Advice needed on dealing with startup output offset - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State
Home Forums Rules Articles diyAudio Store Gallery Wiki Blogs Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Solid State Talk all about solid state amplification.

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 7th April 2016, 10:28 PM   #1
diyAudio Member
 
Join Date: Dec 2014
Location: Amstelveen
Default Advice needed on dealing with startup output offset

Hello. im a newcomer to this site and i'm hoping to get some advice

As a first audio orientated project i'm working on a hybrid Headphone amp. It is mostly pretty simple. A 6N16B-V tube in Class-A mode operating at a voltage of aprox 85v is to amplify the input signal of which the output is then decoupled to a solid-state stage consisting of a OPA2134 and TPA6120a2 in a composite setup.

My problem is that early tests and simulations show that during startup the decoupling capacitor between the tube and solid-state stage will get a bit of a inrush resulting in a sizable offset (~2v for 3 seconds) to the solid-stage and thus also to the final output. I need advice on how to best deal with this as the last thing i want is to send 2v dc into headphones when turning the device on.

Right now i'm thinking of either using a additional buffer between the tube and composite with it's own output decoupling capacitor (having that deal with initial offset rather then output-stage). Another one would be Output capacitor at the end to protect the headphone (tough that would cost a bit of space), but yeah input from experienced people would be very much appreciated.


Speaking about Output capacitors. If i were to use that as a solution and my solid-state stage is dual-rail powered with a nominal DC-offset of nearly 0v (give or take a few mV). would i need to use a bi-polar? or would a well-rated polar cap work fine aswell?
  Reply With Quote
Old 7th April 2016, 10:58 PM   #2
rayma is offline rayma  United States
diyAudio Member
 
Join Date: Apr 2011
Location: Upper midwest
Quote:
Originally Posted by Nighter3D View Post
during startup the decoupling capacitor between the tube and solid-state stage
will get a bit of a inrush resulting in a sizable offset (~2v for 3 seconds)
Can you post the schematic? You can certainly reduce the tube circuit's output capacitor
value to reduce the pulse. It's likely to be sized too large anyway, if the pulse is that long.
  Reply With Quote
Old 7th April 2016, 11:43 PM   #3
diyAudio Member
 
Join Date: Dec 2014
Location: Amstelveen
Quote:
Originally Posted by rayma View Post
Can you post the schematic? You can certainly reduce the tube circuit's output capacitor
value to reduce the pulse. It's likely to be sized too large anyway, if the pulse is that long.
I don't got a finalized schematic yet. but ive added my current test-bench. that is what i have right now for testing and the simulation result of the output of the decoupling capacitor. im using ltspice for simulating.

Reducing the decoupling capacitor further is something i hope to avoid as that would put the cutoff point above the 10hz and still leaves me with a initial peak of nearly a volt for the first 1-2seconds.
Attached Images
File Type: jpg New Bitmap Image (5).jpg (119.7 KB, 116 views)
File Type: jpg simulation.jpg (172.9 KB, 113 views)

Last edited by Nighter3D; 7th April 2016 at 11:58 PM. Reason: Forgot to add the simulation result
  Reply With Quote
Old 7th April 2016, 11:49 PM   #4
diyAudio Member
 
JonSnell Electronic's Avatar
 
Join Date: Dec 2012
Location: The Jurassic Coast, England. GB
Send a message via Skype™ to JonSnell Electronic
Valves settle down when warm.
__________________
Support for Flying Mole Class D, PMC DS-001 (Hypex),Valve Equipment and designs both new and old. www.jonsnell.co.uk
  Reply With Quote
Old 7th April 2016, 11:54 PM   #5
diyAudio Member
 
Join Date: Dec 2014
Location: Amstelveen
Quote:
Originally Posted by JonSnell Electronic View Post
Valves settle down when warm.
I know that and the circuit settles after a short while, but i need advice in keeping everything from sending a potentially headphone killing pulse (assuming that a 2 second lasting pulse of 1v can kill) into my headphones when turned on and everything is slowly settling. :<

i want to improve the safety of my amp and need advice on what ways to minimize it.
  Reply With Quote
Old 8th April 2016, 03:10 AM   #6
diyAudio Member
 
Join Date: Jun 2014
Location: New Hamburg, Ontario
I've built a couple tube hybrid amplifiers. They had a microcontroller protection system that would start the tube heater circuit first, thirty seconds later it would start the main supply, then about 15 seconds later,it would start monitoring for DC offset on the output. If all was good it would engage the speaker relay. This could easily be connected to a headphone amp.
  Reply With Quote
Old 8th April 2016, 04:03 AM   #7
rayma is offline rayma  United States
diyAudio Member
 
Join Date: Apr 2011
Location: Upper midwest
Quote:
Originally Posted by Nighter3D View Post
Reducing the decoupling capacitor further is something i hope to avoid
as that would put the cutoff point above the 10hz.
As a check, delete the U1 circuit and just connect R4 to ground.
Do you still get the same transient after C3? If so, is the duration the same?
If so, then a simple fix is to use a NC relay to short the output of C3 to ground
until the warmup time is over, with the relay being energized after that.
For turn-off, the timing capacitor would have to be shorted to ground via a diode
that is connected to a positive DC voltage that collapses quickly, like the filament's
DC supply input capacitor. Of course, the main output could be shorted to ground instead,
if there is a resistor in series with the output.

Last edited by rayma; 8th April 2016 at 04:32 AM.
  Reply With Quote
Old 8th April 2016, 09:22 AM   #8
diyAudio Member
 
vzaichenko's Avatar
 
Join Date: Jan 2010
Location: Moscow, Russia
Send a message via Skype™ to vzaichenko
Yes, as jwilhelm just mentioned - I would never use a DC-coupled circuit with my speakers or headphones without a proper power-on delay and DC offset protection. It will simply connect your headphones when the circuit is settled and disconnect them immediately in case of trouble.
__________________
If you can't explain it simply, you don't understand it well enough (c) Albert Einstein
See my projects at VIRTUAL ZERO distortion AUDIO
  Reply With Quote
Old 8th April 2016, 09:39 AM   #9
Mooly is offline Mooly  United Kingdom
diyAudio Moderator
 
Mooly's Avatar
 
Join Date: Sep 2007
Simulation is one thing, reality can often be quite different. The circuit in post #3 will kill the opamp at switch on because R1 and C3 couple the opamp directly to the 140 volt rail. That will put a high voltage transient on the opamp input... zap ! It could well do the same at power off too as the opamp input pin gets pulled to far negative.

You need two small fast signal clamp diodes on the opamp - input to clamp the voltage at that point to no higher than V3 + 0.6v and V4 - 0.6v.

Getting a truly silent switch on/off can only be done with a series switch element (relay or solid state) to the headphones. You could use a relay to short the output to ground for a few seconds but that would only work if you were prepared to add a few ohms of series resistance to the output so that the feedback network around the opamp was still functional.
__________________
-------------------------------------------------------
Installing and using LTspice IV. From beginner to advanced.
  Reply With Quote
Old 8th April 2016, 12:23 PM   #10
diyAudio Member
 
Join Date: Dec 2014
Location: Amstelveen
Quote:
Originally Posted by rayma View Post
As a check, delete the U1 circuit and just connect R4 to ground.
Do you still get the same transient after C3? If so, is the duration the same?
If so, then a simple fix is to use a NC relay to short the output of C3 to ground
until the warmup time is over, with the relay being energized after that.
For turn-off, the timing capacitor would have to be shorted to ground via a diode
that is connected to a positive DC voltage that collapses quickly, like the filament's
DC supply input capacitor. Of course, the main output could be shorted to ground instead,
if there is a resistor in series with the output.
Everything stays the same. The reason it lasts so long is cause the inrush is originating from the RC-chain filtering the supply to the tube. that one is taking time to fill up and the capacitor seems to be following that curve (a slow start-up can still be seen as a signal i suppose! ^^
A relay on the capacitor sounds like a simple idea i could easilly integrate. Those solid-state relays aren't really big nor expensive nowadays. would just need to figure out how to keep it conducting for about half a minute. would need a very small footprint timer...or use a timer capacitor as you suggest (in which case could you point me to a resource on how to determine that?)

Quote:
Originally Posted by jwilhelm View Post
I've built a couple tube hybrid amplifiers. They had a microcontroller protection system that would start the tube heater circuit first, thirty seconds later it would start the main supply, then about 15 seconds later,it would start monitoring for DC offset on the output. If all was good it would engage the speaker relay. This could easily be connected to a headphone amp.
Hmmm. do you have any examples of this? I don't know how i would have a microcontroller measure the DC-offset. Probably not something i want to do right now, but definitly worth learning about for other amps!

Quote:
Originally Posted by Mooly View Post
Simulation is one thing, reality can often be quite different. The circuit in post #3 will kill the opamp at switch on because R1 and C3 couple the opamp directly to the 140 volt rail. That will put a high voltage transient on the opamp input... zap ! It could well do the same at power off too as the opamp input pin gets pulled to far negative.

You need two small fast signal clamp diodes on the opamp - input to clamp the voltage at that point to no higher than V3 + 0.6v and V4 - 0.6v.

Getting a truly silent switch on/off can only be done with a series switch element (relay or solid state) to the headphones. You could use a relay to short the output to ground for a few seconds but that would only work if you were prepared to add a few ohms of series resistance to the output so that the feedback network around the opamp was still functional.
Relays seem to be pretty popular here. a relay on the output is doable. the 6120a2 gets a bit unstable anyway if there isn't a minimal impedance. Also using diodes to protect the input from exceeding the common-mode. didn't think of that.

Ok so i got to look into the use of Relays, Diodes to protect the input from exceeding common-mode limitations and the possible use of a microcontroller to monitor DC-offset. Right now im leaning for Rayma's solution of shorting the capacitor output to reference during startup. this would solve the problem of keeping the start up transient from reaching the solid-state stage and seems to be a solution that wouldn't take up a lot of boardspace (im kind of trying to fit everything within the Cadsoft eagle freeware 10*8cm limitation).

Nobody answered my question tough about the if i were to opt to try to protect the headphone through a sizable output capacitor (still considering even if it's "just in case"). would a bi-polar be a must or can a polar be used since the dc-bias would be zero??

Last edited by Nighter3D; 8th April 2016 at 12:47 PM. Reason: forgot to add my current preference
  Reply With Quote

Reply


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
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
0V output with a 7809 at startup 00940 Power Supplies 22 15th February 2016 05:51 PM
Advice needed: Output stage in Sony TA-E9000ES (CXA8042AS DAC) Groundloops Digital Source 2 22nd June 2011 04:20 AM
Dealing with an 8" driver offset AllenB Multi-Way 8 19th August 2010 05:26 PM
Offset nulling - kind request of advice beppe61 Solid State 13 9th May 2007 12:47 PM
protecting output tubes at startup drembo Tubes / Valves 10 14th November 2003 06:24 PM


New To Site? Need Help?

All times are GMT. The time now is 10:56 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
Copyright 1999-2016 diyAudio

Content Relevant URLs by vBSEO 3.3.2
Wiki