im still not convinced with these voltages.
everything works well no problem, but when you turn it right up its wandering in and out anything between -6v to +4v, and im not sure i would like to put that through a set of loudspeakers
everything works well no problem, but when you turn it right up its wandering in and out anything between -6v to +4v, and im not sure i would like to put that through a set of loudspeakers
MMM ive just realised that is without a load, and with a load at half volume, just very quickly, the centre doesnt realy move and with the scope, the sine looks the same realy
why on this model does it wander, yet when i do the same with a 3130 for example the centre doesnt move at all regardless
why on this model does it wander, yet when i do the same with a 3130 for example the centre doesnt move at all regardless
What I would do is look with the scope at the speaker output with no load and see if it goes crazy because you are approaching clipping. That will be obvious on the scope, if the signal level reaches close to the rail voltages then it will level off at that point and that can cause many amps to do weird things.
At zero volume you should have a clean trace with no signal of any kind present. The basic signal with the scope on 20ms/div is F=1/T which is 50Hz (mains hum).
The 'furriness' could possibly be high frequency oscillation but not conclusive at this point.
Was the half volume and full volume done with no audio signal present?
Have you got the scope ground probe connected to the speaker negative terminal?
Do you see a clean trace if you leave it as it is now and just connect the probe tip to the where the probe ground is clipped?
What is the amplitude of that signal and what is the scope volts per division vertically? and can you hear any hum from the speaker with the volume on minimum, then half and then full volume (no signal of course)
The 'furriness' could possibly be high frequency oscillation but not conclusive at this point.
Was the half volume and full volume done with no audio signal present?
Have you got the scope ground probe connected to the speaker negative terminal?
Do you see a clean trace if you leave it as it is now and just connect the probe tip to the where the probe ground is clipped?
What is the amplitude of that signal and what is the scope volts per division vertically? and can you hear any hum from the speaker with the volume on minimum, then half and then full volume (no signal of course)
a tiny bit of hum at zero volume but you have to put your ear to the speaker to hear it
hum and his at full volume
all volumes were taken with no signal
yes ground to speaker terminal
'Do you see a clean trace if you leave it as it is now and just connect the probe tip to the where the probe ground is clipped?'--not sure what you mean with this
not sure about the amplitude
this was 1 volt per div
next 3 attachements as suggested standard set up 100mv per div
hum and his at full volume
all volumes were taken with no signal
yes ground to speaker terminal
'Do you see a clean trace if you leave it as it is now and just connect the probe tip to the where the probe ground is clipped?'--not sure what you mean with this
not sure about the amplitude
this was 1 volt per div
next 3 attachements as suggested standard set up 100mv per div
Attachments
A little bit of hum is going to be normal on those amps. For testing hum levels at higher volume settings then shorting the inputs will give a lower overall value.
Are we saying all good hum wise then?
The DC point should not alter under no signal conditions at any point of the volume control. If it only alters when playing music then look at the output on the scope with music playing and see if it only happens as the clipping point is reached.
If you had 1 v/div for those images then you would have 2 volts peak to peak of hum and that would be incredibly loud. Even a few millivolts of signal is very audible in a speaker.
Are we saying all good hum wise then?
The DC point should not alter under no signal conditions at any point of the volume control. If it only alters when playing music then look at the output on the scope with music playing and see if it only happens as the clipping point is reached.
If you had 1 v/div for those images then you would have 2 volts peak to peak of hum and that would be incredibly loud. Even a few millivolts of signal is very audible in a speaker.
Are you sure that’s 100mV per division? That level of hum (and hiss for that matter) would be driving you out of the room. You wouldn’t need to put your ear up to the speaker to hear it. Does your scope switch 10X probes automatically or do you have to hit a switch to tell it which you’re using? I‘ve gotten it wrong before.
thats what it said on the horizontal adjustment, presuming thats what it means
the vertical is the timing right?
the vertical is the timing right?
Vertical is the amplitude, horizontal is time. The trace is drawing out the voltage in real time as it changes.
So 20ms/div means the spot tracing the signal out moves 1 square to the right every 20 milliseconds.
All that furriness on the trace could be high frequency noise. If you expand the timebase out (so make the time per division less) then you might see what that noise is. It could be a rough sort of sinewave riding on top of the other signal at perhaps 1Mhz in frequency or more..
If it were 1Mhz then you would need to alter the timebase to 1us (1 microsecond) per division to see one cycle in one horizontal division. Oscillation could be higher still in frequency which would be an even faster sweep speed.
Try this simulation to see how it works. This shows a 50Hz sine of 1 volt along with a 1Mhz sine of 0.1v I've set it up so it looks like your scope and one cycle takes 20ms of time. So the frequency is 50 Hz.
The trace is furry because of the 1Mhz present. 1st image.
In the sim you can put the cursor over a bit of the trace and highlight just that bit. 2nd image and 3rd image which highlights an even smaller part.
The sim takes a minute or to run depending how fast your PC is. Tip... don't click on Vout until the sim has completed as that will run it far quicker. Look at the lower left of the window to see when it has 100% completed.
We can now alter the sim .tran to 10us and re run the sim. 4th image. This time its quick and we see the 1Mhz clearly. The 50Hz is still there but you don't notice it at that speed.
So that is how you might use your scope here, to investigate that furriness and see what it really is.
So 20ms/div means the spot tracing the signal out moves 1 square to the right every 20 milliseconds.
All that furriness on the trace could be high frequency noise. If you expand the timebase out (so make the time per division less) then you might see what that noise is. It could be a rough sort of sinewave riding on top of the other signal at perhaps 1Mhz in frequency or more..
If it were 1Mhz then you would need to alter the timebase to 1us (1 microsecond) per division to see one cycle in one horizontal division. Oscillation could be higher still in frequency which would be an even faster sweep speed.
Try this simulation to see how it works. This shows a 50Hz sine of 1 volt along with a 1Mhz sine of 0.1v I've set it up so it looks like your scope and one cycle takes 20ms of time. So the frequency is 50 Hz.
The trace is furry because of the 1Mhz present. 1st image.
In the sim you can put the cursor over a bit of the trace and highlight just that bit. 2nd image and 3rd image which highlights an even smaller part.
The sim takes a minute or to run depending how fast your PC is. Tip... don't click on Vout until the sim has completed as that will run it far quicker. Look at the lower left of the window to see when it has 100% completed.
We can now alter the sim .tran to 10us and re run the sim. 4th image. This time its quick and we see the 1Mhz clearly. The 50Hz is still there but you don't notice it at that speed.
So that is how you might use your scope here, to investigate that furriness and see what it really is.
ok thanks will do, ill have a look to do that tomorrow
Im in the middle of another conundrum at the moment, ive put a thread out
I think I'd call those above caps "deceptively bad". Long ago in the 1970's, Japanese manufacturers were already into the practice of increasing the size of their components for a market where right or wrong, the visual impact of big translated to powerful, therefore better. I was curious about a big old JVC model I once came across, when I noticed huge, chassis mounted smoothing caps of 65mm dia. and about 90 tall. The label stated 15,000uF but they were long dead. Inside though, were 3 smaller diameter caps that had nominal 4,700uF markings though the total probably was at least 15,000 uF when new. 'funny what you see under the hood - often not what you expected at all and nothing like as refined as the seductive blurb and paid review columns would have you believe.
At least with a 3020, there are no pretences to refinement - it's what Mooly describes and it looks and works like something from an earlier generation - lots of warm distortion, odd constructional shortcuts and features so never quite "up there" with the best. Why still popular? Because price is king - particularly among hobbyists wanting to try their luck at fixing things.
At least with a 3020, there are no pretences to refinement - it's what Mooly describes and it looks and works like something from an earlier generation - lots of warm distortion, odd constructional shortcuts and features so never quite "up there" with the best. Why still popular? Because price is king - particularly among hobbyists wanting to try their luck at fixing things.
i certainly notice some things when i look deeper into these units Ian.Im no seasoned expert yet,maybe i never will be(not enough years left lol), but you do find some strange things.
I think out of all the NAD units i work on, the 3020(ser 20) has the most spurious things going on,some done by the factory, and some definatly not.
They are fun to work on and given thier relative simplicity in most cases, they can be 'hard work', but still provide lots of enjoyment for me.
This site and its members has been an inovation for me, and has given my enthusiasm for things in life, especialy learning a real boost 👍
Furry and non furry 🙂 A thickening of the trace is a classic giveaway of high frequency instability. For real here:
https://www.diyaudio.com/community/...-checked-to-see-its-stable-havent-you.191389/
https://www.diyaudio.com/community/...-checked-to-see-its-stable-havent-you.191389/
I think the 3020-20 model was the first and did have a number of bugs and undocumented mods before shipment from Taiwan. The standard 3020 model came later and had the necessary changes from the beginning. The funny stuff with 2SC1400 pinout suggests a sourcing problem if it came with an EBC (European standard) pinout. All 2SC part numbers are allocated as proprietary Japanese codes, so they can only come from the one original manufacturer (NEC) and have ECB pinout. CBE or EBC pinout on a Japanese type simply means its a fake but probably intentional here, for whatever reasons - perhaps to effect a design change at minimal cost. The type was dropped in later versions and changed or reverted to BC559 or 556. Perhaps it was really one of those from the beginning.
Last edited: