Guitar amp for Wes is running!

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Some of you may remember that I was working on a guitar amp for my grandson. Well it is up and running in beta test version. I am very excited about getting it this far. I now have till the 18th to get it tweaked and the initial voicing done.

Did the voltage tests over two nights.

Saturday:
The first thing that I discovered is that the switches on most of the 1/4 jacks were not working (always open). I examined them and discovered that the spring pressure had moved the contacts upwards in the plastic body. I suspect that over zealous soldering softened the plastic enough to damage the locking mechanism. So I used a pair of needle nose pliers to put them back in position and put a dab of super glue on in hopes that this will keep them in place. In the long run I may have to replace them. Anyone else run into this problem before?

When I got to the part that involved plugging it in things went a bit more smoothly. The voltages out of the transformer all fell in line with expectations so I installed the recto and the standby switch worked as desired, bleed resistor was working and no caps exploded. :bigeyes:

I proceeded with installation of the signal tubes and fired up again. The parts that were supposed to glow did and the ones that were not supposed to glow didn't. So I went on to measure voltages starting at the power tubes (the theory being that if there was excessive current in the power tubes I wanted to know about it ASAP) and I got as far as measuring the PI tubes before I quit for the night. The results are

371V - 6V6 Plates
320V - 6V6 Screens
21V - 6V6 Cathodes

306V - B+ for PI
217V - PI Plates
40V - PI Cathodes
30V -PI Grids

290V - B+ for final gain
283B - B+ for input stages

With both input stages and the MV at 50% I could hear a little bit of noise like faint rushing water and maybe a touch of hum with my head about one or two feet from the un mounted test speaker.

Sunday:
Well it didn't take long to get the rest of the voltages. They are as follows.

CF:
290V Plate
111V Cathode 100k
106V Grid

Final Gain:
106V Plate 100k
1.4V Cathode 820R/0.68uF

Normal Channel:
167V Plate 100k
1.1V Cathode 820R/330uF

Bright Channel:
205V Plate 100k
2.2V Cathode 2.4K/0.68uF

BTW, schematic is available at http://tubbythetubie.embarqspace.com/ under the "Guitar Amp Project" page.

Since there was no stray DC on the input jack I decided to hook it up to my keyboard (don't have a guitar) and it did make music. The normal channel was very full and the bright channel was very bright so that is as it should be.

I was able to get some fuzzy distortion with the keyboard output and the preamp gain controls set high enough. With controls set lower the sound was very clean.

With the cascade switch on of course the tone was dominated by the bright channel but the gain was quite noticeably higher and it was pretty easy to get the fuzz without blowing the eardrums out.

I was doing the tests using the Bad Betty (93dB) speaker as I wasn't yet confident enough to risk the Man-O-War.

So all in all functional but there were a few glitches that need to be tweaked.

1. There is a little bit more noise than I like. Hiss and something like a buzz is clearly audible at normal playing volume though not deafening. Fortunately there is no hum. I only used shielded cable on the input to first tube grids and used regular wire for all other connections such as gain and tone controls. I could try going to shielded in more of the signal path.

2. The treble and mid controls work as expected but not as much difference as I expected. Maybe using a guitar will make it more obvious. The Bass however seems to make much less difference, works backwards and seems to have most of its effect in the very first few degrees of travel. I though sure that I had wired it correctly but I may have a couple of wires crossed. It acts like there is a linear pot in there so I will have to recheck and make sure it is log.

Any other things to check in this regard. Could a loading issue be the culprit.

3. The gain and MV pots have a little glitch at the very bottom of their travel just as it comes in. This may just be a feature of the pots and should not be a problem as long as the behavior is confined to the turn on area. The pots are conductive plastic.

A friend of my granddaughter here in town has volunteered his services to test it out so will know more when he gets his guitar on it.

Advice welcome.

mike
 
2. The treble and mid controls work as expected but not as much difference as I expected. Maybe using a guitar will make it more obvious. The Bass however seems to make much less difference, works backwards and seems to have most of its effect in the very first few degrees of travel. I though sure that I had wired it correctly but I may have a couple of wires crossed. It acts like there is a linear pot in there so I will have to recheck and make sure it is log.

One thing I noticed is the wiper in the schematic is on the wrong side of the pot based on others like the Fender Bassman which uses this common tone stack.

Poor resolution near the bottom of the gain and master volume pots is pretty common, I see this a heck of a lot with alpha pots time to time.

The first thing that I discovered is that the switches on most of the 1/4 jacks were not working (always open). I examined them and discovered that the spring pressure had moved the contacts upwards in the plastic body. I suspect that over zealous soldering softened the plastic enough to damage the locking mechanism.

I have had this happen several times with the Neutrik brand jacks.
Though similar to Cliff Jacks, the plastic compound is not very heat resistant, Even a light touch soldering and they loosen up pretty bad.
Switch to the Marshall style Cliff branded jacks will help this or just use the Switchcraft jacks like the old fender amps.

I suspect the 500PF bright cap is a bit extreme (for my taste) I rarely use over 250pf and often only 100pf. That might be contributing to the hiss you mention.
 
Thanks Trout. I have some old switchcraft style jacks in my salvage bin so I am replacing the plastic with these. I am one short so will have to pick up another for the input (though so far it is not giving me any trouble)

When using the metal jacks would you recommend insulating them from the chassis. Right now the one that I have installed is not insulated but I am still running a dedicated ground wire to the signal ground point.

I have a guitar player coming over today to test so I will see if the bright sounds excessive. You are talking about C11 and 12 right?

I did notice last night that the noise is quite reasonable when nothing is connected to the input so it may be something in the keyboard or the cable that I am connecting with that is responsible for the bulk of the noise. After todays testing we will have a better idea.

As to the bass pot wiring, you are right! I drew the schematic wrong and then wired it according to the schematic. :xeye:

Pics are posted here...

http://www.hawthorneaudio.com/forums/viewtopic.php?f=8&t=2139&p=29684#p29684

mike
 
Test report

As promised a short report on the results.

Jake brought one of these http://www.musiciansfriend.com/product/Tribute-by-GL-S500-Premium?sku=511117 a couple of cables and one of those digital multi-function pedals (sorry I didn't get which one). We ran the guitfiddle into the pedal and then into my amp. We started with the output hooked up to the $25 surplus speaker (Bad Betty) until I convinced myself that it was safe to try the more expensive Eminence Man-O-War.

The noise level was much lower than when my keyboard was hooked up to it; very pleased with that. We got very loud levels before the hiss really began to show up. Switching over to the MOW both I and the guitar players felt that it had the better tone as well as being 10dB loader. It actually had really nice bass so I was pleased with that.

Without pedal distortion we were able to produce some preamp distortion using just the amp and guitar controls but feedback was a limiting factor here. We were at ear splitting levels but right on the edge of feedback. I had noticed that the tubes were showing some microphonics when turned up pretty loud (running finger along the chassis could be heard in the speaker) so I suspect that finding some less microphonic tubes might be of help here. I am not well schooled in the feedback mechanisms in guitars but it is possible that the tremendous volume levels involved may also have been exciting the strings and causing incipient feedback. Considering the possibility that direct electromagnetic coupling between circuits might be involved as well, I think it wise that I go ahead and change to shielded wire in as much of the signal chain as I can. This is especially so since with sufficient gain (cascade switch on) we were able to get feedback at more moderate volume levels.

The players also agreed that the bright channel was really brighter than it needed to be so changing those 500pF caps in the bright channel to 250pF is probably a good idea and should further reduce noise. Whether it might help the feedback also I don't know.

All in all I think the noise is acceptable and that it can be reduced somewhat. Feedback now seems to be the major issue to deal with. Reliability is of course an unknown quantity that will only be fully addressed with time.

Many thanks for all of the assistance and moral support.

mike

BTW, this will be a 1x12 combo so microphonics is probably a very important issue.
 
Great, will order some and have them shipped to MD so they will e there when we arrive. The Chinese 6V6s don't seem to be complaining so I may go ahead with them rather than order another matched set of better tubes.

I will also order some 250pF silver micas to tone down the bright channel a bit. I have some 33k carbon film that I think I will sub for the 68k metal films that I am using for the input grid stoppers and see if that helps the noise at all.

I have been informed that I am pushing the 5U4GB a little bit hard with that 30uf 1st filter cap so will order a 22uf to ease up on the recto a bit. It looks like it will increase the ripple on the plates by about 5V or 31% which is a negative I suppose but I don't want to fry the recto. Total DCR of the PT secondary is 137ohms (measured). Any input on whether this is a good idea.

mike
 
A Little fuel for thought,

#1
The guitar you tested with uses single coil pickups, those often feedback more easily as a rule of thumb. I have one Strat that you would swear the pickups were nothing more than microphones near the strings. You can actually talk into them!

My good Strat has the lace gold noiseless pickups and do not produce any odd noises or microphonic tendencies.

#2

You might want to think about using an EH12AY7 as the first pre-amp tube. NOS being best, I use EH12AY7's in the V1 position on most of my amps. reason? less noise, a bit cleaner sound, a smoother onset of distortion/overdrive.

#3 DO NOT USE Sovtek tube in either V2 or V3.

Especially the 12AX7LPS. Sovtek
There has been much talk on guitar amp forums about this tube. They have a known design flaw that could in some cases cause damage.
Read about it here

Considering price, JJ12AX7's generally cost a couple bucks less than most of the tubes listed above, I have had better luck with the JJ over any current production 12AX7.

#4, The cap listed as C12 is also an extremely high value(500pf) for that position. I generally use 100PF and it is still piercing and bright.

I have a guitar player coming over today to test so I will see if the bright sounds excessive. You are talking about C11 and 12 right?

Yes, and also C6, most Fender amps use 250pf for this position, and Fender amps are pretty well known for brightness and chime.

Anyway, those are just a few thoughts, voicing an amp to individual taste is pretty subjective.

Trout
 
I think that Wes's guitar is this one...

http://www.musiciansfriend.com/product/Epiphone-G400-Electric-Guitar?sku=512590

Which has humbuckers. Would that likely be less inclined to feedback.

You might want to think about using an EH12AY7 as the first pre-amp tube. NOS being best, I use EH12AY7's in the V1 position on most of my amps. reason? less noise, a bit cleaner sound, a smoother onset of distortion/overdrive.

Hmm... I have a brand new Electro-Harmonix labeled 6072A/12AY7EH. Would that be a good input tube to try? Would I need to change any of the resistor values? Are they generally pretty good WRT microphonics? I was going to use it for some hi-fi experiments (CFs and phono pre amps) but I could use it in his amp and get something else for my experiments.

Thanks for the heads up on the LPS.

So C6 and C11 of 220 to 250pf and C12 of around 100pf (120 OK?)? Changing C6 to 250 in the tone stack calculator shows a drop of a few dB in the midrange centered a bit above 1kHz but little change above 2 or 3k. Does that sound right to you?

mike

P.S. Trout, what do you think of the resistor and cap changes I mentioned in the previous post. Another person who's experience I value has suggested the CF would be less susceptible to EMI and the smaller value would generate less noise all else being equal.
 
I have been informed that I am pushing the 5U4GB a little bit hard with that 30uf 1st filter cap so will order a 22uf to ease up on the recto a bit. It looks like it will increase the ripple on the plates by about 5V or 31% which is a negative I suppose but I don't want to fry the recto. Total DCR of the PT secondary is 137ohms (measured). Any input on whether this is a good idea.

In 99.99% of the amps I build I use 22uf/500V electrolytic caps from FT (German Made) regardless of whether I use SS or tube rectifiers.

I had a few builds about 2 years ago that experienced arcing of the 5U4 rectifiers once I exceeded the 22uf value.

Another reason I never exceed 22uf is for the legendary sag. If you get to big, the amp gets stiff and has less vintage charm.

Another thing you might find interesting would be ,
This thread

As far as the 12AY7 goes, your current cathode resistors and related bypass caps are pretty much in line with the 1959 Marshall super lead. It is pretty common practice for guys looking to clean it up a bit to just swap the tube.

I would imagine it could be tweaked for perfection, but that once again becomes a personal voicing thing.

Looks like a pretty nice job so far !!

Trout

Edit;
So C6 and C11 of 220 to 250pf and C12 of around 100pf (120 OK?)? Changing C6 to 250 in the tone stack calculator shows a drop of a few dB in the midrange centered a bit above 1kHz but little change above 2 or 3k. Does that sound right to you?

Those values sounds pretty good.
I used to rely on the tone stack calc, but then I discovered than many instrument speakers color the final sound of the amp somewhat.
I ended up tuning to the speaker rather than using the actual mathematical values of the tone calc.
Since I use Jensen Neo speakers 80% of the time, it makes for easy adjustments. When I switched to Jensen C12N all the values went right out the window.
 
Thanks, I will take the amp upstairs and swap in the AY and see what happens. Of course I am back to using the keyboard but I should be able to do an informal test for microphonics compensating for volume level.

So far the only 500V 22uf I have found is an illinois at triode electronics in their bargain cap section. 450V are common though. Given max B+ of 370V or so do you think that 450 is enough margin? Is Illinois cap any good?

mike

Actually I just found the F&T at Watt Amps for $5. That is reasonable and he has the silver micas and switchcraft jacks that I need too. Price is a little higher than some but as long as he can delivery quickly (tues or wed) it is OK.
 
Good news and bad news

Well the good news is that I have fixed the bass control, the microphonics are taken care of and the noise is manageable and will no doubt improve when I get to MD and switch those caps. I can tell because the normal channel can be turned up all the way and the master almost all the way before there is much noise at all. Bringing in the bright channel increases it quickly.

(Schematic linked in post number 3)

Now the bad news. I have not gotten around to testing the preamp out until today (power amp in had already been tested and worked as expected). I checked the jacks for continuity and the resistance between the input side of C9 (input to PI) and the wiper of the master volume is zero when nothing is plugged in and equal to the MV setting when a dummy plug is plugged into either jack.

When I hooked the preamp out to the input of my PA amp and play my keyboard the signal does get through to the PA amp and the gain and MV controls work as expected in that regard but, the music also plays at a lower level through the guitar speaker even with the MV all the way down. Turning the MV up has a little bit of effect on the guitar speaker output but very little. The gain contols on the other hand have a pretty fair effect. The output from the PA very quickly overcomes the output from the guitar speaker but the leakage is troubling to an extent.

When all is said and done I suppose that if it is a problem he can easily unplug the speaker output since I have the 250ohm safety resistor in there but still it would be nice to fix it properly some day. We are leaving in the wee hours of Monday morning so no time to implement something major like output buffer amp but... What is your best guess as to the real fix?

Hmmm... now that I think about it the resistance is zero when the MV is turned down and that shouldn't be...

mike
 
I figured out a work around. On the theory that a shorted input can not develop a signal to amplify I made up a grounding plug. Plugging this into the return (power amp in) jack kills the audio in the guitar speaker without having to disconnect it. I tried it out and it works fine.

I suspect there might be some issues if one were to forget and leave it connected after pulling out the send plug but as long as both treble and MV are not all the way up there would be at least some load on the CF so it shouldn't be too likely to cause damage.

I suppose that (1) finding the source of the coupling and eliminating it or (2) using a different switching scheme for the jacks would be the proper solution but this should get us by since I have to head out with it tomorrow.

I will be taking my stuff with me and checking in back here while on the trip so if anyone has any suggestions go ahead and post.

Next step is painting the face plate. I figure I will go ahead and put the base color and a layer of lacquer on and that way I can just mail him the decals when I get them done.

mike

P.S. turns out I was remembering incorrectly when I said that the resistance was zero with the MV turned down.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.