Bassman aa864 tone stack

[craneclimber]

Hi everyone - I've had this Bassman 50 sitting around for a while I've been collecting parts and trying to find time for it. I figured I'd run into some snags, just not this early on.

History: 1964 Fender Blackface BM50, aa864. Bought on fleabay as 'sounds great, working amp'. Need I say...NOT! Looks like someone who knows even less than I do attempted repairs. Filter caps were under-sized and under voltage, It appears to have been modded to an aa/ab165.
My goal is to restore it to the original aa864 circuit. My only reason for this is that this is how it was brought into the world.

What I've done so far: New filter caps on new fiberboard, new diodes and bias cap. The main fiberboard was such a nightmare to look at that I decided to strip it and rebuild from scratch. I've been working from a schematic from Jack Darrs Amp Handbook and a layout diagram from Schematic Heaven. See links below.

The snag(s): What I see...When comparing the schematic to the layout for the Bass Inputs there appears to be a  difference in the wiring of the bass pot wiper and the deep switch. Also, the bass pot wiper doesn't have a path to ground, which I think it should.
I dug up another schematic from Schematic Heaven, also linked to below, which makes more sense to me, except I'm not sure how the bass pot wiper is wired in. (I'm not familiar with that particular representation of a pot, showing the wiper across the resistor at an angle and not wired to anything). My gut tells me it should go to ground, allowing for a variance to the resistance in that part of the signal path. As big as my gut is, I don't always give in to it, so if anyone can clear this up for me it would be greatly appreciated.

SH = schematic heaven
JD = Jack Darrs Guitar Amp Handbook

[sluckey]

The Darr schematic is wrong. The schematic and layout on schematic heaven are accurate (at least in the tonestack area). The schematic drawing from schematic heaven shows another common way to draw a pot that is wired as a two terminal variable resistor. Looking at the layout may help you understand this way of drawing the variable resistor.

I have the original Darr book from the '60s and I built that preamp along with a twin reverb preamp. I still have the pencil corrections made in my Darr book. I had a real AA864 to compare. Look at my old schematic to see another way to draw that variable resistor bass pot. Electrically, it's all the same.

http://home.comcast.net/~seluckey/76amp/76_Amp_Project.pdf

[FYL]

I've been working from a schematic from Jack Darrs Amp Handbook and a layout diagram from Schematic Heaven.

Most of the schemos in Jack Darr's "The Electric Guitar Amplifier Handbook" have been redrawn, with a few mistakes added in the process. Schemos and layos posted at Schematic Heaven or here (http://www.el34world.com/charts/Diagrams.htm) are repros of the original Fender docs - mistakes, if any, are 100% original.

[craneclimber]

Sluckey, FYL - Thanks for the input and reference mat'l.

I had a pretty rough day at work, so will go over them later.
Thanks again

[craneclimber]

Sluckey,
I've taken the time to draw out the circuit from the layout dwg., doing that and having your dwg to compare it to helped a lot. Confirmed my suspicion that Darr's circuit is incorrect. I'll be working from Schematic Heaven dwg for the rest of this project.
Maybe I should give more weight to my gut instinct next time.
Thanks again

[sluckey]

Stick around and let us know how it goes. We need pics too. Gotta feed the habit!   

[craneclimber]

Looks like it's gonna be a couple of weeks before I can sit down and continue on the main board.
Hope these will hold you over til then.
I'll reflow the solder joints when the boards are installed.

[J Rindt]

In the picture above, of the rectifier, would two of those diodes per side be enough?  Wouldn't two of those new diodes give more back protection than three of those old silver diodes?
Thank You

[craneclimber]

Diodes in series are used to increase the breakdown voltage.
The 1N4008 diodes I used have a PIV of 1000 volts, so yes, you probably could get away with using only 2 on each leg.....possibly even one. Maybe someone else can weigh in on this.
I'm following Fenders design mostly to stay true to their design layout. All of the components I use for this rebuild will be new replacements with tighter tolerances, metal film resistors instead of carbon comp, silver mica caps instead of ceramic disk, etc.

<<those old silver diodes>>
not sure what you mean here?

[J Rindt]

"<<those old silver diodes>>
not sure what you mean here?"

Weren't the original (the ones you replaced)  Fender diodes silver?

[craneclimber]

Ahhh yes,
yes they were.
Fenders originals were silver.
I don't know if that particular style is still available, most repairs I've seen to diode boards have used 1N4007 which are 1.0 amp 1000v.
I said I used 1N4008, I was mistaken.......I used 1N5408, which are rated 3.0 amp, 1000v
Doug carries both.

[J Rindt]

Ahhh yes,
yes they were.
Fenders originals were silver.
I don't know if that particular style is still available, most repairs I've seen to diode boards have used 1N4007 which are 1.0 amp 1000v.
I said I used 1N4008, I was mistaken.......I used 1N5408, which are rated 3.0 amp, 1000v
Doug carries both.

Yeah.....OK, thanks.
One reason I asked is that I installed a new board into a friends AC568 Bassman. I changed the circuit to AB165/AA864 specs. I was thinking I might as well install new rectifier diodes while I am inside this amp. I believe those old diodes were rated at 600 volts. So I was thinking that 2x1N4007 (on each side) would be plenty, and then just one in the bias circuit.

[craneclimber]

If you have a # off the original diode I can try to look up the current rating for it.
If you decide to go with the 1N5408 you will have to open up the eyelets ever so slightly, the leads are a bit fatter than the 1N4007.

[J Rindt]

If you have a # off the original diode I can try to look up the current rating for it.
If you decide to go with the 1N5408 you will have to open up the eyelets ever so slightly, the leads are a bit fatter than the 1N4007.

The diodes are:
RCA 35604
Thanks

[craneclimber]

Sorry, I am unable to find any specs on the RCA 36504 diode.
Concensus seems to be to go with the 1N4007 at the very least.

[HotBluePlates]

I can't quote the specs for the original diodes, but I can tell you why 3 per leg if you didn't already know.

Old solid state had junctions that did not withstand high voltage, outside of selenium diodes (which were very large and already outdated). Maybe each of the "silver diodes" could withstand 350v of Peak Inverse Voltage (PIV). That's how much voltage the diodes is rated to withstand when biased off.

When you charge a filter cap through a rectifier, then have the negative a.c. cycle at the power transformer, you have perhaps 350v * 1.414 = ~495v on the filter cap, and another -495v peak present at the power transformer winding. The rectifier sees 495 * 2 = 990v from one side to the other.

Well, when you have a diode with not enough voltage rating, you stack them in series just as you would with filter caps. Three 350v diodes in series will tolerate 350*3 = 1,050v which is plenty for our example transformer and filter cap circuit. The series diodes only pass as much current as any 1 can pass, so with 1N4007's we are limited to 1A continuous. You will see that a single 1N4007 per leg is likely sufficient, and 3 is massive, massive overkill. We don't have any guitar amps that need to have a rectifier rated for 3kV (or approximately 1.5kV at the filter cap). The 1N5408 is overkill in another dimension, as 3A is much more than we need.

1N4007 is used everywhere because it is common and cheap (and smaller-rated diodes are also physically smaller and harder to handle without damaging). Use whatever you are comfortable with, but these days 3 diodes in series is only for maintaining original looks.

[craneclimber]

HBP - Thanks for filling in the blanks on the diode question.

<<< 3 diodes in series is only for maintaining original looks. >>>

The main board on this amp is being completely redone, hopefully with correct original layout, not necessarily the original look 'cause of orange drops, metal film, etc.

I just got back from a quick trip cross country, and if things stay fairly calm at work I should be able to stuff that main board this week.
I'll post some pics when the board is done.

[PRR]

> Old solid state had junctions that did not withstand high voltage, outside of selenium diodes

Selenium rectifier breakdown is low.... under 100V?

But they are basically "plated washers", with a fairly cheap plating (more than zinc, a lot less than pure Silicon). They were stamped out like washers and "stacked" to the desired voltage: a few for 100V, a whole bunch for 1,000V. It is about as exotic as shoving fender-washers in your Chevy until the door and fender are in line.

The most basic Silicon breakdown is 7V. You can tilt this by doping. But if you over-dope, the diode doesn't conduct well. And the required doping must be precise in amount and depth/gradient. It took some years to get this cookie-baking consistent. Which is why you have the 1N4001-1N4008 series. I remember when '01 was cheaper and there was no '08 (it stopped at '07 which was too costly to use casually). Today the '01 and '08 are the same price (in DIY quantity) and there is no reason to use anything less than '08.

In the "2-diode" rectifier plan, the PIV is more than twice your DC voltage. Strictly, it is twice the no-load voltage, plus overhead for line surges. So when using a 50V CT winding to make 70V DC, you need at least 140V PIV and anything less than 200V-300V PIV is risky long-term.

By the same thinking, your "450V DC" amps need 900V diodes, and 2,000V is a better plan. But even well-baked Silicon hardly goes over 1,000V per junction. (Which is already pretty fabulous if you bust-open a diode and look how thin the wafer is, much less the layer which blocks 1,000V.) Moreso since everything today is 5V or 3V and high-volt diodes are going out of fashion. Sluckey had some 10,000V rects and I do not want to know how much his company paid for those.

[craneclimber]

PRR-
This is why I really like this forum, getting feedback from you and Sluckey makes me realize how much I have forgotten over the years.
I remember enough to get by on minor repairs and rebuilding a board, but a lot of the theory and complete understanding of the circuits has been lost to me over the years.
I think I'll dust off my old class notes on power supplies and filters and review them.
Much obliged.