Summoning the electronics gurus: Helllllllllllp!

[Roboburger]

Subject: 1968 Fender tape echo. The erase head is super weak, as is the sound and performance in general. I figured I'd replace the power supply caps. Schematic here:

http://www.schematicheaven.com/effects/ ... fe1000.pdf

We're working with the power supply, post T1 and rect bridge. If you look at the schematic at C14 positive, you can see that it's supposed to go to, and ONLY to, R33.

But here''s photos of the actual unit:
http://picasaweb.google.com/Roboburger/ ... 3218513314

and the three pictures before this one (you navigate with the arrow icons above the photo) to the left.

I have a hand drawn version of the reality here:
http://picasaweb.google.com/Roboburger/ ... 4628578706

Here's the deal: It's like someone added a continuity dot to the schematic on the negative side of C17!

That's a huuuge difference in what is on the schematic and what is really there. It connects T2 Green Middle to the DC supply, where schematically, it's not supposed to be!

In short: "Huh?"

Do I want to cut the traces and return the unit to the schematic as issued? I don't know enough about it to know if it should be one way or the other!?!? Is the schematic correct or is the reality correct!! Could they have made a mistake like this in production and not caught it?

[pjwhite]

Unless the circuit board never worked right in the first place (which I doubt), the PCB layout must be correct and the schematic is missing a connection dot.
Don't start hacking traces. Try replacing Q4, then the capacitors and finally the resistors associated with the erase circuit until it starts working. An oscilloscope would be helpful to see if you are getting the correct waveforms as shown on the schematic.

[Roboburger]

Yeah, the 'scope has been packed away since the move- this will get it out. I dunno if it's my voltmeter, But I couldn't get any real readings at T2, where the schematic says it should be a 60v 60kHz signal.

I just thought that maybe someone might be able to drop a tidbit like, "The center tap of T2 should/should not be connected to the C14+ line." I was thinking that it was getting it's needs from the C14- line, via R28 & C17. I wish I knew more about Transistor electronics- the published schematic 'looks' more right to me than the reality- which just might be a production error (they've been known to happen).

[nclayton]

That schematic is totally F'ed. Not only is the connection to the bias transformer center tap missing, but the polarity of C14 is indicated wrong. Look at the rectifiers and the polarity of all the other caps.

The connection to the center tap of the bias transformer is actually a negative voltage, and its definitely required (and definitely has to be negative) in order to make Q4 conduct.

The other poster said to replace Q4, and that's definitely a better place to start rather than in the power supply. Q4 is part of the bias oscillator, and it seems pretty clear the bias signal is not being generated or else is very small. That would cause your erase to not work, and normal recording would be disorted -- very gritty crossover type distortion.

But if the machine is erasing a little bit, that probably means your bias signal is not non-existent, just very low. That SORT OF implies Q4 might be OK. I'd say maybe the absolute best place to start would be C17. If that cap died the oscillator would lose a lot of gain. Any of the other caps and resistors attached to Q4 or the bias transformer could be suspect as well, but since C17 is a relatively low voltage electrolytic it seems like the most likely to go.

the polarity of C17 shown on the schematic is correct. The positive side should go to Q4 emitter, the negative side to the center tap.

Ned

[Roboburger]

That schematic is totally F'ed.

Ned

You mean the actuality (my hand-drawn schematic) is F'ed, and not the original schematic, right???

I'm gonna have time on wednesday to work on this, I'll post again soon.

[nclayton]

No, your schematic looks fine to me. On the original schematic there's the error you found, but also at least one other error, the polarity of C14 is DEFINITELY indicated wrong. That schematic is probably nice to have, but definitely take it with a grain of salt. The way it's actually wired shown on your schematic looks correct to me, the original definitely does not.

[nclayton]

I think a big part of the confusion is that when you look at a schematic, usually the positive voltage supply lines are drawn toward the top of the page, and negative ones toward the bottom of the page. It's the opposite here. That transistor Q4 sits between its 2 voltage supply lines, but the one on top is negative, the one on bottom is positive. Probably we can assume that the guy drawing the original schematic wasn't paying attention to this and just put the "+" sign on C14 where he normally would, toward the top, which is wrong.

Anyway, I think you were thinking that the strange center tap connection you wanted to cut was to a positive voltage, and if you were, you'd be justified to think the connection's existence was very strange and inappropriate since it would be going against all the other polarity markings (besides C14), not to mention the conduction direction of the PNP Q4 transistor. Just look at the circtuit again knowing that the polarities are "upside down" and it will make sense to you.

Current has to flow out of the bottom positive rail through R28 into Q4's emitter. then it flows through the collector into the transformer, and then out the center tap to the negaive supply rail (the mysterious C17 connection). Without that connection, there's a dead end for the current and no circuit.

C17 is actually an emitter bypass capacitor. It could just as easily go to the positive supply rail in parallel with R28 as it could go up to the negative rail, or it could go to ground, it really doesn't matter. The reason they sent it to the opposite rail from the emitter resistor is to make basically be humbucking. That way the emitter has the same amount and same phase of power supply ripple on it as the collector, and Q4 will basically be working between a ripple free DC voltage. None of that matters. The point is that C17 bypasses the emitter resistor R28 to eliminate degeneration from emitter negative feedback. If that cap is a goner than Q4 will lose a lot of gain. The Q4 circuit is the bias oscillator. If Q4 loses gain it's hard to say what will happen for sure, but the oscillation could make a stable but much too small signal, it could make no signal at all, or it could oscillate erratically.

If the oscillator is working correctly it should be making the 60v 60khz signal you weren't able to measure. that signal is the erase signal and also the record bias signal, so if it's gone that perfectly explains why the machine doesn't erase and also why when it records it probably sounds really bad.

I hope this makes a little more sense than my other post.....

ned

[Roboburger]

I think a big part of the confusion is that when you look at a schematic, usually the positive voltage supply lines are drawn toward the top of the page, and negative ones toward the bottom of the page. It's the opposite here. That transistor Q4 sits between its 2 voltage supply lines, but the one on top is negative, the one on bottom is positive. Probably we can assume that the guy drawing the original schematic wasn't paying attention to this and just put the "+" sign on C14 where he normally would, toward the top, which is wrong.

Holy Sheep Shit, you're hitting the nail on the head! and I'm having flashes of comprehension. I was mentally orienting all the caps the wrong way, assuming that the positive was up. Now I gotta go back downstairs and look at it again, retrace the "actual reality" schematic.

Current has to flow out of the bottom positive rail through R28 into Q4's emitter. then it flows through the collector into the transformer, and then out the center tap to the negaive supply rail (the mysterious C17 connection). Without that connection, there's a dead end for the current and no circuit.

C17 is actually an emitter bypass capacitor. It could just as easily go to the positive supply rail in parallel with R28 as it could go up to the negative rail, or it could go to ground, it really doesn't matter. The reason they sent it to the opposite rail from the emitter resistor is to make basically be humbucking. That way the emitter has the same amount and same phase of power supply ripple on it as the collector, and Q4 will basically be working between a ripple free DC voltage. None of that matters. The point is that C17 bypasses the emitter resistor R28 to eliminate degeneration from emitter negative feedback. If that cap is a goner than Q4 will lose a lot of gain. The Q4 circuit is the bias oscillator. If Q4 loses gain it's hard to say what will happen for sure, but the oscillation could make a stable but much too small signal, it could make no signal at all, or it could oscillate erratically.

Ooomph, that's mostly out of my league. I started out good, then got heavy.

If the oscillator is working correctly it should be making the 60v 60khz signal you weren't able to measure. that signal is the erase signal and also the record bias signal, so if it's gone that perfectly explains why the machine doesn't erase and also why when it records it probably sounds really bad.

I hope this makes a little more sense than my other post.....

ned

I think the biggest problem here is the fact that when I took 101, Reagan was president. I'm someone who needs the reality to match the diagram or I get lost quickly. I'm in over my head here. I was really hoping just to swap a few caps and not have to actually LEARN anything...

Okay, back up from the basement. It get's wackier. I got a better VOM, with a frequency counter. Voltage across C19 (Green High & Green Low) is 20 Volts 56 kHz. Voltage across erase head is 60 Volts, 56 kHz. Voltage is going up across T2, and not down...

It is still only partially erasing the previous pass, but yeah- the tone of the recording is less Fart-o-rama. If only the motor could pull the tape at a constant speed!

Y'know- I'm gonna take the time and re draw the section. I just noticed that the published schematic features TWO Q4 s, and I don't see an L1-R22-C16 on my circuit card... I gotta go to work now anyway. Lousy Improv Comedy at the club tonight.

[Roboburger]

Yeeeeeeaaaaahhhh! Got the new erase/record head in the mail today, slapped it in and Whammo! It erases beautifully.

http://picasaweb.google.com/Roboburger/ ... 0396226914

Now the pinch roller goes to be redone. Many thanks to Joe Dundovic for helping me think through the situation and having a new head for just 60 bucks! He informed me that the voltage should go up to 50-60 volts across the T2, and not down to 20.

http://www.magneticheadcompany.com is his website. Lotsa cool PDFs there.

he also hipped me to this website:

http://www.srdpc.com/witt/

and he can rebuild pinch rollers. There's a serious flat spot on this current roller. thus the inability to pull consistently.

I'll be taking some time to catalog all that is wrong with the Schematic vs Reality issue that kicked my ass. So far, in the published schematic, we have a cap oriented the wrong way, a missing point where there should be continuity, two items labeled Q4, the part numbers for the two tape heads are swapped, and the voltages for the bias tranny are also swapped...

up for modification: the owner wants the thing to be able to feedback onto itself like his space echo can, but the feedback knob maxxed out is set to a "safe" level for the typical operator by Fender. Let's see if the Maxwell Smart of circuitry here can get that going on. Maybe Lenny from Of Mice and Men is a better analogy. Would Red Greene register with anybody? Howzabout SuperDave Osbourne?

Spec to the decimal place, mark with spray paint, strip wire with an ax, solder with a plumbers torch. That's how I roll.