Webcor conversion, odd distortion - diyAudio
Go Back   Home > Forums > Live Sound > Instruments and Amps

Instruments and Amps Everything that makes music, Especially including instrument amps.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 30th September 2011, 07:36 AM   #1
diyAudio Member
 
Join Date: Sep 2011
Default Webcor conversion, odd distortion

Webcor Amp.jpg

This is my first post, and I am looking for help in eliminating an unusual and objectionable distortion from an otherwise great-sounding amp. I will attempt to include some scope views and a schematic.

I purchased a Webcor 166-1 amp on EBay. Originally this was built in the 1950s as a phonograph amp, which included a second channel for a microphone. It uses a 6AT6 preamp tube, a 12AX7 phase splitter, and a dual 6V6 power section, with a 5Y3GT rectifier. It has a 10'' speaker. I am hoping for a good-sounding light weight amp and a DIY adventure with a happy ending. My prior projects were restoring some hifi amps (Fisher X202-b and a Magnavox SET home-brew conversion). The guitar amp I have been playing sounds great but is too heavy for me to carry happily (Semour Duncan Convertible 100).

There is some limited information on the web about others who have made this conversion. The 6AT6 input tube invites comparison to a Fender Harvard, and its cathode biased power tubes invite comparison with a Fender Deluxe 5E3, and reportedly DIYers have made some conversions with those classic amps in mind. I recapped the amp, modifying only the 6AT6 input section to match the Harvard 5F10 schematic. It was necessary to modify the input section, as it previously was intended to serve as a microphone input using a circuit that I don't fully understand, and an input jack I have never seen before. I added two input jacks with 68K resistors to the grid of the 6AT6 preamp tube for the high gain channel (formerly the microphone), and the 12AX7 phase splitter for the low gain channel (formerly the phono input). Prior to recapping, the amp passed only the weakest signal, so I replaced all caps with modern equivalents within 10%, at higher voltage limits in most cases.

Recapping was successful, as the amp then sounded really excellent up to the point of slight overdrive, at about 12 o'clock on the volume control. Really special tone up to that point, enough to make it worth continuing. Both channels worked fine, each with their own volume control, and the high gain channel with plenty of volume to really be useable direct from a guitar. Unfortunately, a buzzing sound starts as the volume is increased beyond 12 o'clock. It is a harsh, mechanical sounding buzz, like a loose part. At full volume, the amp certainly is usable and strong but with a fairly raspy distortion that is not musically endearing. The buzzing persists through a known good speaker cabinet, proving that it is an electronic issue, not the speaker or cabinet.

On the oscilloscope (MacCRO, freeware) injecting a 1KHz sine wave, the sine wave is pristine until the volume reaches 12:00. Then, it breaks down immediately by 12:30 with appearance of a negative directed deep spike beginning near the crest of the positive phase, and returning to positive before the negative phase of the sine wave. Please see attached scope views.

This occurs through either the 6AT6 channel or the direct low gain channel (previously the "phono" channel), which just goes directly to the 12AX7 phase splitter. It persists with other known good 12AX7 tubes. It is not altered by switching the 6V6 output tubes.

As an asymmetrical distortion I am suspicious of the very asymmetric phase splitter section in this amp. In that this is not a "clipping" pattern but rather a deep "notch", I do not have a ready explanation for what is happening here, and I am not aware of what would happen in the splitter or elsewhere to cause this. I am posting this because I hope one of you may have experience or ideas which would shed some light.

In the attached schematic, please note the unusual feedback loop from the output transformer to the 6AV6 cathodes, which also seems to serve as a cathode bias arrangement with bypass cap. I may want to play with an alternative cathode bias arrangement without the feedback, as in a Deluxe. Also please note the unusual phase splitter circuit, with the two triodes of the 12AX7 in series (it seems to me). I have not seen this in other guitar amp schematics. Any comments on this would be appreciated. Thanks!


Webcor 166-1 schematic.jpg

1Khz 6at6 preclip 12;30 0.75mac vol.jpg

1KHz 6at6 1;00 0.75mac vol.jpg.
  Reply With Quote
Old 30th September 2011, 08:06 AM   #2
Enzo is offline Enzo  United States
diyAudio Member
 
Join Date: Sep 2004
Location: Lansing, Michigan
The phase splitter is basic enough. Your signal feeds into V2, the 12AX7, where it is amplified. That signal in turn is sent to the grid of the upper 6V6.. R11 and R12 are the grid return resistance for that 6V6, but they also form a voltage divider, and the signal at that grid is sampled by that divider, and the result is used to drive the other triode in the phase splitter. The original signal is inverted coming out of the first 12AX7 triode, and so is exactly what the second triode needs to provide opposite polarity signal to the lower 6V6.

You could have a parasitic oscillation going on. Look up that term. Assuming the original amp was stable when it left the factory, I would be suspicious that the construction you did around the inputs may be an issue here. You report either input will create this symptom, but does the 6AT6 stage volume affect the symptom when using the other input? In other words are the two "channels" interactive?

If you can cause the symptom using the low gain direct input, does removing the 6AT6 make a difference?

Did you run any signal carrying wires from point A to point B anywhere? In other words did you add something like a control or jack wiring that adds a few inches of wire? Such wiring can act like an antenna, picking up and radiating signal through the amp. Many home builds have had to correct wiring "dress" problems to make them stable.
  Reply With Quote
Old 30th September 2011, 04:35 PM   #3
diyAudio Member
 
Join Date: Sep 2011
Thanks, Enzo, for the quick response. This is just the kind of help I was hoping for. It will take a few days for me to follow these suggestions, and then I'll re-post.
  Reply With Quote
Old 2nd October 2011, 05:37 AM   #4
diyAudio Member
 
Join Date: Sep 2011
Default Webcor progress report

Enzo suggested that this might be a parasitic oscillation and made some suggestions and proposed some tests. The two channels are not interactive in creating the distortion. Removing the 6AT6 tube does not affect the distortion while using the low gain channel.

I went through the construction I did around the inputs and found several areas for improvement in wiring "dress". The two new inputs, which are open cage type, now have the shortest possible length of cable possible, and now I am using shielded cable with the grounding shield soldered to the ground point shared by other preamp components. The open cage inputs are now insulated from the metal chassis, decreasing the possiblities of a ground loop. I have rerouted mains current and the ground terminal for the three-prong power cord at a distance from the inputs. I also found two connections that seemed to never have been soldered at manufacture (back in the 1950s), and soldered them.

The amp gained some sonic improvement and lost some hum, which is great. The offensive distortion now kicks in with the volume at about 3:00. The amp is therefore more useable at low level play, but still not satisfactory.

I added a defeat switch for the power tube negative feedback loop. This switches off connection from the OT secondary to the 6V6 cathodes, and replaces it with a combined cathode ground for the tubes -- a 200 ohm resistor in parallel with 25uF cap. The original circuit provides cathode grounding with parallel resistor and cap at the center tap of the OT secondary, as a part of the negative feedback circuit.

This defeat switch gave a noticable gain in power and more headroom before the objectionable distortion kicks in. The distortion still prevents using the amp wide open and enjoying it. I installed the switch on the panel and will keep this feature for now.

As suggested by Enzo, I have been reading up on parasitic oscillation (thanks for pointing me in this direction). In addition to careful shielding and dress of the input section, it seems like having proper grid resistors is important. Looking at the schematic for the Fender Deluxe 5E3 as well as the Harvard 5F10 (similar period amps with similar tube complement to my Webcor), it seems that a grid resistor in series with the the coupling capacitor is standard (1500 ohm). The Webcor has 470Kohm resistance between grid and ground, but only a coupling cap between the phase splitter tube plate and the 6AV6 grid. Could this be the source of my parasitic oscillation? The Webcor, of course, was not envisioned as a guitar amplifier, and its circuit might have to be adapted to solve my problem.

So, with the above changes, here are some scope views. I have to believe that the clues are all here, and that someone out there will recognize them. Now, with a 100 Hz sine wave and the volume at 3:00, here is the distortion seen previously:

Attachment 242757

It shows a nice symmetrical soft clipping pattern (sounds so nice) with a wicked sharp brief downward spike just as the waveform passes the isoelectric point. This does not repeat on the downward limb of the wave, only on the upstroke. Its downward deflection is much greater in amplitude than the clipped sine wave itself. What allows that high negative voltage to get through?

Now, switching off the negative feedback to the 6V6s, and changing nothing else, we see that we have gained headroom against the wicked downstroke, which is now much smaller:

Attachment 242758

There are two new spikes appearing however. These are just after the crest and trough of each wave, and are also sharp spikes in the opposite direction of the prevailing voltage change. Possibly the negative feedback is successful at supressing them. I am not clear on what creates such sharp spikes, however.

Finally, at full bore with negative feedback off, here is the waveform:

Attachment 242759

The pattern looks so specific. I wish I had the electronics background to interpret it. Would a parasitic oscillation be so sharp? Could a capacitor be triggered to fire like this with a falling voltage? Would love to hear opinions on this.

Failing a specific diagnosis, I may have to fall back on copying the phase splitter circuit from the schematic of a benchmark amp such as the Deluxe or Harvard. I'm thinking about adding a grid resistor to the 6AV6 as seen in these amps, as a first step.
  Reply With Quote
Old 2nd October 2011, 03:50 PM   #5
diyAudio Member
 
Join Date: Sep 2011
Location: Texas
I am a newbie at all of this amp building. My experience is limited to one kit tube amp that I built. I really have no business giving advice, but here goes . I'll share my experience.
I had a problem that sounds similar to what you are experiencing.It was a parasitic oscillation. I used what is commonly referred to as "chopsticks" to arrange and move wires and such while the amp is oscillating to find the offending wires or parts. A tiny amount of movement can make a big difference! Route the wires right next to the metal chassis where possible, for as much of their length as possible. I also replaced a couple of wires with shielded cable, while also paying attention to routing them next to the chassis as mentioned. I eliminated almost all oscillation using this method. I eliminated the remainder by adding a grid stopper resistor when I had gone as far as I could with the chopsticks. 68k is the recommended value for a grid stopper. I didn't have a 68k, so I used a 56k, it worked great, and the oscillation is gone. Good luck, it sounds like a cool project.
  Reply With Quote
Old 2nd October 2011, 06:01 PM   #6
diyAudio Member
 
Join Date: Sep 2011
Default Webcor distortion

I just checked my post and found that the graphics did not come come through. I am just repeating the part with the scope graphics, as they seem to me like they may lead to a specific diagnosis:

"So, with the above changes, here are some scope views. I have to believe that the clues are all here, and that someone out there will recognize them. Now, with a 100 Hz sine wave and the volume at 3:00, here is the distortion seen previously:

100 Hz 6AT6 yNFB 3;30 10-1-11.jpg

It shows a nice symmetrical soft clipping pattern (sounds so nice) with a wicked sharp brief downward spike just as the waveform passes the isoelectric point. This does not repeat on the downward limb of the wave, only on the upstroke. Its downward deflection is much greater in amplitude than the clipped sine wave itself. What allows that high negative voltage to get through?

Now, switching off the negative feedback to the 6V6s, and changing nothing else, we see that we have gained headroom against the wicked downstroke, which is now much smaller:

100 Hz 6AT6 nNFB 3;30 10-1-11.jpg

There are two new spikes appearing however. These are just after the crest and trough of each wave, and are also sharp spikes in the opposite direction of the prevailing voltage change. Possibly the negative feedback is successful at supressing them. I am not clear on what creates such sharp spikes, however.

Finally, at full bore with negative feedback off, here is the waveform:

100 Hz 6AT6 nNFB fullvol 10-1-11.jpg

The pattern looks so specific. I wish I had the electronics background to interpret it. Would a parasitic oscillation be so sharp? Could a capacitor be triggered to fire like this with a falling voltage? Would love to hear opinions on this.

Failing a specific diagnosis, I may have to fall back on copying the phase splitter circuit from the schematic of a benchmark amp such as the Deluxe or Harvard. I'm thinking about adding a grid resistor to the 6AV6 as seen in these amps, as a first step."

Thanks, faulkner 1953, for the chopsticks idea. Maybe doing that while the amplifier is on the scope will be informative. Regarding grid stoppers, I have 68K resistors on the input and phase splitter grids. The power tube grids have 470K and 480K in grid-to-ground resistance, but no series resistance from the plate of the phase splitters. The specs of the 6V6 seems to allow me another 20K maximum resistance to add before reaching the maximum allowable for a cathode biased circuit (0.5 meg). I may try adding 10K in series on the grids of the 6V6 to see if that helps.
  Reply With Quote
Old 3rd October 2011, 04:50 AM   #7
diyAudio Member
 
Join Date: Sep 2011
Default Webcor distortion fixed

It looks like the odd spiking distortion is fixed. I added 10Kohm grid stopper resistors in series from the phase splitter coupling capacitors, with no improvement. The thing that fixed the problem was removing the 10MFD capacitor from the OT secondary center tap. This cap does double duty for the NFB circuit and as bypass cap for the cathode bias resistor. This may be an unusual NFB circuit, as it does not appear in other guitar amp schematics I have seen. As the unit was originally intended for hifi reproduction (or at least calling square dances in school gym class), this feedback circuit may be overwhelmed by double coil pickups played as we do today. At any rate, this cap stood out as the likely source of a spike too large and sharp to get through the soft clipping output tubes -- as it follows the tubes and has a direct connection to the speakers. Once the cap was removed, the spikes went away:

before:

100 Hz 6AT6 nNFB fullvol 10-1-11.jpg

and after removal of the 10MFD cap from the OT secondary center tap ground:

nNFB full on less NFBcap.jpg

The NFB loop can still be switched on, without the bypass cap, and looks like this:

yNFB full on less NFB cap.jpg

(okay, I could have just said it looks the same) The tiny spikes at the end of each crest, positive and negative, are also gone. They might have been echoes of the large spike, picked up as interference by the input circuit, then split into two by the phase splitter and injected back into the output tubes (I don't know why the time delay). They were blocked well by the NFB circuit, as shown in the last post.

Anyway, thanks for the help with this project. I had my chopsticks ready, and was ready to try that next. The amp does sound great, is loud and full of touch sensitive distortion, nice and clean when you back off a little. A light to moderate strum on a six string chord sounds clean-ish and pleasant, with the amp full on. A little more attack and there is a nice overdriven sound, with a nice swirling thing -- sounds like a guitar, in other words. It seems to like double or single coils pretty well.

I'll be trying to fit a 12 inch speaker in the amp (in place of the 10 inch). Hoping to be able to carry this to practice and have the drummer able to hear me -- we'll see. I'll post the final circuit as modified and some pictures once this is done. Thanks to all!
  Reply With Quote
Old 4th October 2011, 12:48 AM   #8
Enzo is offline Enzo  United States
diyAudio Member
 
Join Date: Sep 2004
Location: Lansing, Michigan
Great, always good when a problem gets sorted.

Just a thought, what if that cap was OK at ohm meter voltages, but when subjected to signal levels in that part of the circuit was breaking down? So then did you try tacking in another cap of similar rating just to see if it was the particular cap itself or just the presence of ANY cap there.
  Reply With Quote
Old 4th October 2011, 04:21 PM   #9
diyAudio Member
 
Join Date: Sep 2011
Default Webcor distortion

Enzo, you asked whether it might be a faulty 10MFD cap at the OT secondary to ground position. I happened to do this experiment (unintentionally) earlier along in recapping the amp. I had not ordered enough 10MFD caps the first time around, and substituted a 25MFD cap in that position, (too eager to hear if the amp would work at all). It had the same distortion by ear and on "scope" -- in fact, the first graphics I submitted were with this cap in place. I later switched it for the correct value cap, seen in the second submission. I know this 25MFD cap works, as it is now part of the NFB defeat circuit in the amp, as cathode bypass cap.

I am wondering if this unusual part of the circuit is acting like an automotive ignition, with the cap charging with each crest of the wave from alternate 6V6, and discharging through the transformer if the voltage across the 200 ohm resistor becomes negative. If everything were balanced, and the OT was new, possibly that voltage would never drop below zero. If the "center tap" is not exactly centered on the waveform going through the OT, possibly a negative voltage could briefly appear there. Just speculating, of course.

Here is the schematic before modifications:

Webcor 166-1 schematic.jpg

And after:

Webcor mod schematic.jpg

The amp, with the NFB circuit active, but no 10MFD bypass cap in place, sounds good. Maybe that cap was helpful for hifi purposes, but the guitar sound does not seem to be lacking so far. The NFD circuit noticable tightens the base response while cutting the volume a small amount. Without NFD is my slight preference for now, but having both available seems worthwhile for now.

Pictured below are the two guitar inputs at right, and the NFD defeat switch at extreme left. The other control knobs are original.

2011-10-04_12-12-12_811.jpg
  Reply With Quote
Old 4th October 2011, 11:30 PM   #10
Enzo is offline Enzo  United States
diyAudio Member
 
Join Date: Sep 2004
Location: Lansing, Michigan
OK, not the cap itself then. I can't fault the design, I doubt the circuit behaved this way in general, at least within its expected use.


And with that parallel 200 ohm resistor, I have to think that cap is not doing a lot of discharging back into the inductance of the transformer.

BAck about 1968-69, I had a Rickenbacker semi-hollowbody bass guitar and no amp. A little Webcor tape recorder, with its little oval speaker, became my amp for a long while. Absolutely terrible for guitar, let alone bass. But hey, what do you do?
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Distortion Neutralizer for Odd Harmonics smoking-amp Tubes / Valves 60 15th November 2007 12:25 AM
Webcor Amplifier Speaker Charlie C Tubes / Valves 2 11th June 2005 04:53 PM
Odd distortion problem soundNERD Chip Amps 10 25th November 2004 12:16 PM
Odd distortion problem wintermute Solid State 16 10th October 2004 01:40 AM
What is worst - even or odd order of distortion DarkOne Solid State 18 8th February 2002 12:42 PM


New To Site? Need Help?

All times are GMT. The time now is 09:56 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2