Strange issue: Power/Regulators OK, but poor power to ICs

[synth-dude]

Hi,

I'm trying to revive an old musical keyboard, its been sitting in the shed for about 15 years now.

When I power it up the lights come on but no sound.  I investigated its got an op amp front end, probing that IC shows it has no power.  So next I checked the regulators, its has 2 7805 regulators and a LM337 for a -6.5v rail, these all check out OK, voltage is good.  Next I checked the main CPU, a Z80A, and its datasheet says it should be 5v at Vcc, but in the circuits its getting about 1.45v.

So basically the power supply and regulators appear to producing the correct voltages, but once distributed across the board the voltages either reduce greatly or go to zero.

I'm not an expert in repairing things so thought I'd ask if there were any common ways this can happen with old circuit boards?  Its a double layer board, but not multi layer.

I'll continue tracing things across the board and maybe something will pop up.

Thanks

   

[Ian.M]

Corroded vias linking top and bottom tracks, and/or dry joints.   

[Cloud]

Connect one probe of multimeter to the output of the voltage regulator and with other follow the track. When resistance will increase significantly you could insulate area with the problem 

[synth-dude]

I pulled the board out completely and cleaned everything, followed all the traces and checked resistance, everything seemed ok.  So I put it back together and now it works.  Good that its fixed but bad that I dont know what the problem was and may emerge again.  I suspect maybe a ground link wasnt conducting and removing the board fixed it.  Anyway. thanks for the ideas, always good to have more solutions in the toolkit.   

[Ian.M]

Maybe, but more likely there's a dry joint somewhere.  Follow the power and ground traces examining each joint with a magnifier under a bright light and resolder any that look iffy.  The classic visual signs of an open circuit, intermittent or high resistance dry joint are a grey ring round the centre of the joint, and when you touch the lead end just above the joint with a freshly tinned soldering iron, you can often melt the middle of the joint without melting the outer ring of solder - which is worth trying, because then you *KNOW* the joint was bad so can be confident you've improved the situation by resoldering it.   One that was only partially cracked will usually reflow completely, and you should keep on looking elsewhere.

[synth-dude]

Ok, I'll do that now, cheers.

[Ian.M]

I should add that the grey ring is quite hard to see, unless the joint has been carrying significant current and the crack has opened up due the the bad joint cooking itself.  Its by no means a sure thing - you can get dry joints that have failed to tin the component lead properly but look fine.  On a few occasions I have actually been able to slip a resistor lead out of the solder fillet!  Far more common are joints with a suspect ring that are still making reliable contact.  Some boards have a large number of joints like that presumably due to problems with the wave soldering process.

Lead free solder makes your life harder as for most alloys there is far far less visual difference between a dry joint and a good one.  When reworking a board, you need to identify whether its Sn/Pb or lead free, as mixing different solder systems frequently results in dry joints.  A trace of the wrong solder in the joint doesn't matter (except legally for ROHS) so desoldering, cleaning up then resoldering is OK.