Help me fix my 8516

[sweetbeats]

View attachment 135524

If i unplug the heads MOST of the noise is gone and the vus return to rest.

I think it’s time to scope the audio path to/from the record/play amp cards. Scope the noise. I’m wondering if you have some HF oscillation going on that you can’t hear but the meters can read.

 

[ecc83]

that is a great idea, iv been trying to track down a milivolt meter but i cant seem to find one used locally. Will havr to look into this build a bit more.

Levell TM3B A.C. R.M.S. Microvoltmeter Volt Meter | eBay

<p dir="ltr" style="margin-top:0; margin-bottom:0;">Levell TM3B A.C. Microvoltmeter Volt Meter. In working order. Checked that it measures a signal which seemed fine however it may need a service/calibration. Been modified to be powered with a 9v supply via red and black wires on the back. There...

www.ebay.co.uk

Naturally a bit risky and it will need the calibration checking but Level kit was very good. The input attenuator will for instance still be accurate. That unit is probably battery powered and that is fine because it avoids ground loops. The early models had a rather lethal mains supply! SMPSU replacements are of course now cheap as chips.

You can still buy an analogue AC mV meter which seems to be a 'clone' of the old Levells* but be the thick end of $300!

*Note, many of those meters were scaled in dBV, some dBu as well so be aware. IMHO, dBV is more logical anyway!

Dave.

 

[happybirthday]

I think it’s time to scope the audio path to/from the record/play amp cards. Scope the noise. I’m wondering if you have some HF oscillation going on that you can’t hear but the meters can read.

Im stumped. Noise on all channels, but coming from the heads???

I need a little push.

 

[sweetbeats]

Not coming from the heads. But with the heads connected it completes the circuit. The first step I was suggesting is just to get a literal pocket, using the scope, of what the noise is. What amplitude, what wave form, what frequency.

 

[happybirthday]

im seeing 200mVac on the 0v rail and the bias osc. when using a card extender. im having a hard time catching it on the scope.

heres a schem.

There is some other mACv hanging around the amplifier but the bias and 0v are the strongest.

 

[happybirthday]

Ok i have a real update... i found the fuckery

 

[happybirthday]

At first glance.... i was measuring continuity between the osc output and the 0v rail. didnt make sense that they were shorted. i dunno.

Pulled the board out to get a better access with my probes.... holy mackrel, someone butchered this thing

Desoldered these caps to see what the brown stuff was. Looks like glue, they look like they were reused out of some other electronic project....

A FEW broken traces and broken "user" p2p solder joints............................



Im almost excited right now, sweet hell, ill try to sort this out the best i can, i really hope the master OSC isnt fried, but i wouldnt be surprised if it was.

Luckily i found this whole board online, now to ship from EU....

Good evening gentelmen, thanks for all your help so far, iv learned 10x what i knew when i started tearing into this thing. Noone is going to fix it for me, but i really appreciate the time taken to teach me what you know.

 

[sweetbeats]

The caps are C2 and C4? It always helps to tell the reader what parts you’re holding relative to the schematic.

 

[happybirthday]

Yes c2 c4.

 

[happybirthday]

 

[sweetbeats]

im seeing 200mVac on the 0v rail and the bias osc. when using a card extender. im having a hard time catching it on the scope.

heres a schem.

View attachment 135614


There is some other mACv hanging around the amplifier but the bias and 0v are the strongest.

I’m sorry but I can’t follow what you are saying here…0V of what? Tell me what you’re probing by referencing the schematic…there are multiple 0V references. You’re looking for ripple on the power rails (+15V, -15V, etc.), not the ground reference. But this is all moot because clearly your filter PCB assembly is roadkill and you need to fix it. If the traces are hashed cut them out and put in jumper wires, or use the tails of new components as point-to-point jumpers.

Those caps…those were in C2 and C4??? Those are supposed to be 1000uF parts. Those “2m2” caps are 2.2uF. Total and complete butcher job. Make it right and re-check. Post any questions.

 

[happybirthday]

My apologies for being unclear. Let me explain for archival purposes.

Without really knowing what im doing, i was scoping and metering ac current on the amplifier cards (using a card extender) U and L connectors. (Will post a diagram below)
I was looking for ac noise, it showed up on a few pins on the connector for the individual amp cards however was the strongest on the 0v rail and the bias input pin. I was unable to catch the waveform on my scope. It was out of control, kept changing.

This is when i pulled the amp unit out of the machine and took it apart finding the filter pcb. It is in an awkward location, after trying to scope and meter awkwardly i decided to pull the little board out and that when i found this mess.

Im really hoping i can make this board work, ill try my best, but if the Master OSC is busted i may be out of luck, gotta figure out how to test it/take it apart.

Should have some time today to play around with it.

Question does the 0v rail = ground, or no?

 

[sweetbeats]

Question does the 0v rail = ground, or no?

Short answer: maybe.

Point to consider: that doesn’t really matter if you are analyzing a DC power supply’s output for AC ripple, proper voltage level, etc.

Longer answer: “ground” is a bit of a misnomer of a term, because there are different types of “ground”. You have an actual ground or “earth” reference in a domestic, commercial or industrial power service, where the service panel has a ground buss bar physically strapped to the earth via a copper stake driven into the ground…it’s there to dissipate current that, due to a failure, goes where it’s not supposed to go and in order to avoid it going through you we have the “safety ground”…electricity, like water, always seeks the path of least resistance. Then devices have a chassis ground, a common reference for the device and this chassis ground may or may not be connected to “earth”…the chassis ground may “float” (and you can know if this is the case if the power cord has only two prongs, and not a third ground pin, and the chassis is not somehow otherwise connected to the service safety ground). Then you have DC power supply common references. And then you have circuit “ground” or common references, and this includes signal ground or shielding. This also may or may not be connected to the chassis. It depends on the device and the circuit. Usually it is. So regarding your power supply, and “0V”, think of it this way…a +15VDC supply is 15V right? 15V from what? The 15 volts DC is a power differential measured between two levels. You have to have a common reference from which to measure that power differential. We call that common reference the “0V” reference. It in and of itself is not a power rail…neither is the +15V. Together they make a complete circuit and comprise the power rail. It could be +13V and -2V. It’s still 15 volts. So, sure…call it a ground. Call it the common reference. Or call it the 0V reference because that’s what it is. This is even more relevant when there is a common 0V reference or null shared in a dual power supply, where you have +15V and -15V, for instance. The 0V reference is at the center of the 30V differential between the + and - outputs, and the current ”flows” through the electrons at each output, albeit in opposite directions, because of the bi-polarity in a dual supply. If we call it a “ground” that’s kind of a misnomer because “ground” comes from “earth” and that has to do with the safety ground in the electrical service, and, again, the chassis, power supply common reference(s) and circuit grounds in a device may or may not connect to “earth”…they may float instead. This is why, when measuring for things like AC ripple at the output of a regulated DC supply it’s important to measure between the 0V reference and the supply output. If you measure between the chassis (for instance) and the supply output, you may get false results if the 0V supply reference is not strapped to the chassis via a low or “zero” resistance path. You’re trying to measure for ripple across the supply differential. Not a supply output and a separate supply output…not a supply output and a signal output, not a supply output and some other supply’s 0V reference.

Hopefully that helps.

 

[ecc83]

And when measuring ac signals and especially with a scope, find the "0" reference or "0 volts" point as close to the point of measurement as possible. Don't form a big loop.

Dave.

 

[happybirthday]

Thank you both, this clarifies a lot for me. I understand some of the information i was providing was in accurate and confusing.


Here are some findings from today:

Recapped the Filter PCB, cleaned the board the best i could, then measured resistance between all the pins, in and out.

Found a short on the VCO, between pins 14 and 13. Pin 13 is also shorted to the 0v input, which is why i think i was having the issues in the first place.

I removed the Oscillator from the board and without everything is measuring correctly as far as i can tell resistance, continuation.

I ripped the cover of the VCO unit, the circuit board is epoxied, im going to try to melt/remove the epoxy, maybe i can reverse engineer the unit after testing.

Any advice greatly appreciated, will report back as soon as

 

[sweetbeats]

I’m a bit lost…sorry…why are you measuring resistance across all the pins? After recapping what is the problem you are having? Why are you tearing the master bias oscillator apart? Maybe I’m missing it or there’s some information to add, but you found a bunch of roadkill solder work, and some very wrong and very zorched caps. You fixed all that. What is still not working right? What did you find operationally deficient when you repaired and recapped the filter PCB assembly and powered the system back up?

 

[happybirthday]

Sweet; when i put it all back together pins 13 and 14 are shorted together. This threw me off. 13 and 14 are both shorted to pin 4,5,6 - 0v input from PS, which i believe was the reason for the crazy oscillation in the amp channels (bias signal coming thru the 0v rail)

Without the VCO soldered to the board there is no short between pin 13/14/456 on the filter pcb and the output legs of the oscillator which go to 13/14 are shorted , which is why i didnt run it.

Your thoughts?

 

[tmcstudio]

I think it’s time to scope the audio path to/from the record/play amp cards. Scope the noise. I’m wondering if you have some HF oscillation going on that you can’t hear but the meters can read.

RF from Capstan Motor/Bearings might be an issue. Close proximity to playback head. I had to replace motor/bearings as last maintenance I did on the deck. A VTVM is what I use when tracing anything analog, along with a dual scope if needed. Sorry about the board tracing being in that condition! It's curable though Good luck.

 

[tmcstudio]

RF from Capstan Motor/Bearings might be an issue. Close proximity to playback head. I had to replace motor/bearings as last maintenance I did on the deck. A VTVM is what I use when tracing anything analog, along with a dual scope if needed. Sorry about the board tracing being in that condition! It's curable though Good luck.

 

[seankerns]

Why are there so many "empty" posts on this thread? Makes it impossible to follow the narrative.