ABC's of presence control and phase inverters in A/B amps

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Emperor-TK

Well-known member
Joined
Jul 14, 2004
Messages
1,076
Location
NJ, USA
I need help understanding the basics of how a phase inverter and presence control works for an A/B tube amp. I'm in the process of converting a Laney A100H into a Marshall (model 2204). The two are similar enough where I just went with the power section of the Laney, but the tone is very bright, to the point that I can't turn the treble up past 1 or 2. I'm wondering if the presence section might be to blame.

For reference:

EDIT: Image deleted due to hosting issues

I used the 5K presence pot from the Marshall schematic, but for some reason I put a 680nF cap in instead of the 100nF used by both Marshall and Laney (I can't remember why).

So my questions are:

1. How does the pot resistance and the capacitor value effect tone/performance?

2. From what I've read, I see that the inverted signal is taken from the transformer to drive the bottom of the phase inverter section. But I can't figure out what function the 4 resistors in the T-configuration serves.

A brief walkthrough of how this section works would be greatly appreciated.

Thanks,
Chris
 
Good morning, Chris.

This guy does a better job at explaining the LTP than ill ever do, check it out....


http://aikenamps.com/LongTailPair.htm
 
Chris,

What capacitors are you using in the PA section?
Did you keep the originals?
I have recently re built the same amp.

Sorr
 
Kit, That was very helpful, thanks. I only understand about 50%, but I sure know more about the section than I did before. I am going to try his suggestion about scoping the two outputs and playing around with the bias resistors, plate resistors, and feedback section.

Sorr, I completely reworked the amp, making a new turret board. I went with all Solen polypro's.

I actually have the Marshall 50W myself, but so far the Laney/Marshall sounds quite different. I was doing the conversion for a friend who loves my Marshall.

So far, the differences/problems are:

1. 120Hz hum.
2. Too much treble (sounds good with treble at 1 or 2 though)
3. Turning up the presence increases volume, and also adds a buzzy distortion.

Number two I can live with, but I suspect that there are problems leading to 1 and 3. Bad filter caps maybe (the amp hasn't been used for several years). The buzzy distortion and gain boost from the presence control suggests something is wrong with my phase inverter still. That's why I want to scope it.

I've already learned a bunch on this project, even if I don't get an amp as good as the real Marshall. However, in certain ways, this amp is already better than the real thing.

-Chris
 
Chris,
Disconnect the output from the preamp to the phase splitter and see how much hum you have.
Then feed a signal (from CD etc) into the input of the phase splitter and see how it sounds.
If your source doesnt have a variable output, use a pot to control the volume.
Connect the ground of the source & earth end of pot to the bottom of the phase splitter.

Sorr
 
What does "A/B" have to do with anything here?

This is a Fender 5F6-A Bassman derivative. Yeah, it has more tubes and was built by Laney and modded in the Marshall fashion.... it's still a late tweed Bassman.

I can't read those part values. There are better Fender and Marshall schematics around.

FORGET the Presence control. Take it OUT. Do NOT put it back until the amp works good with all knobs at normal settings ("2" is not normal for guitar).

Does your religion allow test equipment? I'd sweep the frequency response, whole-amp and individual stages. The output stage shown, withOUT the Presence network, with a resistor load, should be dead-flat 80Hz-6KHz. If you use the conventional Fender/Marshall tone stack, lift the bottom off ground. The preamp stages should also be flat through the guitar range, with maybe a droop in the bottom octave if you use Marshall's 0.68uFd cathode cap instead of Fender's usual 25uFd cap.

If your guru bans meters, use 100 ohm and 10 ohm resistors to pad-down your iPod's earphone output to guitar-pickup level, and listen to favorite trax through the amp and speaker. Full-band music and speech should sound full-range, with a rising high end. A guitar amp should not make a huge kick-drum sound, but it should "thudd", not "tapk". Cymbals may not have full sizzzle, but should sound like slightly dull cymbals. Speech should sound like a good table radio with a little treble boost.

If you don't have near-flat (ignoring tone-stack), then you got a miswire or a wrong part value. Your symptoms suggest to me that you have a major loss of bass somewhere, which you have to balance by knocking-down Treb. When you find and fix your bass-loss, you may have a ton more gain.

Oh: how do you know the feedback loop is correct polarity? If it is wrong, the amp may squeal at full power, or it may sit not-quite squealing but on the edge, making the treble way too hot.

> what's the purpose for the two different plate resistor values (82K, 100K).

Signal is fed to V104A. It is amplified directly to V104A's plate, and also passes through V104A's cathode to V104B's cathode and amplified to V104B's plate. The short-path has a little more gain than the long (cathode-to-cathode) path. The 82K resistor fudges V104B's gain down similar to the V104A-V104B path's gain.

It may, or may not, balance the push-pull drive. Balance may, or may not, be a good goal for a guitar amp. I do not know if Leo did it this way because he liked the sound, or just because it seemed logical. Certainly, if you have 100K and don't have 82K, use 100K for both. Save this tweak for after the amp is 97% perfect and you are looking for the last two percent.

> for some reason I put a 680nF cap in instead of the 100nF used by both Marshall and Laney (I can't remember why).

I don't know why either. The 0.68uFd (some say 680nFd) cap goes in the cathode of the first stage after the input jack for a Marshall light-bottom slant. The "Presence" control is goofy to begin with, and I'm not even going to think what it might be doing with arbitrary values. Millions of similar amps do not have Presence; many which do have it never use the Presence.... leave it out! Save it for final voicing!! Use Leo's values!!!!
 
1. 120Hz hum.

Like you said, probably dried up filtercaps.

2. Too much treble (sounds good with treble at 1 or 2 though)

Does it have the "classic" treble bleeder cap across the volume pot?
If so disconnect it.

3. Turning up the presence increases volume, and also adds a buzzy distortion.

That is strange......like PRR said leave it out for now. Fix it later.

This PI configuration actually has rather "buzzy" distortion. Its hard clipping when the plates hit B+. We can live with it cause we roll of bass and the standard guitar speaker wont move much past 5-6 khz. Giving a "centered" middy sound.

Also, tubes matter to some extent. Beware of grid current in the PI.
JJ ecc83 would not work in my 70´s marshall 50W head.
 
>>>What does "A/B" have to do with anything here?

Probably a misunderstanding on my part. I'm still learning. I thought push/pull = A/B = an amp with a phase inverter, whereas a single ended amp doesn't use a phase inverter. I guess that isn't correct?

>>> FORGET the Presence control. Take it OUT. Do NOT put it back until the amp works good with all knobs at normal settings ("2" is not normal for guitar).

Will do. So would it be correct to leave the voltage divider (100K series, 4K7 shunt) and just pull out the presence pot and capacitor?

>>>Does your religion allow test equipment?

Yes, we are very progressive that way (Church of the Enlightened Tweeker). It's more a matter of not knowing how to use my o-scope completely (again, I'm still mostly a newb here). I have no problems poking around on line level/low voltage circuits, but a tube amp scares me (at least until I can find time to read my scope's manual cover to cover). I'm also unclear about the right way to load a tube amp when under test. A 5Khz test tone coming out of a 4x12 at high volumes = no fun.


PRR, Sorr and Kit, Very good info, much appreciated. I feel like I have some homework now :thumb:

-Chris
 
You can disconect one end of the cap to disconnect the presence

Some of the newer fenders use two resistors at the same spot the 82K is used. The two are in series and add up to about 100K like the other PI resistor I guess to balence current. The output is taken from the junction of the two resistors and not the plate(91k and 8.2K IIRC). I believe this is in the blues deluxe and some ohers.

But if you look at a bunch of schematics some amps use two 100K and don't seem to care about the balance VOX ac30.....
 
another good site

http://www.obsoleteelectronics.com/The_Archives/the_archives.htm

look for the Cathode Phase Inversion PDF
 
But if you look at a bunch of schematics some amps use two 100K and don't seem to care about the balance VOX ac30.....

Yep, that has to do with the value of the "tail" resistance. We can make it larger to better approximate a current source, giving equal (close enough anyway) swings on the plates. So an AC30 PI is actually better "balanced".......current and voltage. But this is done at the expence of headroom and gain.

I think the tail R in the AC30 is 47k?

Fender/marshall is closer to 10K.
 
I disconnected the presence control and get a high pitched squeal independent of the gain or master volumes. That sure suggests that something is wrong in my LTP section. I traced the schematic and my layout about 20 times and can't find a discrepancy though. I've tried a couple of different tubes to make sure that wasn't the cause.

Does the squeal suggest a specific problem to anyone?

I'm learning how to properly use my scope, so that's my next plan of attack.

So what would happen if the feedback was comming from the wrong secondary of the output xformer?

-Chris
 
You have probably connected the primary wrong, switch the connections of primary to output tubes' plates or switch connections between PI outputs and power tubes' grids.
 
replace solens with mallory 150's or SoZo caps.

solens sound like crap in guitar amps sterile and way too much high end. i dont believe this is the root of your problems with the amp, but one of a few for sure.
 
dramadisease

Don't be so quick with the cap stuff. You don't "know" the amp. What works in one amp does not always work in another.

Have you tried solens? I have an amp with solens that people who tried like a lot.

It depends on the iron, the tubes and the operating points, the circuit. If the iron "blurs" to much you might not want the 150 "sound"

Caps are a taste issue
 
As Alex mentioned, sounds like the negative feedback is positive!
Have you changed any of the wiring in the power amp/ phase splitter section?
If the feedback polarity is reversed it will oscillate or just be on the verge of oscillation.
ie wires from the O/P transformer to the O/P valves anodes?

Sorr
 
Guys, thanks again. The suggestions are really appreciated. I will see if I can get some time tonight to switch grid wires and see what happens. Regarding the phase issue of the feedback: How would I use my scope (dual channel) to determine positive or negative feedback? Would I be looking for polarity inversion of a sine wave input or for addition/cancellation?

I did rework the PI section (made a new turret board). I've traced it multiple times by eye, but it wouldn't surprise me it I still missed the problem.

Regarding the capacitor suggestion: The whole project started out as simply recapping the amp with Solens. Another friend has the same amp with Wima polypros which used to put this one to shame. The Solens work really well in this amp, but of course that is a matter of preference more suited to one of the many "what is the best capacitor..." threads.

-Chris
 
I flipped the grid wires...bingo! It was definitely positive feedback. The presence control now works in the opposite (correct) direction. Originally, I had just assumed that I had the outer legs reversed. I should have mentioned that before. I can also disconnect the presence pot without the squeal. The tone is now back to normal too.

The power-amp section is now working perfectly, and without the hum. I've sent various signals into the power amp by lifting the coupling cap from the preamp section and all was dead quiet. The hum seems to live in my pre-amp section. I think I will try moving some ground wires around and see what happens. Besides the filter caps and grounding issues, is there any other source of hum that I should be looking for?

On a side note: boy I sure do like the sound of a solid state pre into a tube poweramp. Talk about tight, chunky, aggressive clean sounds....completely the opposite of the thin and glassy vintage Marshall clean sound. While testing, I ran the headphone out of my Dean Markley practice amp into the Laney poweramp. I liked it so much that I am considering also building-in a solid state preamp with a "glass/sand" switch. Musicman and SG Systems, they were on to something. The hybrid amps of the 90's with tube preamps and solid state poweramps had it backwards IMHO.

-Chris
 
Back
Top