feed all this circuitry like IT said.If it only stops oscillating when you put the 100uf solen in the final part of your power supply,you still have issues because an electrolytic shouldn't make that oscillate when its running correctly.It would have to change the impedance by quite a bit and I can't see an electrolytic doing that much damage in that position.
If the power supply waveform isn't pretty,neither is the sound in most cases.

 

RE: Power supply, posted on April 6, 2009 at 05:08:27
danlaudionut
Audiophile

Posts: 5480
Location: Schenectady
Joined: June 6, 2002
Michael

Hammond 157Gs (30H/596DCR) and 2-20uFd Solens
So it is 30H/40uFd/30H/40uFd (LCLC) filter

DanL


 

RE: Phono Stage Motorboating, posted on April 6, 2009 at 05:34:00
danlaudionut
Audiophile

Posts: 5480
Location: Schenectady
Joined: June 6, 2002
Michael

The Silk LCR RIAA filter is just that -
Inductance/Capacitance/Resistance filter module.
More at thread link below.

It's a hi mu octal gain stage DC coupled to a
hi mu octal cathode follower with CCS loads.
Maximum gain with a 600 ohm output impedance.
The 600 ohms needed for accurate RIAA response
from the Silk LCR filter module.
The only low mu octal is the final stage (6SN7)
because extra gain was needed for my system.
Transformer is there because I needed to invert the output
to correct the phono stage invertion of the signal.

>> I can't see an electrolytic doing that
>> much damage in that position.

Exactly why I posted "Strange Thing Happening".

DanL


 

RE: Phono Stage Motorboating, posted on April 6, 2009 at 07:48:39
dave slagle
Manufacturer

Posts: 5448
Location: NYC
Joined: April 27, 2001
Ahhh got it. I didn't realize it was there as part of the CCS. I presume you are using the CCS for ease of adjustment of the current through the CF which allows you to adjust the Z-out to 600R?

dave

 

RE: Phono Stage Motorboating, posted on April 6, 2009 at 08:00:54
danlaudionut
Audiophile

Posts: 5480
Location: Schenectady
Joined: June 6, 2002
Dave

Yup you got it.

DanL


 

RE: Phono Stage Motorboating, posted on April 6, 2009 at 08:20:16
artsybrute
Audiophile

Posts: 751
Joined: March 19, 2002



A standard LCR RIAA with 600 ohm in/out should look like this.



Triamp... Take a load off!

 

Artsy..If I built this circuit you posted,this seems like this is what I need, posted on April 6, 2009 at 18:56:47
Michael Samra
Dealer

Posts: 36118
Location: saginaw michigan
Joined: January 30, 2005
to sweep phono stages when injecting my signal generator..Would this be ok for the EQ you think?
If the power supply waveform isn't pretty,neither is the sound in most cases.

 

VR Tubes: A more effective way to LF isolate stages, posted on April 6, 2009 at 20:42:00
edit

 

VR tubes: properly implemented are generally a solution..., posted on April 6, 2009 at 21:11:46
edit

 

RE: Artsy..If I built this circuit you posted,this seems like this is what I need, posted on April 6, 2009 at 21:25:21
danlaudionut
Audiophile

Posts: 5480
Location: Schenectady
Joined: June 6, 2002
Michael

NO you would need the inverse of that filter.
You need to attenuate the lows instead of the highs.

DanL


 

RE: VR Tubes: A more effective way to LF isolate stages, posted on April 6, 2009 at 21:31:24
danlaudionut
Audiophile

Posts: 5480
Location: Schenectady
Joined: June 6, 2002
Thom

A few things -
1) I have too low of a B+ to be able to use VR tubes.
2) I have to wimpy of a transformer to waste the energy.
3) By stiffening the supply with more capacitance
the tendency to oscillate will be at a higher frequency
and therefore even more audible not less audible.
I added a 50uFd cap and the frequency of the motorboating
doubled proving this conclusion.

DanL


 

RE: Artsy..If I built this circuit you posted,this seems like this is what I need, posted on April 7, 2009 at 03:58:25
artsybrute
Audiophile

Posts: 751
Joined: March 19, 2002
I think what you are looking for is an inverse RIAA circuit. I've come across a bunch of them on the web over the last few years. Try googling "inverse RIAA". They are normally simple IC based, since they are just used for scope output, not sound.

As for building an LCR RIAA, from what I understand the difficulty is in the implementation. Getting the values precise (and measuring them) requires some delicate work. Also IIRC, the circuit is normally potted immediately because otherwise some of the values drift. I don't know why.



Triamp... Take a load off!

 

RE: Phono Stage Motorboating, posted on April 7, 2009 at 05:35:21
Stephen R
Audiophile

Posts: 1436
Joined: January 11, 2002
The way to reduce motorboating is to increase the time constant in the power supply i.e decouple the stages from one another that are involved with the EQ where you have a huge LF boost in a phono stage.

You won't increase the tc by just adding capacitance. You need to add resistance as well. For the sorts of current you are pulling, about 4k7 to 10k would be about right with about 20uF plus as the decoupler.

cheers,

Stephen

 

Totally agree.., posted on April 7, 2009 at 08:23:29
Michael V
Audiophile

Posts: 295
Location: Washington, DC
Joined: February 9, 2007
I would stay away from the later vintage of VR tubes (late 70's), however, since IME they seem to have much worse quality control.

 

More Storage Capacitance = Better Bass ! Yes !!!, posted on April 7, 2009 at 14:15:42
Interstage Tranny
Audiophile

Posts: 3063
Location: Eastern
Joined: October 4, 2006
It's simple and real, Dan. Every electronic device has impedance, capacitance, reactance, even resistive(very high resistance in your example) and inductive coupling is possible. Since all stages feed off the same B+ tap, the wiring and device reactances are "reacting" even though your preamp tubes are well regulated/constant current. Parasitic oscillations occur in the strangest places, often perplexing designers.

A tube power supply is charging and discharging the power supply cap storage bank at the speed of the power supply, which is 120 times per second with a fuul wave rectifier. Think of a car battery in winter. More storage capacity yields easier starting, more electrical oomph. In our tube gear, adding storage capacitance usually results in better bass impact, detail and control.

 

Found an inverse RIAA circuit, posted on April 11, 2009 at 17:41:39
artsybrute
Audiophile

Posts: 751
Joined: March 19, 2002
I found a complete writeup and schematic of an inverse RIAA for use with a signal generator when testing RIAA stages, but the one I found is not on the web.

It's in Morgan Jones' construction/testing book, "Building Valve Amplifiers", ISBN 978-0-7506-5695-5.

Page 326 has the finished schematic, pages 319-327 have the complete writeup including theory.

Hope this helps, and that you are still getting responses to this thread.



Triamp... Take a load off!

 

Page processed in 0.038 seconds.