Question on Bassman AB165

[JammyDodger]

Hey All,

I just picked up a '66 Bassman. I opened it up to look at the circuit and found that the bias had already been changed over, so far so good. Then I started looking into the feedback loop. This has been modified as well. There is the 820 Ohm resistor except it is not connected to the OT it is open on one side! There is a 270k Resistor where the 100 Ohm should be to ground. Any reason for this?

The amp sounds 'decent' although there is room for improvement.

Any ideas why this was done?

Cheers, Mike

[Bob-I]

Some people like the more aggressive sound of no NFB.

[jaysg]

270k?!?! I'd put it back to stock before going much farther.

[Wayne]

This sounds like a prime opportunity to explore. You've already located the components in question, and one of them is even unhooked. Now that you know what your amp sounds like with no feedback at all try it with the original FB network values, and then start experimenting with progressively larger or smaller values - you may find something you like, and you can always revert to factory values (or none, as it was when you got it).

W

[Firestorm]

Disconnecting the feedback loop is a pretty common mod. But the (what should have been) 100R is part of the cathode circuit for the PI. I'm kind of surprised it works with 270K. I'd get that puppy back to stock right quick.

[M Fowler]

If you do not have a presence control then .1 mfd cap across the 47k or 100k resistor where the 820 ohm resistor is and can use 10k, 15k or 22k to ground where the 270k resistor is now.

Blackface driver circuit has .1 mfd 600v cap, 100k, 82k on bottom V and 100k on top. The feed back circuit has .1 mfd, 1m, 470, 1m going to 6.8k resistor in the middle from there to the left should be 4.7k to ground. Also off the 6.8k in the middle to the left should be 470k-1m resistor to ground. They all vary.

Mark

[Ron Worley]

Hey Mark, did you draw that after a couple martinis????

Sorry, couldn't resist.... all pun intended....

[M Fowler]

Hee hee

Thanks Ron wish I had. I need to figure out how to use all the fancy computer imaging stuff you guys use. I can't draw worth a crap using a mouse either. After words I thought I should have just imaged that picture and edited what I wanted, oh well.

Mark

[Scottie]

Just looking at the parts in your picture, your circuit looks more like the AA864 than the AB165.

I would definitly put a 100r in place of the 270K. Whether you want to keep the 820 NFB connected is up to you.

As a guess, I'd say that someone along the line tried to convert the circuit back to more of a standard fender NFB scheme using the AA864 setup.

The stock AB165 feedback setup is kind of odd for a Fender. The feedback line runs through a series cap and they feed the 6L6 plates back to the PI through 220K resistors. Note the carbon on the chassis where these probably used to be.

Where they got that 270K, though, I'm not sure.

Scottie

[M Fowler]

My favorite the tremolux has 820 NFB and 100R to ground, AB763 which the CBS period driver. I used to change this to the brown tolex mod pre-CBS with 25k pot and 4.7k to ground

Bassman NFB can vary from 470K to 1M. The ground of the driver can be 100R but the stock bassman is 470k and modified usually with 220K.

From my take on this they were experimenting for a certain tone with the 270K

Should work just fine with the 270K if not lower it until you like it. Hook up the 820 to the ext. speaker jack for NFB.

Mark

[Scottie]

Really? I haven't seen a lot of them, myself. I have a '64 Tremolux, and it has the standard 820/100 setup. This what you have?

If I understand, you have modded the the feedback resistor with a pot to make a kind of "soul" control? Cool. I did that on my Blues Deluxe.

I guess I'd still recommend that Mike put the more standard AA864/AA165 NFB values in. The AB165 is a bit odd.

Scottie

[Firestorm]

Ummm ... the resistor we're talking about (I think) is the load resistor for the NFB loop. It's chosen in tandem with the feedback resistor itself to establish the ratio of feedback you want: 820R/100R (or sometimes 47R); 27K/5K (usually implemented as a Presence pot); also sometimes 100K/5K; 56K/4K7 paralleled by a 25K pot. By '63, the 820R/100R(47R) values had become established.

The thing to consider is that that resistor sits between the PI's tail resistor and ground (so it gets added to the PI's ground circuit). Anything over 50K or so should pretty much shut off ALL gain in the PI. And if the feedback loop were connected would put huge amounts of feedback into the PI making it sound pretty anemic.

If you do reconnect the feedback circuit, a resistor that large is even potentially dangerous because the PI will find a much easier ground reference through the 820R and through the SPEAKER. There is a small amount of DC present here and running DC through a speaker is very naughty. I've actually seen a Blackface Fender kill a speaker when the PI ground connection came loose.

[Scottie]

Ummm ... the resistor we're talking about (I think) is the load resistor for the NFB loop. It's chosen in tandem with the feedback resistor itself to establish the ratio of feedback you want: 820R/100R (or sometimes 47R); 27K/5K (usually implemented as a Presence pot); also sometimes 100K/5K; 56K/4K7 paralleled by a 25K pot. By '63, the 820R/100R(47R) values had become established.

I agree, lift the FB resistor if you like, but floating the ground reference for the PI could have some weird things happening. Mike, even if you don't want any NFB (that is, keeping the 820 disconnected), Fire is right -- use something way less than 270K to tie the PI to ground.

Scottie

[M Fowler]

I was wrong the ground of the driver should be 100R. Did this amp motorboat with that 270k resistor in that spot?

Mark