advertisement


Begging for help in diagnosing fault with minimal equipment?

B

bugbear

pfm Member
As you may have guessed, my Sugden Au51p, after a rather lovely 18 months enjoying Pete Maddex's recap, is on the blink.

I noticed a bit of crackling during explosions on sci-fi films; tracking down a handy youtube bass sweep

(
)

shows that at (say) 20-30 Hz, at reasonable volume I get a lot of high frequency crackle. At very low volume things are OK.

I tried swapping over inputs, and speakers, and am certain that the fault is right channel only.

Following some email advice from Pete I have physically checked all the capacitors that he replaced for lifted tracks, and found nothing.

I have also checked with my multimeter that all tracks leading to/from the caps show continuity, and that all tracks that might have been bridged show NO continuity.

Given the symmetric power rails, and NPN/PNP transistor distribution, it seems reasonable that the amp is a traditional biased/AB. Assuming that I only have one fault, I could use a oscilloscope on the output waveform, and infer which "half" of the board is faulty. But I have no scope. (I also have no signal generator, although bizarrely, due to extreme good fortune, I do have a varaic https://www.pinkfishmedia.net/forum/threads/identify-a-variac.210736/ )

Am I wasting my time trying to diagnose with so little test equipment, or are there useful/simple techniques of which I am ignorant?

BugBear (kinda' desparate)
 
Your problem might actually be to do with the source rather than the amp.

Are you mixing down 7.1 or 5.1 sound tracks to a smaller number of channels?

PC's and even many dvd/blu ray players do this very poorly and end up clipping the DACs for the remaining channels which produces exactly the effect you describe. You even get it on broadcast satellite channels on Sky.
 
colasblue said:
Your problem might actually be to do with the source rather than the amp.

Are you mixing down 7.1 or 5.1 sound tracks to a smaller number of channels?

PC's and even many dvd/blu ray players do this very poorly and end up clipping the DACs for the remaining channels which produces exactly the effect you describe. You even get it on broadcast satellite channels on Sky.
Click to expand...
It's just stereo out of the TV (in the case of YouTube, it's an Amazon Fire Stick feeding the TV via HDMI). I have no multi channel equipment at all.

Since it's volume dependant, I'd assumed it was the amplifier; moreover swapping in (which I have, for the last 4 months) my partner's little Arcam Alpha fixes the issue.

But just to be sure, I wired the Sugden in, and played Bjork's Hyperballad from CD, which has a rather splendid bass intro.

Result - Crackle, snap, spit from the tweeter!!

BugBear
 
Looking at the layout it got a parallel output transistors I would have a look at the connections on the output transistors at the bottom of the board and the big green resistors.
It could be one half of the output cutting in and out.

Pete
 
Pete MB&D said:
Looking at the layout it got a parallel output transistors I would have a look at the connections on the output transistors at the bottom of the board and the big green resistors.
It could be one half of the output cutting in and out.

Pete
Click to expand...
Heh. Thus far I'd been hoping it was 1 on the 10 caps (or their fitting). Because if I have to extend my analysis to everything, I'm going to struggle. :)

In the meantime, I have at least discovered a way to stop messing about with YouTube.

http://www.wavecor.co.uk/testdisk.html

BugBear
 
I traced the final output components - very simple and brutal.

op_parallel by plybench, on Flickr

EDIT; the resistors leading to the bases of the all the transistors are 2.2 Ohms. DAMHIKT. :D

EDIT2; big oops. The NPN transistors are on the side of the board with the negative rail, the PNPs are on the side with the positive rail. So my schematic is wrong.

EDIT3; as you were - my notes on the power transistors were wrong, the schematic is right

BugBear
 
Last edited:
Just a thought, but could there be an inadvertent "mix" of "leaded" solder and lead free solder.
qazr.
 
Could create a bad connection, especially if there is solder residue on the PWB track and/or the component "legs".
qazr.
 
The amp is old enough to have been made with 60/40 so I can’t see it causing the problem with the amp.

Pete
 
(from another thread)
Juancho said:
Understood. As an alternative a pair of headphones and capacitor and resistor might help to track whether the noise is there at the driver stage or only in the output stage.
Click to expand...
That's interesting. Coupla' questions (from the ignorant).

What resistor/cap values would I need (I assume they're to drop voltage and block DC).

Where (in a push/pull class A/B amplifier) does the signal become assymmetric?

BugBear
 
It doesn't become asymmetrical and you can't trace the fault with headphones. It's a feedback amp (like virtually all amps) and any fault will pretty much appear everywhere in the bad channel. Can't be arsed with long explanation so just trust me on that...
 
Might be worth giving all pots and switches a good clean, if that does not work carefully use some freezer spray on small areas of the board to see if it gets better or worse, be very careful around the bias transistor as if you cool it too much the bias current could get silly very quickly.
 
I've spent the last coupla' weeks getting test kit - leads, plugs, oscillscopes, signal gens.

Turns out the lowest voltage the gen will make is 0.5V, so I desoldered (*) some TINY resistors from an old CD player to make a divider network. Fiddly.

But - wish me luck - I'm going IN!

testing by plybench, on Flickr

BugBear

(*) DIY hobbyists would have racks with odd resistors left over from projects
 
(Assistance/Advice requested)

I tried to find out what triggered the problem; the crackling/breaking up is worst at around 45Hz, when .162V (pk to pk) will cause it.

Moving on (as advised) I measured the power rails.

On the left hand (working) channel, as mentioned before, the pins of the orange-cabled-connector are:

52.7, 59.2, -59.3, -52.9

On the (mirrored) right hand connector I find:
-45.9, -59.2, 58.6, 52.9

The Sugden circuit appears to use high voltage, low current for the "input" amplification, and lower voltage, HIGH current for the output transistors. It looks like the negative output supply has a problem. -45.9 should be -52.9

Annoyingly, if I disconnect the connectors, the pins on the PSU board shows completely different voltages, even on the left hand channel. It looks as if there is some kind of connection/feedback going on.

At this point I am at a loss as to what to check next. I think I've found a clue, but I don't know how to follow it.

Advice/tips very much welcomed.

BugBear
 
Further messing with the signal gen (frequency and volume) reveals that the crackling comes and goes like a beat frequency between the signal and 50Hz.

I sort-of-suspect I'm somehow draining the reservoir capacitors of the PSU.

The waveform (on my not-really-an-osciloscope, using 10x probe) looks like this:
wave_crackle by plybench, on Flickr

BugBear
 
Have you had the psu board out and checked the soldering?
Or give it a tap when it’s running to see if you have a dry joint.

And Scope the rails to see if you have lots of ripple on the bad channel.

Pete
 
You must log in or register to reply here.


advertisement


Back
Top