Squealing

[MikeSafreed]

I have an Express clone that I built and sold a few years ago and the customer asked me to install a PPIMV. So, I did that. Amp now squeals like the OT primary leads are reversed, but it only does that when the Impedance selector is set to 8 ohms (into an 8 ohm cabinet). When I change the selector to 16 ohms the squeal goes away. Would that impedance mismatch cause the squeal to disappear? My initial checks indicate that the PPIMV is wired correctly to each half of PI and that the OT leads are connected to the correct tubes. But, obviously, I've done something wrong. Ideas? (I know I can simply reverse the OT primary leads and test this or reverse the PPIMV/PI leads and see what happens but I also want to understand how the change to the impedance selector (changing the secondary selection) "cures" the squeal.

[pdf64]

Connecting a 8 ohm load on to a 16 ohm output will reflect a lower impedance load back to the power tube plates and reduce the gain of that stage.
Same as lowering the value of plate load at any other stage.
If that stage was in the positive feedback loop that oscillated, then lowering the gain should increase its margin of stability.
Layout and lead dress are critical with amps that run 'on the edge'.

[Stevem]

are you using shielded cable on the master volume pot mod?
Does the amp have feedback and or a presence control?
Where are you landing the ground for the pot?

[MikeSafreed]

Shielded cable on the wires running from the pots to the PI and also used shielded cable on each wire from the pots to each power tube (to the 1.5K resistors on each tube). All are grounded to the bias connection of the pots.

I really don't think there are any lead-dress issues in this amp. That's an area with which I'm super-careful.

I'm thinking that it's really a simple fix - just reverse the OT primary leads - but I just wanted to understand how changing the reflected impedance would temporarily "fix" the problem.

[larsvictor]

The NFB wire is connected to the 8 Ohm lug of the impedance switch together with the 8 Ohm tap of the OT, right ?

What happens if you just disconnect the NFB? If it´s somehow an issue with reversed OT primary leads the squal should stop then.

Lars

[MikeSafreed]

Lars,

Yeah, that will be the first thing I do tonight when I get off of work.

I will post the results tomorrow morning, or late tonight

Thanks.

Mike

[Stevem]

The key here is to run the amp set for 16 ohms into a real 16 ohm load ( not just a resistor) and see if it acts up, if so then reversing the OT primary leads should be the fix!

[MikeSafreed]

Lars (and all),

Yes, it was the OT primary leads that were reversed. I guess I knew that all along, but what I couldn't understand was that this amp was used for over 2 years, into various speaker cabinets/loads, and it didn't exhibit any negative characteristics related to squealing or howling. Only when I installed the PPIMV did it start acting up. So, my first thought was that I messed up that install. But, it was correct. The only thing that makes sense to me is that the amp must have been on the edge of acting up all along, and the PPIMV just threw it over the edge. And I did check the color coding on the OT several times and, according to the colors it was installed correctly, but this is not the first time I've had an OT that needed to have the primary leads reversed. I don't build a lot of really high-gain amps - I build mostly Fenderesque-style amps and it's always immediately obvious when you have the OT primary leads reversed. What threw me was the fact that the amp would sound just fine when there was an impedance mismatch on the secondary side of the OT. Just another "Senior Moment" I guess!!! Thanks for the help...

[larsvictor]

Mike,

I'm glad you fixed it.
Mystic thing in that power amp - I still do not understand it. Do you know what impedance was selected by your customer?

Lars

[MikeSafreed]

Lars,

The customer had been doing two things and neither caused any squealing or howling. in one case, he was using a 16 ohm cabinet with the selector on 16. Then, he was using an 8 ohm cabinet and the selector was on 8 and sometimes on 16 (he would forget to change the setting), and neither caused any noise. So, I installed the PPIMV for him. When I tested it, the amp would howl and squeal when I used an 8 ohm cabinet with the selector on 8 ohms. When I changed the selector to 16 ohms the howl and squeal disappeared. So, I made sure that the leads from the PPIMV were correct, and they were going to the correct sides of the PI and to the correct tubes. I knew that the problem had to be that the OT primaries were reversed but that didn't explain why the amp behaved OK before the PPIMV.

Mike

[angelodp]

Curious as to how one can be certain, during installation, that the OT Primaries are wired in correctly? In the amps I have built I follow the color code on the layout. Pulling the NFB wire is a good trick.

[pompeiisneaks]

Curious as to how one can be certain, during installation, that the OT Primaries are wired in correctly? In the amps I have built I follow the color code on the layout. Pulling the NFB wire is a good trick.

My understanding is that although the color code layout 'should' always match, on occasion the manufacturers accidentally swap them. Since the fix is so easy, they probably don't stress it too much.

That's the fastest/best one I know of for sure, is removing NFB and seeing if the issue goes away, one wire disconnect is about as simple as it gets.

~Phil

[angelodp]

Yes, is it equal to swap the secondary wires if this problem occurs. The primaries get kinda buried in Marshall's.

[sluckey]

Swapping the secondary leads becomes complicated if there are multiple impedance taps involved. For this reason I always recommend swapping primary plate leads. The tube sockets are usually easily accessible. You may have to use some wire stretchers.

There's another way to easily correct NFB phasing for push/pull amps. Simply swap the grid wires to the power tubes. This is very easy to do on the board for most Marshalls. Usually easier and faster than swapping OT leads.

[angelodp]

Thanks Steve, hopefully I will not have this issue upon power up. Been waiting for some parts from valvestorm, the US mail is very stressed.

So with a PPIMV installed, just swap the board wires where the 220k's were, right?