So definitely the amp then. A couple of checks you can do (but you need to be super careful with the measurement) is to see what the voltage is across R19 and R21 which are the 0.1 ohm in the output stage. Measure with no signal applied to the amp. The 306 should be able to deliver 1 watt from its Class A driver stage and it can do that without the output transistors even in place.
The simulation shows 5.7 millivolts across R23 and 2.4 millivolts across R19. So these are very small values we are measuring.
Also check the voltages across R17 and R18. They should be equal and around 18 volts across each.
Is the sound distorted at any volume level, even very quiet?
I always think the best scope to begin with is an older analogue type rather than a digital one. Forget those USB things, they are far to limited. You need at least a 20Mhz bandwidth (most will be more anyway) and most are dual trace. The scope is just a voltmeter that traces voltage over time out on the screen so although they look complicated the basic function is just voltage and time measurement.
The simulation shows 5.7 millivolts across R23 and 2.4 millivolts across R19. So these are very small values we are measuring.
Also check the voltages across R17 and R18. They should be equal and around 18 volts across each.
Is the sound distorted at any volume level, even very quiet?
I always think the best scope to begin with is an older analogue type rather than a digital one. Forget those USB things, they are far to limited. You need at least a 20Mhz bandwidth (most will be more anyway) and most are dual trace. The scope is just a voltmeter that traces voltage over time out on the screen so although they look complicated the basic function is just voltage and time measurement.
R17 18v
R18 19V
R19 03.1mv
R21 00.3mv
R23 0v
I'll check to see if I can hear the distortion at low volume and get back to you...
Thanks for the tip. I did look at those USB scopes, but will avoid now.
R18 19V
R19 03.1mv
R21 00.3mv
R23 0v
I'll check to see if I can hear the distortion at low volume and get back to you...
Thanks for the tip. I did look at those USB scopes, but will avoid now.
OK, these all sound in the right ball park then. It's difficult now without a scope to see what is happening.
See whether it is OK at very low level and as before, is there anything you have altered signal wise that could be causing this. Any mods, any changes anywhere, anything really.
See whether it is OK at very low level and as before, is there anything you have altered signal wise that could be causing this. Any mods, any changes anywhere, anything really.
They sound fine at low levels but get progressively worse as the volumes turned up. When I switch it off with the music still playing, the left channel keeps producing music (although distorted) for a lot longer than the right channel.
And just to be clear, its only the left channel that's faulty?
We have to cover all bases on this (sorry for asking 🙂) but you are measuring those voltages on the faulty channel and not the good one?
The reason one channel plays for longer when you turn it off is most likely because that channel is for some reason drawing less current and so the reservoir caps can keep it going for longer.
Crude test. TR7 is the final Class A driver/output transistor. Are they both running at similar temperatures if you touch them?
We have to cover all bases on this (sorry for asking 🙂) but you are measuring those voltages on the faulty channel and not the good one?
The reason one channel plays for longer when you turn it off is most likely because that channel is for some reason drawing less current and so the reservoir caps can keep it going for longer.
Crude test. TR7 is the final Class A driver/output transistor. Are they both running at similar temperatures if you touch them?
Yes, only the left channel is faulty. And yes, I've only been measuring the left, faulty channel.
Should I check T7 with music playing? With no signal or speaker connected, they both feel cold to lukewarm.
Should I check T7 with music playing? With no signal or speaker connected, they both feel cold to lukewarm.
With no signal I think.
We might be reaching a point where DC checks alone are not going to find this.
We might be reaching a point where DC checks alone are not going to find this.
I've asked on another forum If I can borrow a scope - https://pinkfishmedia.net/forum/threads/oscilloscope-in-orbit-a-pfm-diy-resource.174164/page-2
Wow 👍 interesting.
As far as DC checks go I think we've covered all the main points.
1/ Supplies correct at -/+ 38v approx.
2/ Current flow in Class A output stage correct. That is the voltage across those 0.1 ohm resistors and the 560 ohms.
You can always compare readings with the good channel and see if anything stands out.
Distortion can be pure distortion of some kind or it can be caused by the circuit perhaps oscillating at high frequency and doing strange things. Look at any and every mod you have done and make sure everything is OK. No wrong values anywhere, that kind of thing.
This is the kind of scope that would be suitable to start out with... not that this is a recommend that this one is OK of course, just the kind of thing to look at.
https://www.ebay.co.uk/itm/334361507345?hash=item4dd97fb611:g:IywAAOSwyhpiLIzE
Analogue scopes are tough and teach you so much more than a digital one where you tend to accept what it says without questioning.
As far as DC checks go I think we've covered all the main points.
1/ Supplies correct at -/+ 38v approx.
2/ Current flow in Class A output stage correct. That is the voltage across those 0.1 ohm resistors and the 560 ohms.
You can always compare readings with the good channel and see if anything stands out.
Distortion can be pure distortion of some kind or it can be caused by the circuit perhaps oscillating at high frequency and doing strange things. Look at any and every mod you have done and make sure everything is OK. No wrong values anywhere, that kind of thing.
This is the kind of scope that would be suitable to start out with... not that this is a recommend that this one is OK of course, just the kind of thing to look at.
https://www.ebay.co.uk/itm/334361507345?hash=item4dd97fb611:g:IywAAOSwyhpiLIzE
Analogue scopes are tough and teach you so much more than a digital one where you tend to accept what it says without questioning.
I've found in the past that the large majority of the time any problems were caused by me, so It's highly likely something I've done.
In your simulation you have L3 as 1u8, in the Quad service manual it's 1u5, but both mine are now air-core and measure 1u3 (assuming my DE-5000 is correct). When I made them, I know I made sure they matched what was already there (for inductance and resistance), and it's highly unlikely they've changed because they are pulled together very tightly with cable ties and glue (measured after the cable ties were fitted). Both channels have been fine for nearly two years with these inductors, so I can't see it being a concern, but I am interested to know how much difference the inductance makes. Unfortunately, I don't know where the original ones are.
I looked at one of those Hameg scopes, and wondered if it was worth paying the extra for a unit in apparently excellent condition, calibrated, and with probes?
https://www.ebay.co.uk/itm/134052389104
Reading around, I noticed quite a few people prefer the old analogue scopes. Something about instant results?
In your simulation you have L3 as 1u8, in the Quad service manual it's 1u5, but both mine are now air-core and measure 1u3 (assuming my DE-5000 is correct). When I made them, I know I made sure they matched what was already there (for inductance and resistance), and it's highly unlikely they've changed because they are pulled together very tightly with cable ties and glue (measured after the cable ties were fitted). Both channels have been fine for nearly two years with these inductors, so I can't see it being a concern, but I am interested to know how much difference the inductance makes. Unfortunately, I don't know where the original ones are.
I looked at one of those Hameg scopes, and wondered if it was worth paying the extra for a unit in apparently excellent condition, calibrated, and with probes?
https://www.ebay.co.uk/itm/134052389104
Reading around, I noticed quite a few people prefer the old analogue scopes. Something about instant results?
L3 is one the most critical parts in the amp... in theory anyway... as it forms part of the bridge used for removing distortion from the unbiased output transistors.
If there is doubt over the coils then maybe we could try swapping L and R over. In the simulation the amp can become totally unstable if the bridge balance is altered to much but whether that translates to the real amp I wouldn't like to say.
I'm very suspicious though if you have gone to air cored from ferrite cored. The Quads (thinking of the 405) can be very sensitive to transistor types as well.
If there is doubt over the coils then maybe we could try swapping L and R over. In the simulation the amp can become totally unstable if the bridge balance is altered to much but whether that translates to the real amp I wouldn't like to say.
I'm very suspicious though if you have gone to air cored from ferrite cored. The Quads (thinking of the 405) can be very sensitive to transistor types as well.
Anything you get has to be your choice 🙂 If its something you think you will use quite a lot then its probably worth going for something decent.
Left Channel
R24 21.99R
R25 22.07R
R22 561R
Right Channel
R24 21.91R
R25 21.91R
R22 561R
I swapped L3,R24,R25 and R22 over to the other channel, and still got the same problem on the same side.
I'm running out of ideas now without being able to see what this distortion looks like on a scope.
Just a thought. When you have checked the rail voltages was that done without a speaker attached? If it was then can you measure if there is any DC voltage across the speaker terminals with no load attached. The 306 would behave differently if there was a DC offset, it wouldn't harm the speaker but it would skew the rail voltages.
Just a thought. When you have checked the rail voltages was that done without a speaker attached? If it was then can you measure if there is any DC voltage across the speaker terminals with no load attached. The 306 would behave differently if there was a DC offset, it wouldn't harm the speaker but it would skew the rail voltages.
All testing was done without anything attached to the amplifier (apart from the power cable).
With no load or anything attached -
Left channel
00.3mV
Right channel
00.8mV
Perfect. So that blows that theory out of the water then.
We really need to see in what way a signal is being mangled to get a handle on what is happening here and what this distortion might be. If there are no obvious voltage anomalies between the good and bad channel then I cant think of any other way to do it really, and we have covered all the main areas to check.
We really need to see in what way a signal is being mangled to get a handle on what is happening here and what this distortion might be. If there are no obvious voltage anomalies between the good and bad channel then I cant think of any other way to do it really, and we have covered all the main areas to check.
I'll see if I can borrow a scope, if not I'll buy one. Whatever happens, I'll report back here...