Bassman AA864 vs AB165

[mresistor]

Just want to discuss the differences I notice in these two Bassman amp circuits.  I have read that the AA864 is the best blackface Bassman and was made under Leo's leadership, and the AB165 was the first amp in the Fender line that CBS started messing with, during and shortly after the takeover of the company.  First off I notice that the bias circuit in the AA864 is typical BF style and the plate load resistors on the PI are unbalanced, 100K and 82K..  pretty much blackface style. The PI to 6L6 coupling caps in AA864 are .1uf where in the AB165 they are .022 which in my thought would cut down on bass freq but would make upper bass freq tighter. The feedback on the AA864 is from the OT to the PI through an 820 ohm resistor in normal blackface style, while the AB165 goes through a 45K and then a .1uf cap and ties into pin 2 of the 12AT7.  AB165 also has localized feedback applied from the 6L6 plates back to the input of the .022uf coupling caps. Why did they do this? There might be some minor differences in the preamp but I haven't looked at that yet, rather I am concentrating on the PI downstream.


I think conversion to the simple blackface bias circuit as in the AA864 is a definite good thing to do. That balance bias circuit does what it should, but could cause distortion in the output at higher levels, until you unbalance it again. In my thoughts there are other ways to eliminate hum, and I question whether it was a non-existent problem that CBS was trying to fix. (figures  )


I don't like the localized feedback with those 220K resistors on the 6L6s. Why would CBS want to do that? Fender didn't do it on the AA864 Bassman nor on the AB763 Deluxe, Super or Twin.


I don't like that feedback arrangement from the OT either and it's not like the other  blackface amps.


So wouldn't it be prudent to implement these AA864 characteristics into an AB165 Bassman?  Especially if you are going to play it out and wish to have better tone?


BTW   another question, this AB165 Bassman I have has three wires coming off the OT and the layout and schematic only show two. I have the black ground and a green and green/yellow stripe with the latter being hooked up and the former capped off. Is this amp setup for 8 ohms instead of 4 or vice-versa? 


http://el34world.com/charts/Schematics/files/fender/Fender_bassman_aa864_schem.pdf

http://el34world.com/charts/Schematics/files/fender/Fender_bassman_ab165_schem.pdf

[sluckey]

I suspect that OT is not original. Notice the splice on the blue wire?

[mresistor]

Yes Steve I did notice that splice and wondered myself. The transformer says  Fender EIA606 412 on it.  I'll go back and look at the pics in the other thread.

[mresistor]

In Pauls Bassman pics that Sluckey posted before in another thread, that Bassmans OT has chromed/silver end bells and it looks like black/brown/red input and black and green output wires. The black input wire could be a very dark blue. So this OT could be a replacement. Who knows. The dates on it would indicate Dec 1964, which would very much coincide with the other transformer codes and the serial number.

[HotBluePlates]

... So wouldn't it be prudent to implement these AA864 characteristics into an AB165 Bassman?  Especially if you are going to play it out and wish to have better tone? ...

For what instrument? Bass? Or guitar?

Just want to discuss the differences I notice in these two Bassman amp circuits.  I have read that the AA864 is the best blackface Bassman ...

GW strikes again! (just kidding)

If you'd like to delve into what/why the changes do, we can help with that. I'm guilty of having owned an AB165 Bassman and converting it to AA864, before further modifying one of the preamp channels into a 5F6-A/Marshall-style preamp. In hindsight, I realized (as Sluckey noted in the other thread) that the amp was intended for bass players... So the changes aren't necessarily bad.

I notice Dave Funk added a dig at Gerald Weber in his book, saying essentially "A guitarist wouldn't be playing through something as lowly as a bass amp in those days, especially one with no effects!" I think Dave was mostly talking about the 5F6-A Bassman, but same goes for the 60's amps as well.

As a preview of a discussion of the changes:

... The feedback on the AA864 is from the OT to the PI through an 820 ohm resistor in normal blackface style, while the AB165 goes through a 45K and then a .1uf cap and ties into pin 2 of the 12AT7.  ...

I don't like that feedback arrangement from the OT either and it's not like the other  blackface amps. ...

It's not like the other blackface amps in that the injection point for feedback is moved from one side of the long-tail inverter (pin 7, "tail resistor") to the other side (pin 2).

The long-tail already had a means of blocking d.c. to pin 7 in the form of a 0.1uF to "ground". That has to be manually included when the feedback is moved to pin 2 (which accounts for the 0.1uF seen in the AB165).

The series resistance in the feedback loop is increased, because the overall circuit impedance at the new injection point is different (higher) than the original 100Ω to ground.

So it's not necessarily "bad" it's just a different implementation. It's not immediately apparent to me (because of the multiple paths to a.c. ground) whether the amount of feedback has been changed. With amp-in-hand, it would probably be just as easy to run tests on both circuit arrangements to see if the AB165 is less sensitive (requires more phase inverter input voltage for same speaker voltage/power), which would then prove more feedback (except we know we have local feedback at the output tubes, which will impact the result).

You should know that if you switch to the AA864 feedback loop, you have just changed the feedback polarity (by switching sides of the inverter), so you'll have to swap OT primary or secondary leads. And one of those primary wires is already trimmed too short to reach the other output tube...

[HotBluePlates]

... The transformer says  Fender EIA606 412 on it.  ...

In Pauls Bassman pics that Sluckey posted before in another thread, that Bassmans OT has chromed/silver end bells and it looks like black/brown/red input and black and green output wires. ...

The stock colors are Blue, Red, Brown for the primary and Green, Black for the secondary.

The code you posted indicates Schumacher, but there should be another set of numbers on it to indicate the model of transformer. The "...412" means "12th week" of '64 or '74 or '84 or '94 (we don't really know).

I don't recall seeing a Fender transformer with a Green/Yellow secondary. Then again, I don't recall seeing Fender OT's with multiple taps, but that doesn't mean they don't exist. I had an all-original '67 Princeton Reverb with an Export power Transformer, which had multiple capped-off primary taps.

[mresistor]

Thanks HPB - I'll get back after I digest what you have said  for now   the other numbers on the OT are 021009 .


Also   do you remember what the bias setting was for your AB165? Did you run it hot-warm-cold?

[mresistor]

HPB   one could also switch the secondary wires on the OT. And this amp is to be played by 6 string guitar. He's a guitar teacher. I also forgot to mention I need to take the death cap out of this amp.

[Willabe]

HPB   one could also switch the secondary wires on the OT.

You should know that if you switch to the AA864 feedback loop, you have just changed the feedback polarity (by switching sides of the inverter), so you'll have to swap OT primary or secondary leads. And one of those primary wires is already trimmed too short to reach the other output tube...

[HotBluePlates]

Thanks HPB - I'll get back after I digest what you have said  for now   the other numbers on the OT are 021009.

The original OT part # would be 125A13A or 022871.

Also do you remember what the bias setting was for your AB165? Did you run it hot-warm-cold?

The schematics for the AA864 and AB165 Bassman show a bias voltage of -44 or -45v. In all likelihood, if you have average 6L6's that voltage will be a good bias point. This is because the output stage is designed to take a certain-sized voltage input (about 43-44v peak to one 6L6 grid) to deliver maximum clean output power.

Back when I had a Bassman (which I sold sometime in the late-90's) I would have only known enough to follow the Ken Fischer/Gerald Weber rule of bias by idle current. The "70% Rule" didn't exist back then, so most of us followed the "idle at 35mA rule" which GW printed (even though he said there's no one number to idle at, and apologetically offered 35mA as a target that should be safe, most of the time, with 6L6's).

If you switch to adjustable bias and fiddle the bias pot while playing, there will be a wide range where the amp sound doesn't change. If you can even get the amp biased cold enough, there will be a spot near maximum bias voltage/minimum idle current where the amp sounds weak and/or distorted on small signals, but fine when turned up. If you bias very hot, max clean output power will be reduced (meaning the amp will distort sooner) and you might hear an increase in low-midrange.

If the amp ran class A, maximum output power would only be about 30w, when using real 6L6's. If you do bias very much hotter than the -44v that's stock, the amp should be tested up near max volume while watching the output tubes, to be sure they don't redplate with loud signals.

[HotBluePlates]

... this amp is to be played by 6 string guitar. He's a guitar teacher. ...

Gotcha. That defines what he may be shooting for then.

Just want to discuss the differences I notice in these two Bassman amp circuits.  I have read that the AA864 is the best blackface Bassman ...  First off I notice that the bias circuit in the AA864 is typical BF style and the plate load resistors on the PI are unbalanced, 100K and 82K..  pretty much blackface style. The PI to 6L6 coupling caps in AA864 are .1uf where in the AB165 they are .022 which in my thought would cut down on bass freq but would make upper bass freq tighter. ...  AB165 also has localized feedback applied from the 6L6 plates back to the input of the .022uf coupling caps. Why did they do this? ...

Overall, because this was a bass amp, and Fender made changes to make it a better bass amp. Plugging a guitar into it (because it's a lot cheaper than the other blackface amps now) is a relatively recent occurrence. What people have to say about the amp relates to this not-as-designed use.

The AA864 has an adjustable bias voltage, while the AB165 has a bias balance circuit. Since Fender wanted you to use the design-bias when replacing output tubes, the switch to a bias balance allowed you to use 2 tubes whose idle currents didn't exactly match. The result is you can plug in 2 unmatched tubes, yet get zero-difference in idle current, leaving all of the OT core's capability for transferring power to the speaker. Unbalanced d.c. in the OT would eat up some of the core's transfer capability at low frequencies first. Obviously that's no good for a bass amp.

The phase inverter itself is identical to the AA864 phase inverter, outside the feedback injection point and the coupling caps to the output tubes. The coupling caps may have been changed due to the added local feedback at the output tubes; you have a feedback loop inside a feedback loop. The low end roll-offs of each loop have to be staggered (or feedback reduced) to avoid negative feedback turning into positive feedback. As you note, leaving the coupling caps at 0.022uF might be a good sonic result even in an AA864 circuit. If nothing else, the 6L6's will recover more quickly from grid-input overload (i.e., "farty distortion").

As for the local feedback around the 6L6's, I'm not certain what Fender was trying to do other than make the output stage as clean as possible. When a loop is over multiple stages, you're never really sure how hard any individual stage is being cleaned up; you only see the end result. If any stage in the loop runs into a hard limit (which would cause clipping), the how section inside the loop loses feedback control and distorts abruptly. Fender may have been trying to clean up the 6L6's individually, or make them less sensitive and less likely to be the weak-link in a clean output stage. Again, trying to keep things clean for bass amplification.

The AB165 has a local feedback loop around the 3rd gain stage of each channel. The AA864 had given a 3rd gain stage only to the bass channel, but had knocked down the signal entering it by half. The added series resistance of the upper part of the divider could contribute noise, and Fender made the change in the AB165 to a different means of controlling signal level at that stage. 470kΩ against 220kΩ sets the gain of that stage at a bit over 2 times rather than ~56 times. Yet there's a bypass for the cathode resistor in this stage as well, because all the lost gain is being used to keep the stage very clean.

[mresistor]

I hadn't been looking at the preamp- but now that you said that the fbk on the 7025 sticks out -


HPB - thank you for taking the time to share your knowledge of this amp with me. Very useful info.

[p2pAmps]

Having built and restored a ton of Bassmans my ears like a AA864 for guitar.  95% of the Bassmans I build the owners request AA864 and are very pleased in the end.
No way to hate a Bassman :)

[mresistor]

Very cool - I'm looking forward to making these modifications ...   thanks mscaggs

[Willabe]

Having built and restored a ton of Bassmans my ears like a AA864 for guitar.  95% of the Bassmans I build the owners request AA864 and are very pleased in the end.

What channel do most guys plug into, the bass or the normal channel?

[HotBluePlates]

Having built and restored a ton of Bassmans my ears like a AA864 for guitar.  95% of the Bassmans I build the owners request AA864 and are very pleased in the end.

What channel do most guys plug into, the bass or the normal channel?

I'd normally plug into the Normal channel, as the Bass channel tone stack caps are sized such they don't adjust guitar signals as well.

But I also noticed I had a leftover triode after converting to AA864 specs, and used it to build a 5F6-A preamp on the Bass channel side. There's a big volume jump and a different sound when you use a footswitch to go from the Normal side to the Bass(man) side in that arrangement. Very nice for solo boosts.

[p2pAmps]

Having built and restored a ton of Bassmans my ears like a AA864 for guitar.  95% of the Bassmans I build the owners request AA864 and are very pleased in the end.

What channel do most guys plug into, the bass or the normal channel?

I think it all depends really.  Many that I build have the Marshall-esque 2204 preamp on the left so it all depends.  I never much liked the stock Bassman left channel for guitar but thats just me. 

[sluckey]

I never much liked the stock Bassman left channel for guitar but thats just me. 

I can't imagine anyone liking the bass channel for guitar. But the 3 triodes make a good platform to modify to be a guitar friendly preamp. Less work to convert the AA864 power amp to look like a more popular AB763, but converting the AB165 is not difficult either.

[p2pAmps]

I never much liked the stock Bassman left channel for guitar but thats just me. 

I can't imagine anyone liking the bass channel for guitar. But the 3 triodes make a good platform to modify to be a guitar friendly preamp. Less work to convert the AA864 power amp to look like a more popular AB763, but converting the AB165 is not difficult either.

I fully agree Steve

[mresistor]

Under the doghouse I find there are no 220K 1W resistors strapped across the 70uf caps.  5F6a style I guess. Might this partly explain why the plate voltage is a little high?

[sluckey]

no

[mresistor]

Oh crap - there they are  under the caps.  RC filter.

[Ed_Chambley]

I prefer the normal channel, but I also mod the Bass channel like MS Scaggs.  The only difference is I use a little more gain for a slightly more overdriven tone.  I have also used a EF86 in this position as well.


The Bassman is a great platform and I have really only kept 2.  One is a early 62 Brown with presence and the other is a silverface Bassman I converted to the AA864 and put it into a head.  It was basically a chassis and the PT was good, but I changed it too.  The chassis already had holes drilled so no loss.


This one I installed a tube loop using the "extra triode".  A Bassman with a loop is a very nice things to have to me as I really like modulation effects after the preamp.  I just set up the normal channel to sound best with single coils and the Bass channel for Humbuckers by lowering the first Bypass to a 3.3uf remove the boomyness and increasing the plate resistor to get more grind.  Really turned out nice.


There is a lot of room in these chassis and I have moved the power tubes using the rectifier hole and had enough room to add a reverb.  Since I prefer plate verb, that is why I like the loop.

[Willabe]

Ok, thanks guys. 

[SoundmasterG]

Here is an article about converting an AB165 to some of the AA864 specs from Andy Ruhl. I've tried these mods in a project amp and they sound great.


Greg


http://acruhl.freeshell.org/mga/main/AA864_vs_AB165.html

[mresistor]

Here is an article about converting an AB165 to some of the AA864 specs from Andy Ruhl. I've tried these mods in a project amp and they sound great.
http://acruhl.freeshell.org/mga/main/AA864_vs_AB165.html

Thanks Greg. -  I read that article and several other sources of info prior to posting this thread.  I have converted to AA864 type bias,  BF bias. And got rid of the local feedback on the 6L6's. Then some .022uf caps in the bass channel with slope resistor changed to 56K. The bass channel is a monster now. Very usable for guitar, with a lot more sustain and not really that overdriven. The deep switch seems to work backwards of the way you would expect. Oh I changed the 6L6 input coupling caps to .047 METs. 

[mresistor]

HPB or anyone else - what is going to be the effect of changing the AB165 PI input capacitor from .1uf to .001uf ?

[HotBluePlates]

HPB or anyone else - what is going to be the effect of changing the AB165 PI input capacitor from .1uf to .001uf ?

Have you changed the overall negative feedback around the output stage?

If not, I'd leave that 0.1µF cap going into the phase inverter. I suspect it was raised to that value from the 500pF used in the AA864 to avoid capacitive voltage division with the 0.1µF coming to the same point from the feedback loop.

The input impedance of the long-tail inverter is much more than the 1MΩ resistor you see due to bootstrapping. This is why the original 500pF coupling cap was able to be used, in a bass amp.

[mresistor]

No sir, I haven't changed anything with the global feedback topology.  Think I will leave it alone . The amp sounds very good as is. Might upgrade the blue original caps in the normal tonestack  with some cork sniffer caps.. Mojo dijon and some French company that has an S logo.

[DummyLoad]

French company that has an S logo

solen? it's good stuff.

--pete

[mresistor]

That's the company Pete.  Solen Fast caps.

[HotBluePlates]

... Might upgrade the blue original caps in the normal tonestack  with some cork sniffer caps...

Do those caps look like the ones below?



If so, they have already attained "cork sniffer" status. Sozo already makes a replica style of those caps, in the same way the Dijons are intended to replicate the "mustard caps" found in some British amps.

[mresistor]

Yes I understand that people like the blue caps in old Fenders.  Only reason I have taken some out is to change values.  HPB  I have been using some old brown caps out of that Hammond organ, in other projects,  they have the black band on the negative (foil) side. They sound really nice. I'll post up a pic.
I think I'll leave the blue caps in the normal tone stack. IOW  leave it alone.


Funny thing I noticed in comparing the AB165 schematic to a layout, is that the layout shows that pins 1,2,3 are used on the 3rd gain stage (7025) but in the schematic   it shows pins 6,7,8 being used.
Confusing   at first.

[HotBluePlates]

...  Only reason I have taken some out is to change values.  ...  I think I'll leave the blue caps in the normal tone stack. IOW  leave it alone. ...

Gotcha. Yes, if you're changing tone stack values, it's reasonable to replace them (but save them if you can use them elsewhere). I interpreted your earlier post to mean a more widespread replacement of those blue caps.

... the layout shows that pins 1,2,3 are used on the 3rd gain stage (7025) but in the schematic   it shows pins 6,7,8 being used. ...

I don't remember what's actually used in the amp. Regardless, the other triode isn't being used... I'm sure you'll sort it out. 

[mresistor]

my mistake   -  I was looking at that third 7025  but its the second one that is being used for the third gain stage.