Tone control oscillating - UK Vintage Radio Repair and Restoration Discussion Forum

Wendymott · 16th Jan 2020, 10:49 pm

Hi Peeps..... I am pulling my hair out now... My Eddy EC10 rebuild is nearing completion... however I am stuck on one problem that seems to be causing me major grief.
The audio stage was supposed to be a simple job, as in... pre amp, Tone filter and LM386 output stage.
When I was "playing" initially, I used the stage I used in my 3 band transceiver, however the tone controls were replaced with fixed resistors, after I had decided on the audio response required..
It was just a case of copying the pcb etc.... however on test I noticed a high frequency hiss, looking at the waveform it was a 100khz ..ish .. sinewave.
I rebuilt the module fitting variable pots as in the attached schematic.. infact I copied this circuit...with the Baxendall circuit it is inevitable that there is at least on long run of pcb track..
In desperation. I relaid the pcb this pm with the shortest possible track to pin 6 ..input..and tested tonight...... again with the treble pot advanced more than 50% it squeals like a stuck farm animal....
Any ideas ?? I really dont want to relay another pcb...... is there an error with values ??.... it was a published circuit.

kalee20 · 16th Jan 2020, 11:04 pm

Hmm.

I'd expect the oscillation to be most provoked when treble CUT is applied, because that is when most feedback exists. Is that the case?

And... What type is the op-amp? Is it easy to swap for a different type which is pin-compatible?

As first try, I'd connect 3.3μF tantalum capacitors from each supply pin to 0V.

Radio Wrangler · 16th Jan 2020, 11:22 pm

And then try adding a small capacitor to the opamp from output to inverting input. It may be suffering from high frequency lag due to unavoidable strays. 30 or 100pF to start with.

David

Wendymott · 17th Jan 2020, 11:34 am

Thanks David and Kalee.... The Op amp is a TL072...... the first half is a buffer amp.. not shown.
Added full O/p stage

G6Tanuki · 17th Jan 2020, 2:22 pm

Op-amps can be fickle beasts - some of them can present a negative resistance at the output under certain conditions. When feeding some load-values this can cause strangeness (I remember an op-amp circuit which was particularly uncouth about feeding certain lengths of screened cable)

What are you loading the output with during your tests? Try a 1K resistor *in series* with the output, feeding into whatever your load is, so the output of the op-amp never gets to 'see' a resistive load of less than 1K.

Then take your NFB signal from the downstream side of this 1K resistor.

I also don't see any signs of supply-rail decoupling.

Wendymott · 17th Jan 2020, 3:49 pm

Hi Tanuki.. Lots of supply decoupling..R82/C79..R78 / C89 and C80/81 as the bias tap for the opamp. But I will try a decoupling resistor also from pin 7... Thanks

cmjones01 · 17th Jan 2020, 4:04 pm

What are C90 and C91 for? They look like 1nF and 330nF capacitors straight from the op-amp's output to ground. That will certainly send it unstable. If I've read the diagram correctly, remove them and see if that helps.

Chris

kalee20 · 17th Jan 2020, 4:16 pm

Just seen the full circuit... Completely agree Chris!

Capacitive load on an op-amp is bad news.

Wendymott · 17th Jan 2020, 5:10 pm

Hi Chris, David and Kalee...Those caps were to stop RF, and limit the HF noise.... which they did... however I see your point. C92 330 nf was not fitted.. my error on the schematic..the C91 1 nf is fitted.
However..full marks to David..adding a 560pf across the pins 6 / 7 cured the problem.. 100pf was ok too as was 1 nf, so I opted for a half way solution.
Tanuki, the 1K in series with the pin 7, partially cured the problem but not enough to justify it, as now C88 is fitted. As the schematic is now a little busy, I placed C88 in the baxendall control area, although its directly on pins 6/7
The other thing I noticed was self oscillation of the LM386 at certain levels, this was only on the lower half of the sinewave, the Zobel network should look after this... I increased C93 to 220nf. I also found that if C93 was 100nf ceramic SMD, it still "squawked" bit ......a MKT type was ok... however as I am not relaying the pcb to accommodate an MKT, I opted for a 220nf SMD ceramic. As I am not bothered about the full audio spectrum.. its ok.
Thanks peeps for your inputs.

16th Jan 2020, 10:49 pm	#1
Wendymott Octode Join Date: Jun 2013 Location: Huddersfield, West Yorkshire, UK. Posts: 1,795	Tone control oscillating Hi Peeps..... I am pulling my hair out now... My Eddy EC10 rebuild is nearing completion... however I am stuck on one problem that seems to be causing me major grief. The audio stage was supposed to be a simple job, as in... pre amp, Tone filter and LM386 output stage. When I was "playing" initially, I used the stage I used in my 3 band transceiver, however the tone controls were replaced with fixed resistors, after I had decided on the audio response required.. It was just a case of copying the pcb etc.... however on test I noticed a high frequency hiss, looking at the waveform it was a 100khz ..ish .. sinewave. I rebuilt the module fitting variable pots as in the attached schematic.. infact I copied this circuit...with the Baxendall circuit it is inevitable that there is at least on long run of pcb track.. In desperation. I relaid the pcb this pm with the shortest possible track to pin 6 ..input..and tested tonight...... again with the treble pot advanced more than 50% it squeals like a stuck farm animal.... Any ideas ?? I really dont want to relay another pcb...... is there an error with values ??.... it was a published circuit. Attached Thumbnails __________________ Should get out more. Regards Wendy G8BZY

16th Jan 2020, 11:04 pm	#2
kalee20 Dekatron Join Date: Feb 2007 Location: Lynton, N. Devon, UK. Posts: 7,081	Re: Tone control oscillating Hmm. I'd expect the oscillation to be most provoked when treble CUT is applied, because that is when most feedback exists. Is that the case? And... What type is the op-amp? Is it easy to swap for a different type which is pin-compatible? As first try, I'd connect 3.3μF tantalum capacitors from each supply pin to 0V.

16th Jan 2020, 11:22 pm	#3
Radio Wrangler Moderator Join Date: Mar 2012 Location: Fife, Scotland, UK. Posts: 22,874	Re: Tone control oscillating And then try adding a small capacitor to the opamp from output to inverting input. It may be suffering from high frequency lag due to unavoidable strays. 30 or 100pF to start with. David __________________ Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done

17th Jan 2020, 11:34 am	#4
Wendymott Octode Join Date: Jun 2013 Location: Huddersfield, West Yorkshire, UK. Posts: 1,795	Re: Tone control oscillating Thanks David and Kalee.... The Op amp is a TL072...... the first half is a buffer amp.. not shown. Added full O/p stage Attached Thumbnails __________________ Should get out more. Regards Wendy G8BZY Last edited by Wendymott; 17th Jan 2020 at 11:39 am. Reason: Add schematic

17th Jan 2020, 2:22 pm	#5
G6Tanuki Dekatron Join Date: Apr 2012 Location: Wiltshire, UK. Posts: 13,998	Re: Tone control oscillating Op-amps can be fickle beasts - some of them can present a negative resistance at the output under certain conditions. When feeding some load-values this can cause strangeness (I remember an op-amp circuit which was particularly uncouth about feeding certain lengths of screened cable) What are you loading the output with during your tests? Try a 1K resistor in series with the output, feeding into whatever your load is, so the output of the op-amp never gets to 'see' a resistive load of less than 1K. Then take your NFB signal from the downstream side of this 1K resistor. I also don't see any signs of supply-rail decoupling.

17th Jan 2020, 3:49 pm	#6
Wendymott Octode Join Date: Jun 2013 Location: Huddersfield, West Yorkshire, UK. Posts: 1,795	Re: Tone control oscillating Hi Tanuki.. Lots of supply decoupling..R82/C79..R78 / C89 and C80/81 as the bias tap for the opamp. But I will try a decoupling resistor also from pin 7... Thanks __________________ Should get out more. Regards Wendy G8BZY

17th Jan 2020, 4:04 pm	#7
cmjones01 Nonode Join Date: Oct 2008 Location: Warsaw, Poland and Cambridge, UK Posts: 2,677	Re: Tone control oscillating What are C90 and C91 for? They look like 1nF and 330nF capacitors straight from the op-amp's output to ground. That will certainly send it unstable. If I've read the diagram correctly, remove them and see if that helps. Chris __________________ What's going on in the workshop? http://martin-jones.com/

17th Jan 2020, 4:16 pm	#8
kalee20 Dekatron Join Date: Feb 2007 Location: Lynton, N. Devon, UK. Posts: 7,081	Re: Tone control oscillating Just seen the full circuit... Completely agree Chris! Capacitive load on an op-amp is bad news.

17th Jan 2020, 5:10 pm	#9
Wendymott Octode Join Date: Jun 2013 Location: Huddersfield, West Yorkshire, UK. Posts: 1,795	Re: Tone control oscillating Hi Chris, David and Kalee...Those caps were to stop RF, and limit the HF noise.... which they did... however I see your point. C92 330 nf was not fitted.. my error on the schematic..the C91 1 nf is fitted. However..full marks to David..adding a 560pf across the pins 6 / 7 cured the problem.. 100pf was ok too as was 1 nf, so I opted for a half way solution. Tanuki, the 1K in series with the pin 7, partially cured the problem but not enough to justify it, as now C88 is fitted. As the schematic is now a little busy, I placed C88 in the baxendall control area, although its directly on pins 6/7 The other thing I noticed was self oscillation of the LM386 at certain levels, this was only on the lower half of the sinewave, the Zobel network should look after this... I increased C93 to 220nf. I also found that if C93 was 100nf ceramic SMD, it still "squawked" bit ......a MKT type was ok... however as I am not relaying the pcb to accommodate an MKT, I opted for a 220nf SMD ceramic. As I am not bothered about the full audio spectrum.. its ok. Thanks peeps for your inputs. Attached Thumbnails __________________ Should get out more. Regards Wendy G8BZY