Oscilloscope

[Davey]

Connecting (+) of C77 to chassis ground made no difference, and disconnecting it while 'live' did not either, so I conclude that the solder joint on the underside is good.

[tony sayer]

Right thats very interesting.

Look to see where C89 is located. There will be a wire more likely than not a screened one coming from the balance control to that point.

Disconnect it on BOTH channels at the junction of C89 or pull C89 out and its partner on the other channel, and tell us if the hum comes or goes. If it does then put C89 back then located R100 and its counterpart on the other channel. This will then leave the balance control in circuit and if you look closely that forms a potential divider in conjunction with R100 an if thats temporary disconnected the amp is then left with its balance pot "connecting" the input of the power amp section to an earth point, this may well be significant.

It does .. thats how it works in conjunction with R100 and its counterpart R on the other channel..

;)...

>

[tony sayer]

That looks about right thanks...

[Davey]

You want me to disconnect one or both R100s from their circuits, yes?

I'll report soon what effect disconnecting both C89s has. It's easiest to pull the balance pot wire off where it joins the pin just before C89, I did one already last week (or the week before).

[Davey]

Hmmmmmmmm......

[Davey]

Silence is golden! Power surge noise as power comes on, then nothing. Still disconnected at the moment, but can be re-done when required.

Both R100s are on the bastard board which suffered the busted R60 the other time I worked on it, so I want to be absolutely sure what you want me to do there before I loosen that board again. Please elaborate.

I still need help finding the right replacement for R60 on the RS website, too, please.

[tony sayer]

In article , Davey scribeth thus

Yes I'm just working back from the output so see where the humm is introduced. If you can't remove R100 then there should be a lead from that to the balance control and then that in turn goes of to C89 if you look at the diagram. What I want to do now is to "connect" the input of C89 to earth via the balance control to see if that starts the hum up which it should not.

Do you understand where I'm coming from?..

If you can't do that then try pulling one end of C87 from the board that will do a similar thing but not quite the same but see if you can do the other first..

And as above do that on BOTH channels...

680K ohm

[Davey]

I'll look at the diagram to see what you mean, should be ok. See below.

Ok, I'll see what I can do. Removing either R100 is possible, but so is balancing on one hand, just not that easy......Incidentally, the two R100s are not physically identical, same as the two C93s weren't.

I see what you want to try. I'll see what is possible without major upheaval, or further damage.

Ah, ok. I had already seen that, but it doesn't match the physical design of the dead one, which is more of a cylinder with end-caps and feet, and I didn't know what it's type description was. "Not Wire-wound" was all I knew, from an earlier post. I have a selection of unused resistors from old projects, and of course, that one isn't amongst them. Thanks.

[Davey]

Gotcha! I first reconnected the two flying leads at the C89s, and then started to look at the R100/C87 area. C87s are both EIRE caps, and I already have replacements in my RS Basket, but not yet ordered. I want one more final purchase, and I am already arguing with them over the stealth £4.95 charge for Delivery for private customers, not mentioned on the Shipping Charges page, and their sending of two invoices: One without Delivery, but with VAT; the other with VAT, but not including Delivery, so neither showed the actual sum charged!

But back to the board. I identified the two C87 caps, and noted that they would eventually be coming out anyway when they were to be replaced with new ones. I looked at the three pins on the board, one each for the two signal wires over to the Power boards and C89s, the middle one being where these shielded cables' shields were connected together, and to earth; except they weren't connected to the earth pin. Close, very close, but not connected. Fixing that was easy, and I have now heard the cassette player playing without a hum-filled background! So your trouble-shooting process led to the exact spot. It is easy to say that I should have spotted this before, but I would have had to know exactly how the circuit worked while viewing this part of it, and it looked fine, there was nothing to shout out "Missing Connection". I'll see if I can find it on an early photo. Yes,

It shows the yellow shrink-wrap and the pin to which the wires inside it should be connected (I also have a photo taken from above which makes it look as though it is perfectly ok). To each side, you can see the different R100s, and behind, the two C87 EIRE caps. to be replaced. Behind the left-hand R100 is a resistor R94 (?) which physically matches the broken R60.

So, it looks as though the ultimate cause of the annoying hum has been identified, and as long as it doesn't return while I'm not looking, has been fixed. Many, many thanks to All, especially Tony Sayer. Now it remains to order the remaining parts, and install them, and reassemble it all back to proper condition. Externally, the only thing wrong is the rubbed-off white lettering identifying what the Volume Control/Power Switch/Speaker Selector does.

Soon I'll be able to play my freshly rebuilt Pioneer PL-12D deck through the Goodmans!

[The Natural Philosopher]

YAY!!!

Its so easy for us to say 'well a finger on that screen would have affected the hum level and taken us immediately to the core problem' but when you aren't used to poking on live kit and poking probes into it, its not that easy to get there as quickly.

[Davey]

When I worked in industry, mostly in the USA, it used to be common to debug a large electrical control panel while it was open and operating. Then various 'Elf and Safety rules came in, and when I finished working, we were having to wear stifling shock-proof suits if we even opened a panel door, and the thought of putting a DMM's probes in there would send the Safety Officer into shock. A huge amount of this was due to overzealous interpretation of the rules, but there was rarely any appeal to a Safety Decision. We knew what we were doing, but that didn't count. One of my two most memorable experiences concerning control panels was after a modification had been done to a panel, nothing special, just adding a motor contactor and associated kit. I checked the installation out, closed the panel door, then pushed the button to start the new motor. There was a loud bang, the whole surround of the panel door was illuminated by a bright white flash, and everything then it went dark. It turned out that the added lugs for the feed to the new fuses and contactor were so close to the subpanel that they shorted to it, through the paint, when power was applied to them. Everything went dark because the fuses in the bussplug failed, as they should have done. It was lucky that I had fully closed the panel door, or it would have been all in my face, at high speed. So yes, poking around inside, especially near large capacitors, is something that I am very wary of!

[The Natural Philosopher]

I did (part of) my prenticeship at Marconi Radar where the klystrons and magnetrons were powerful beasts and had tens of KV at amps of current available.

In order to prevent idiot squaddies in the field killing themselves, opening the rack doors cut the power to these.

The standard way to circumvent this was to use a sixpenny bit, which, wedged between a convenient ledge on the case and the microswitch, would restore power to the units.

-------------------

On a different but related topic, the most unusual fault I ever had was when designing a power amp - I used to do a lot of that. If they went unstable - and engineering the best compromise between high frequency performance and being a radio transmitter in meltdown was a large part of that - much of the outputs stages would simply melt down, and it was then standard practice to go back to the last stable setup, replace the blown devices and try again.

On on occasion as soon ans the replacement devices got hot, massive distortion resulted. I guessed that an output transistor had gone, and so it proved. Now when they do that, the standard failure mode is punch through from collector to base, easily identified with a curve tester or even a multimeter. Of which I had both. This one was different. It had a base emitter short. More oddly, it was a real short. Not the sort of 'I am still sort of a semiconductor, but broken' short.

Puzzled, I gazed at the TO3 package, examined it, and could see nothing. I asked the boss for permission to use his mechanical worskhop which featured a vice and a hacksaw. "Why?" "I have a blown transistor that's blown in a way I have never seen, I want to cut the top off and see why, maybe?" "Just replace it and firget it" "No, if one dsoes that, I want ti understand why, in case another one ever does" "Oh wll, okay them"

So I removed the tin top and got a magnifiying lens on it, and yes, I could see what had happened.

The bond wire to the base was well off centre and ALMOST touching the emitter pad. And there was the tiniest of bits of slightly molten arcy looking whisker bridging the two.

Those of you who are familiar with transistors will know that at best about 0.8v exists across a base emitter junction when its working or maybe -1v or so backwards. But that had been JUST enough with the heat and thermal expansion of the bond wire to bridge that gap and cause a teeny whisker. Normally the bond wire would have been so far away it wouldn't have been a problem or so close it would never have passed factory test. The one in a million chance had dropped into the stores.

"So what was the point of doing that" the boss asked when I showed him. "To make sure it wasn't something I should design to deal with" I said. "This is one in a million, if it ever happens again it will be just a failure that you will repair and send back, not a design flaw.

I think he is, 30 years on, still selling that amp design.

[Davey]

I have had a few similar items in my toolbox at times! Allen-Bradley even sold a relay defeater, for testing purposes only, of course.

[tony sayer]

.

:!...

Humm ...seemed very odd that the balance control didn't do anything which o course it couldn't as it wasn't connected to the O volt line!..

Just part of the tone controls..

Letraset;?..

Enjoy:)...

[tony sayer]

In article , Davey scribeth thus

Theres a vid on Youtube, was a CCTV recording on a bloke being slaughtered when a cabinet blew out with a short. A man from UK power networks was telling me he once saw the bolts holding a cover plate on a

11 kV transformer being blown off after a rather odd short;!(...

[Davey]

My only reason for mentioning that was because the resistor is a style match for the R60. If the RS one mentioned yesterday is ok, then that's fine. Since I will have 10 of them, I will probably change both channels' R60, so they have a new matched pair.

Yes, my thinking too. Then maybe some lacquer over it to preserve it.

[Dave Plowman (News)]

They have a minimum order charge. And IIRC it shows up on the final sum which you pay.

CPC don't - and free postage on web orders too.

[Davey]

Their big Delivery page said that Delivery was Free to all account holders ordering more than £20 worth, which was well covered. The

4.95 charge for private accounts is hidden deep in the T&Cs, certainly not displayed on the Delivery page.

To me, that is mis-representation.

I'll check CPC (Farnell) again, for this second order. Thanks for the suggestion.

[tony sayer]

In article , Davey scribeth thus

It doesn't do anything critical.. a matched pair?, the makers wouldn't even thought of it and I wouldn't either a standard 5 or 10 % tolerance will be fine no power as such is involved;!....

[Davey]

That makes that simple, then!

Thanks.