Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: nateflanigan on July 18, 2010, 09:17:32 am
-
Hey everyone, I fired up my DRRI to "super-vibro-man" build this weekend. For the most part things are good, but their are a few issues I could use a hand with.
1) If I turn the reverb knob past about 3/4 the amp kicks into what I'm going to call self oscillation. It's a high pitched squeal that over powers the guitar signal. I'll I'll quadruple check the wiring and parts placement in that section, but in the mean time assuming all that is correct, what might be the cause and how might I fix it.
2) I've got some hum. Not over the top, even what I'd consider acceptable for jamming with friends but a bit much for studio use. Before I go at it with chopsticks if we could narrow the field a little bit that would help.
-with the volume down all the way on both channels the hum doesn't change.
-If I put the amp in stand by the hum goes away.
-I followed Doug's grounding scheme, I think I got it all correct, but I'll check again.
-It's hum not hiss
Thanks!
-
Does it squeal with the reverb tank disconnected?
-
Nate asks a good question. If it squeals with the tank disconnected, the problem is in or near the recovery stage.
I had greater or lesser versions of the reverb squealing with all three Blackface amps I've built. Here are some suggestions:
1 - shielded cables going to and from the Reverb pot
2 - carefully review every mm of wire going to and from the reverb driver tube and reverb recovery tube. Keep plate leads flat on chassis and sensitive grid leads (and reverb return cable) away from everything else. Remember you have 3 dimensions to work with
3 - make sure that the "return" end of the reverb tank is far away from the PT
4 - experiment with reverb cable placement. Inside the tank bag, I run one cable on each side so they aren't touching. I use one or more pairs of cable anchors on the side of the cab to keep the cables parallel but separated as they go up toward the chassis
5 - some guys report success insulating one or both of the reverb jacks (SEARCH here to find recent discussion)
6 - find a quiet tube for the reverb recovery stage. New EH 12AX7s have worked for me. YMMV
7 - try different 12AT7 reverb driver tubes. In one amp, it was the 4th 12AT7 that worked quietly
8 - if you have more than enough reverb at "5" on the dial, think about reducing gain from the recovery stage by removing the cathode bypass cap
None of these may solve the problem but IMHO they're worth trying, one at a time.
HTH
Chip
-
Does it squeal with the reverb tank disconnected?
Yes, much worse. It went from a low squeal to a freak out squeal.
-
Do you have a 220K resistor between grid and ground for the recovery tube? Usually mounted on the reverb return or footswitch jack?
-
Do you have a 220K resistor between grid and ground for the recovery tube? Usually mounted on the reverb return or footswitch jack?
Yes, I do.
-
Does the squeal go away if you put a ground on the reverb footswitch jack?
-
Does the squeal go away if you put a ground on the reverb footswitch jack?
I'm not sure what you mean exactly, so I turned up the verb until it squealed then alligator clipped from the terminal on the footswitch jack that controls the reverb to ground and the squeal stopped. Of course the reverb did as well. What does that mean?
-
Fresh Start, I've been moving around some wires and have been able to get another notch or two out of it. Implementing the shielded cables is going to take a bit more time. Thank you for the tips!
-
Does the squeal go away if you put a ground on the reverb footswitch jack?
I'm not sure what you mean exactly, so I turned up the verb until it squealed then alligator clipped from the terminal on the footswitch jack that controls the reverb to ground and the squeal stopped. Of course the reverb did as well. What does that mean?
Yes, that's what I meant. May be a clue. Do you have a footswitch plugged into the jacks? If so, unplug. Still got squeal?
If you still get squeal, then use your ohmeter to measure resistance between the recovery tube grid (directly on the tube socket pin) and chassis ground with the reverb pan disconnected. Is it 220KΩ?
-
I haven't been using the foot switch. Measuring from grid to ground I'm getting a solid 229k.
Thank you for the help!
-
another clue...
I sat down to do some playing and forgot I had the reverb cables unplugged, when I turned the amp back on it was super hot, the reverb channel stated squealing instantly. Even after I turned the reverb down both vibrato and normal channels were really hot, meaning louder than I left them and really gainy. If I picked with any strength it was like distortion pedal saturated. That's gotta be a clue to something, right?
-
Disconnect the NFB and see how it sounds.
-
I just had a silverface super in with the same issue.I tried everything including shielded cables,ne reverb transformer,cables,tank,lead dress and replaced every reverb component on the board and socket,tubes,etc..
It still squeals.
Now sluckey mentions the negative feedback and that gets me thinking.Hmmmmmm........
Why would that affect the reverb,unless somehow the phase issue is causing the oscillation?
-
Why would that affect the reverb,unless somehow the phase issue is causing the oscillation?
Notice that we now have more symptoms to go by. Both channels are very gainy. Disconnecting the NFB is easy to do and will quickly tell if that is the problem.
My thoughts about the squeal showing up in the reverb channel first is that channel has an extra gain stage and might get upset quicker than the normal channel. It's also possible that Nate jumped on the first symptom he noticed. Since both channels now show symptoms that 'could' be NFB related, I think disconnecting the NFB is a good next step.
On any new build involving NFB, especially when working with an unknown OT, it's a good idea to eliminate wrong phase NFB early on, unless the amp just sounds perfect.
-
Well doctors, I set the vib channel to a reasonable volume, turned up the verb to squeal then backed of a hair. As soon as I disconnected the NFB the squeal kicked in. I had to turn the reverb down quite a bit to stop the squeal. Another thing I noticed, with the amp set on the verge of squealing if I turn DOWN the treble it brings about that wretched noise.
Also, to clarify, the normal channel never squealed, it was just really gainy and freaked out sounding when I disconnected the reverb cables.
Incidentally, aside from the reverb issue, disconnecting the NFB is the best boost pedal I've ever heard. Once I get this all sorted I'll have to work on a footswitch for that.
I'm going to go put shielded cables on the verb pot now, aside from that, what's my next move?
-
Almost forgot...
I'm using mojo "super reverb" transformers. The plate voltage is 450v and the bias is at 31ma.
-
OK, it's not a NFB issue.
-
PROBLEM SOLVED! :grin:
In addition to the shielded cables on the reverb pot, I used shielded cable from the return jack to the recovery tube. I can turn the reverb all the way up without any problems.
Thank you so much guys
Now, I swear to god the amp got hairier. Not noisier, but it's like crunchier, and there's some distortion as a chord rings out. Not something I'm going to get involved in tonight but maybe I'll try some different guitars and speakers and tweak with the bias a bit.
again, thank you.
-
Nate - are the V1b and V2b triodes sharing a cathode resistor? The DRRI uses a shared 820 ohm cathode resistor for those two triodes just like Leo did. However, if you have a separate cathode resistor for each triode, they need to be 1.5K to get vintage levels of gain.
Only mention it because it came up in another recent discussion about converting DRRI amps.
Cheers,
Chip
-
separate 1.5k's a la the hoffman layout.
-
separate 1.5k's a la the hoffman layout.
Time to check resistor values! 3.3 meg reverb mix resistor, preamp plate resistors, cathode resistors, etc. The bad part is that sometimes it's really hard to figure out what your DMM should show for resistance in a circuit. Look at color bands first. Tedious? Sure! But it's a brute force way to make sure everything's what it's supposed to be.
Chip
-
Ok, got some changes to report, and a few more questions.
I put in a 50w speaker instead of the 25w I had, definitely an improvement. Then cooled of the bias a bit more, that really cleaned up the hairiness. The normal channel sounds just right now. Although I noticed one tube is reading 21v and the other is 14.5v, is that a problem? It sounds great so I'm not super worried.
The reverb channel is still too gainy/crunchy for my tastes. Not bad sounding, but I can't get a good clean sound at a volume that would work with a band.
I checked every resistor and cap. Everything's where it should be. Tomorrow I'll double check the wiring.
I'm thinking I should try some different tubes. Right now there's tung-sol's in the preamps, jj ecc81's in the verb diver and PI, EH 12ax7's in the other slots. I don't have any spare 12at7's on hand so I'll have to order some.
-
Don't understand what you mean by tube voltages above.
I've found a couple of really hot Tung Sol 12AX7s - fun in the right place but maybe you've got one where you don't want it.
Chip
-
I noticed one tube is reading 21v and the other is 14.5v
my mistake I meant miliamps, not volts, for the bias readings.
-
I noticed one tube is reading 21v and the other is 14.5v
my mistake I meant miliamps, not volts, for the bias readings.
Even 21ma is running a bit cool IMHO for a 6V6. 14.5 is really cold. Neither channel should be "gainy/crunchy", especially with the bias set that cold.
I'm not a fan of tube matching, but you've got a 30% differential there. Have you tried swapping the power tube spots to see if the difference is the tubes or something else?
Chip
-
Good call, I didn't think of that. They're new Sovtek tubes from the tubestore that we're sold as a matched pair. Not something I would usually care about but the difference seems unusual. I'll swap the tubes around tonight and report back.
-
Swapped the tubes, the bias followed the tubes. I don't have any other 6l6's around but I do have some 6v6's I can try, just to see what kind of numbers show up.
-
Swapped the tubes, the bias followed the tubes. I don't have any other 6l6's around but I do have some 6v6's I can try, just to see what kind of numbers show up.
Nate - those cathode current measures were really low for 6V6s. No wonder the amp sounds funky with 6L6s!
6L6 data sheet: http://www.mif.pg.gda.pl/homepages/frank/sheets/093/6/6L6GC.pdf
Cathode current * Plate voltage = (70% * 30 Watts)
Assuming plate voltage of 400: 0.021 volts * 400 = 8.4 watts
(70% * 30 watts) / Plate Voltage = cathode current goal
21 / 400 = 52.5ma (just an example)
Your goal for cathode current is a function of plate voltage, and that will change every time you adjust the bias. My guess is that you started chasing it and just went the wrong direction. Easy enough to do! Either your bias voltage is way to high (in absolute value; IOW bigger negative number) or something else is wrong.
Also, you can't just plug 6V6s in there. The maximum plate dissipation drops to 12 watts (or arguably 14 watts). The lower current draw of the tubes will let B+ jump upward. Whole new set of calculations.
This may help: http://www.geofex.com/tubeampfaq/taffram.htm
Cheers,
Chip
-
Thanks for the heads up.
I started somewhere around 35ma, and brought it down from there. There was another thread about getting a deeper tremelo, some one was talking about reducing the bias to 17ma working for them, I was futzing around with that.
I added the mid pot to the reverb channel, that cleared up my crunch issues.
Tomorrow I'll bring the bias up to the normal range, and see how things sound.
-
Thanks for the heads up.
I started somewhere around 35ma, and brought it down from there. There was another thread about getting a deeper tremelo, some one was talking about reducing the bias to 17ma working for them, I was futzing around with that.
I added the mid pot to the reverb channel, that cleared up my crunch issues.
Tomorrow I'll bring the bias up to the normal range, and see how things sound.
I'll bet you grounded the Vibrato channel's tone stack for the first time when you added the Mid pot. It may be compulsive of me, but one of the last things I do before first power-up is to go through the schematic and check every single ground connection with the DMM. It won't necessarily turn up a cold solder joint, but it's saved me some time more than once.
The discussion about reducing bias to 17ma for the tremolo had to be in the context of 6V6 power tubes. My AB763 build with 6L6s settled in happily with 43-48ma on the power tube cathodes plus 470 volts on the plates +/- and there's more than enough tremolo. (Voltages HERE (http://www.el34world.com/forms/valvedata/ChipAB763.htm)) I DID make a couple of the changes recommended in the Archives for the bias vary tremolo with 6L6s though.
Chip
-
Thanks for the chart, I'll make my own for comparison.
I have some bad news (for me). I brought my amp over to a friends studio to compare my amp to his stock DRRI. The results were really weird...
I suppose I should focus on the negative...
-My amp was much quieter than the original, which is really weird since I added 6l6's and the mojo OT for super/vibro/bassman
-My amp is still pretty dirty sounding in comparison, maybe it's a taste thing, doesn't seem right.
-My amp has a low frequency problem, I can't turn the bass up on either channel (especially the reverb) past 3. It just get's really dark, muddy, and distorted.
It's not like the amp is horrible, it makes some nice sounds. I was pretty pleased with myself, it's just that in comparison there's definitely some issues. We tried another set of tubes from a bandmaster he had, tried my amp into his deluxe's speaker as well as the 212 bandmaster cab. Those experiments changed the tone a bit but not the nature of the problems.
I've gone over the circuit a skillion times and I'll do it a skillion more, but I must be overlooking the same things every time. So maybe if I take a step back, I'm using the ab763 board, with two 6l6's, and super reverb transformers, into a 50w weber 12, in a deluxe cab, any red flags in that set up?
thanks for your time!
-
any red flags in that set up?
No, just the old Hoffman axiom..... "If were wired right, it would be working"
Double check every resistor (like a 470k where a 470R should be), have you swapped the OT primary wires(?) (disconnect the NFB resistor on one end....any improvement?) You're going to have to divide & conquer
Maybe with some good hi-res pics of the guts, someone here might be able to spot something.....
-
Here's a link to my voltage chart
http://www.el34world.com/forms/valvedata/looks%20ok.htm
also I have 429v coming off the rectifier
went through everything again, sadly I found nothing. Playing the amp at home it didn't seem as bassy, but it's still really dirty. Which isn't all bad, it's a really cranked rock out sound at reasonable volumes but it's also not what I was going for.
I'm going to take a break for a while, then I'll go through and check for continuity at every connection, and take some photo's.
(disconnect the NFB resistor on one end....any improvement?) You're going to have to divide & conquer
I have the nfb on a switch right now taking it out is a cool effect for more distortion but doesn't help anything.
-
Not much has changed but since I've tried a few things and made some observations I thought I'd post ,maybe something will click with some one.
-I've added the mid pot in place of the #2 jack, if I set it so that I'm measuring 6.8k from wiper to ground that's when the amp is really dirty, if I back it off to like 1.5k or less the amp cleans up. This is all with the volume around 3-4, it's a decent volume for playing at home but that's about it.
-I swapped out the 6l6's for 6v6's and rebiased, the bias difference between the tubes was less extreme (one at 17ma one at 16ma). The volume of the amp is about the same, still really early break up though. However the noise floor of the amp is much quieter. I decided to try the swap based on...
Q: You said in your book The Ultimate Tone that you can put 6V6s in a Twin. Are you crazy?
A: No, I'm not crazy. If you carefully study the conditions for this substitution, you will realize how safe this swap is. In TUT our advice was that if the plate voltage is less that 450V, the 6V6 would be okay. We revised this in TUT3, to a limit of 500V, since the 6V6 is rated for 1200V environments in TV circuits. The flash-over point for the tube is just above this.
The 'V' has just over half the transconductance of the 'L', so it tends to draw a proportionate idle current, and yield about half as much power. Any 80W or 100W Twin will easily accept 6V6s. Note that the Fender Deluxe Reverb has a V+ of 420V and bias level of -37V, the same as most 100W Twins.
I'm confused about the -37 bias level so I stuck with the 17ma I've seen elsewhere, at least that number makes sense with the bias equations I'm familiar with.
-
I'm confused about the -37 bias level so I stuck with the 17ma I've seen elsewhere, at least that number makes sense with the bias equations I'm familiar with.
-37 is the bias voltage measured on pin 5 of each power tube.
I don't know where the 17ma figure comes from but it does not compute for me. Conservatively:
6V6 max plate dissipation = 12 Watts
70% = 8.4 watts
Plate voltage * cathode current = 8.4 watts
440 * .019ma = 8.4
Now, assume 14 Watts rating on a 6V6:
http://www.mif.pg.gda.pl/homepages/frank/sheets/135/6/6V6GTA.pdf
14 watts * 70% = 9.8 watts plate dissipation
Add one watt for the screen grid - its current is part of your total cathode current and max dissipation is 2 watts => 10.9 watts
440 plate voltage * 24.7ma = 10.9 watts
You are measuring millivolts dropping across a 1 ohm resistor on the cathode. In this special case millivolts = milliamps. Ohms Law:
I = V / R
Sorry for beating a dead horse, but this amp seems to be biased too cold to me - both with the 6L6s earlier and now with the 6V6s. That's math, although my 24.7ma for 6V6s might be considered a little bit aggressive since I adjusted for screen grid current too. However, that's also about the same plate dissipation that sounded good to me (and some other guys) in my Princeton Reverb builds.
Have you double checked the value of each of the pots in the Vibrato channel tone stack? Just a wild guess, but your voltages look OK. Something has to be off though. Have you measured the value of the 3.3meg reverb mixing resistor?
Argh! Wish I could help here,
Chip
-
it came from this...
http://www.diyguitarist.com/Images/BiasChart-6V6.jpg
But your right, I rebiased for 25ma. It sounds better but that's not what's causing the problems.
Argh! Wish I could help here,
You are, A LOT! Even if you don't have the magic answer, just being able to get this stuff out of my head and bounce around ideas with some one is a huge help. I really think I have the board populated correctly, I've been over it so many times, I think the issue might be in something I'm miss understanding with the wiring. Being that the tone stack seems to have a big effect on the distortion levels I did go over the pots, I can't see the designations so I measured across the outside lugs. Everything looked ok.
I have another friend with another studio who has a SF deluxe and SF bassman, Is there a danger in pulling out the chassis from say his bassman and jumpering the output from my PI to his power amp section and vice versa to try and isolate which sections are problematic?
-
jumpering the output from my PI to his power amp section and vice versa
I would never do this. But you can open the SF Deluxe and get an idea about how to wire a Fender amp to get it working flawlessly. OK, you can find it can appear as a mess, but these amps are working for years. You didn't gave us pics of your chassis's guts, so if you are SURE about components placement and value, tubes characteristics, tubes sockets, voltage, pots and inputs wiring, it's becoming harder and harder to help you further... you got very good advices from other forum members; I cannot see anything to add to help you, unfortunately :sad:
Hope you'll get it soon!
BR,
Danskman
-
RE your voltage chart. V4-7 voltage can't be right. V4-6 voltage doesn't look right either. Recheck both of those. Also check that the plate resistor connected to V4-6 is really 100K. I'm guessing that V4-3 and V4-8 are tied together and have a 820 ohm resistor to ground?
At this point, we really need to see some hi rez pics of your build. We need to be able to read component values and trace wiring. Your exact schematic and/or layout would be helpful too.
Jumpering the preamp of one amp to the power amp of another amp can be very useful. A better way to do this would be to jump the preamp to the INPUT CAP for the PI using shielded wire. Be careful and understand what you're doing.
-
Thank you.
I cannot see anything to add to help you, unfortunately sad
It's not unfortunate, really, I'm totally grateful just to be working through this as a conversation. I'm trying to borrow a better camera, mine is so so. I don't really have much time the next few days. But I'll confirm Sluckey's observations and get some photo's up as soon as I can
Your exact schematic and/or layout would be helpful too.
The hoffman ab763 layout.
-
RE your voltage chart. V4-7 voltage can't be right. V4-6 voltage doesn't look right either. Recheck both of those
My mistake, v4-6 is 268v and v4-7 is .1 mv.
Also check that the plate resistor connected to V4-6 is really 100K.
Check
I'm guessing that V4-3 and V4-8 are tied together and have a 820 ohm resistor to ground?
No, they have separate 1.5k resistors to ground, bypassed by a 22uf cap
At this point, we really need to see some hi rez pics of your build
http://picasaweb.google.com/109914091355686420139/Ab763#
The pictures starts at the normal channel looking at the front of the chassis and move to the right. Then the board is shot from the back of the chassis starting with the normal channel and moving left towards the bias section. Then it's the tubes in numerical order. I don't know if this covers everything well enough or not. If there's something specific any one would like to see, just ask.
Thanks for taking the time to help me with this!
-
thought I had it...
I got some new 12at7's (one of them didn't even work) from banjoworld, put a mesa 12at7 in the PI things sounded clean, it was late, my girlfriend was annoyed so I figured I'd live to fight another day. My replacement 6l6 from thetubestore showed up. So, I put those in, rebiased and plugged in, sounds like a distortion pedal again. I switched back to 6v6's and I've tried every tube I have in every position.
I got nothing.
I think I'll pick up another new 12at7 and see what happens. After that I'm in the mindset to tear it all down and start over.
bah-humbug
-
Dear Mr. Scrooge,
Please do not give up! You need to figure out what is wrong with this build before tackling another build.
Divide and Conquer! You may end up re-building the amp one section at a time, but that's OK. This is NOT a trivial circuit.
Figure out what IS working first. Disconnect the tremolo from the bias supply. Disconnect the negative feedback. Pull the reverb driver tube, disconnect the reverb tank, and ground the grid of the reverb recovery triode.
Do you have another tube amp? Find 4 12AX7s that you KNOW work and sound ok. Grab a set of 6L6s (preferably Svetlana Winged C, but whatever) because that's what you want in the amp, right? The voltages looked OK in your chart, except I still think you're biasing it too cold (cathode ma * plate voltage = plate dissipation with target of 21-24 Watts). Find a 12AT7 that works in the PI position.
What works now?
I didn't want to be critical, but your photos show room for improvement in soldering technique. You can see that several caps have been melted by the side of the soldering iron. Some places the tops of the turrets don't look like there's enough solder. There's a section on soldering in the "References" sticky. Look at Hoffman's board building instruction pages: http://www.el34world.com/boardmaker/BOARD3.htm
I am SURE that I re-soldered every single solder joint in my first Princeton Reverb build. Just because it shows continuity doesn't mean it's a good solder joint. De-soldering braid is your friend - get some and use it. Also, carbon comp resistors are more sensitive to over-heating than metal film. When in doubt, take it out of the circuit (one leg anyway) and measure the resistor's value.
Finally, I still think that jumble of high current wires around the impedance selector shouldn't be anywhere near the reverb jacks and especially the reverb recovery triode.
Sorry if this post is too critical. It's late and I'm tempted to delete the whole thing, but maybe there's something useful here.
Chip
-
Sorry if this post is too critical. It's late and I'm tempted to delete the whole thing, but maybe there's something useful here.
Help is help, it's not like you're calling me a jerk, if something can be done better it's a fact, and a result of my own actions. I'm cool with that.
Figure out what IS working first. Disconnect the tremolo from the bias supply. Disconnect the negative feedback. Pull the reverb driver tube, disconnect the reverb tank, and ground the grid of the reverb recovery triode.
will do. NFB is fine, I'll report back once I've tested the other stuff. I got roped into helping a friend re-model his kitchen this weekend, so I' don't think I'll have much time.
-
Hi Nate, first understand that I know maybe 1% of what most of these guys do. But I do see problems with the soldering also, which might be a big part of your problem? Be sure that you have the iron against the lead and the turret at the same time, and then when you have the joint, remove the iron and the roll of solder at the same time. Should give you a nice peaked, shiny solder joint. Hope this helps! :smiley:
-
After back to back weekends of weddings and house guests I finally got some time to sit down and solder tonight. I revisited every turret on the board, nothing changed. I'm guessing you kind folks have done about all you can for me with the information available. The onlly way I can think of to proceed is to start isolating the different stages by jumpering my pre-amp stages to a different power amp and/or vice versa to try and determine which section isn't working. I brought this up earlier and I think there was enough support for the idea that it's not out of the question. Originally I thought I'd borrow a friends SF bassman, but I really don't want to bitch up some one else's amp. There's an unused PA power amp at my practice space, what issues might there be in wiring that up? I'd have to double check, but I assume that power amp is expecting to see a line level signal at 600 ohms.
-
You really don't need to chop up either amp. You can build a very simple box that let's you listen to your amp's signal at different points in the circuit. Scroll down to the bottom here for Hoffman's listening amp box:
http://www.el34world.com/Hoffman/tools.htm
Here's my take on the listening box:
http://www.el34world.com/Forum/index.php?topic=3950.0
I just plug it into a small, solid state amp I don't use anymore and keep in the shop. Make sure you always turn the volume down to "0" on the box before turning things on though!
Here are Hoffman's tips on tracking down weird noises:
http://www.el34world.com/charts/fenderservice6.htm
Hope that helps some,
Chip
-
Awesome, that's really helpful just when I thought I'd explored every corner of the hoffman site. Your box is super slick, that's a really intelligent idea! I was figuring I'd have to buy a SF bassman or something, do my futzing then sell it once I got my amp working.
-
just when I thought I'd explored every corner of the hoffman site
There's a ton of great stuff tucked away in little corners here and there. I didn't find the best part of the "Board Manufacturing instructions" for at least a year or two: http://www.el34world.com/boardmaker/BOARD2.htm
Don't neglect the "Archives" either. For example, everything you ever wanted to know about fine tuning bias vary tremolo is in there... somewhere.
Just a couple of examples,
Chip
-
Alrighty, finally got around to setting up the hoffman/fresh_start listening box. I really can't thank you enough! It was a very interesting experience tracing the signal through the amp. Here's what I found...
Normal Channel
Signal sounds great all through the preamp, including the mixing resistor and coupling cap to the PI.
Vibrato Channel
Lot's of weird issues but let's focus the other stuff first.
Phase Inverter
Pin 1 clear with some background crackle, kind of like a crappy cable or input jack, really not too bad at all though.
Pin 2 clear
Pin 3 distortion
Pin 6 like pin 1 but louder, meaning both the signal and the crackle are louder.
Pin 7 distortion, an overall more quiet signal than pin 3/8
1st Power Tube
Pin 3 very distorted
Pin 4 very distorted
Pin 5/6 clean with minimal crackle
2nd Power Tube
Pin 3 very distorted but much louder than pin 3 on the first tube with some clean signal mixed in
Pin 4 very distorted
Pin 5/6 clean with minimal crackle
So, I'm a little confused about the distortion around the PI but since there's a clean signal going into the power tubes maybe that's normal? Seems like the problem is in the power amp section, first order of business is obviously going over that section with a fine tooth comb. Which I'll do tomorrow, but in the mean time does anything jump out at any one? Any voltages I should be revisiting?
If the parts placement and wiring are all ok that would leave a faulty or damaged component, seems difficult to track down, any pointers?
-
Once again, I can't find anything out of the ordinary. Looking over the bias circuits page in the library I noticed that the hoffman board smooths the - bias voltage after the bias pot while the typical fender circuit smooths it before the pot. I can't imagine this is causing any problems.
I'm thinking I want to disconnect the bias from the intensity pot and got straight to the 220k resistors. Maybe even pull all the tubes from the vibrato channel. I hate to be a wuss, but I wanted to make sure this doesn't seem dangerous to any one.
-
Nate - what signal are you injecting at the input? Keep the gain (Volume) low and match the signal voltage coming from your guitar. Try 1KHz, 50KHz or 100Khz.
Start signal tracing at the input jack and trace the signal path. Do not "listen" to points where DC is present (i.e. cathodes & plates). Listen to input grids, between the preceding coupling cap and the tube's grid. I don't think you should try "listening" anywhere past the power tube's grids. IOW stay away from power tube plates // OT primaries. There are really big voltage & current swings there you don't want to mess with.
I can't tell what's going on at the PI from your description. To check the input to the PI, test at the junction of the two 220K resistors and the 0.001uf coupling cap. There are caps on both sides, so no DC.
For the power tube inputs (also PI output), check on pin 5. That's the control or input grid. Forget pin 4 (screen grid) and pin 8 (cathode). Avoid pin 4 (plate) for reasons stated above. If the signal is clean on pin 5 of each power tube you should be OK.
"Crackle" sounds like there's a problem. Bad tube, bad solder joint, component burned up by too much heat with soldering iron. Did you use a heat sink when you were soldering in your components, especially caps? See the "References" sticky and go through the material on soldering.
Soldering seems to be like shooting a gun - every American man thinks he's born with the ability and knowledge needed to solder (and shoot) well. I have learned from experience that it is not true in either arena. Most of us have to study, learn and practice to perform either task well.
To de-bug the Vibrato channel, you don't have to bypass the Intensity pot. Just disconnect the lead from the tremolo circuit to the pot. (I don't connect that lead until initial testing is done.) Also, ground the grid (pin 2?) of the reverb recovery triode. That's just what the reverb pedal does. Now you're listening to the Vibrato channel with no reverb & no tremolo. Fix those later.
Sorry but I don't have time to re-read this entire thread. Are you using "known good" tubes? IOW do they work properly in another amp?
I strongly suggest that you start over again with matching the schematic to the layout with a highlighter, checking every single component and connection in your amp one-by-one. Yes, it's tedious.
Use your multimeter to systematically verify continuity at EVERY point in the circuit which should be grounded with the negative lead attached to the safety ground. If the resistance is more than a fraction of an ohm, there's a grounding problem.
BTW if the Normal channel preamp sounds clean, what's the sound coming out of the speaker(s) like?
HTH
Chip
-
Nate - what signal are you injecting at the input? Keep the gain (Volume) low and match the signal voltage coming from your guitar. Try 1KHz, 50KHz or 100Khz.
I used a looper pedal strumming an open E
If the signal is clean on pin 5 of each power tube you should be OK.
When using the normal channel it is.
To de-bug the Vibrato channel, you don't have to bypass the Intensity pot. Just disconnect the lead from the tremolo circuit to the pot.
To clarify, in the Hoffman AB763 old diagram that would be labeled "10". Or in other words, the wire that connects one of the side terminals of the intensity pot to that last .1 cap in the trem circuit.
Sorry but I don't have time to re-read this entire thread. Are you using "known good" tubes? IOW do they work properly in another amp?
My god, that would be absurd, especially on Labor Day. I can verify the 12ax7's, but right now I don't have access to an amp that uses 12at7's or 6v6's. I realize that's an important thing to be absolutely certain about, I'll work on it.
BTW if the Normal channel preamp sounds clean, what's the sound coming out of the speaker(s) like?
Well, when using the listening box, clean all the way up through pin 5 on either power tube.
Here's where it get's "interesting"...
Normally, when using either channel the output is very distorted and quieter than I know a medium powered fender should be.
But...
I went ahead with the a fore mentioned plan, that is using the normal channel and hooking the bias directly up to the 220k resistors. Everything sounds great. Clean, clear, loud, and very very un-noisy.
So, it seem's the next step is verifying all the tubes to eliminate that variable. Then hook the bias back up through the intensity pot but otherwise disconnect the trem and reverb. I'll do that then do a detailed report on the signal trace through the vib channel.
Thank you!
-
I took my tubes over to a freinds to verify them in his DRRI, all is good.
I disconnected the tremelo and shorted the reverb, then set up my listening box situation again, I think that there's more than one problem to sort out on the vib channel but the first is the preamp stage.
Listening at pin 7, and all the lugs on the various tone pots I have a nice clear signal. But at pin 2 of the reverb driver (or any of the caps in between) there's a problem. I would describe it as a sort of clipping, meaning I hear the strum of the chord from the looper pedal starting nice and clear and then it abruptly distorts. Sort of like "strummmmBZZZ!!"
All I can really think to do is replace everything between that side of the preamp tube and the input of the reverb driver.
Thoughts?
-
I had a bad solder joint have the same effect. It looked good but it wasn't, not even sure which one it was I just reflowed them all on 1 tube and problem solved.
Tony
-
A few posts back I resoldered the entire board, but I didn't think to do the tubes themselves. I'll definitely do that before messing around with components.
Thanks for the tip!
-
A few posts back I resoldered the entire board, but I didn't think to do the tubes themselves. I'll definitely do that before messing around with components.
Thanks for the tip!
Look really carefully at each tab on the tube sockets too. I had a tiny whisker of solder - almost invisible - producing an intermittent short on a tube socket that drove me crazy for a couple of hours...
Chip
-
further down the rabbit hole...
I reslodered all the tube pins, and the lugs on the pots just to be sure. No Improvement, but I really really studied the schematic and layout and traced the signal coming off of the first half of the vib channels preamp tube. It sounds great throughout the tone stack, all the way up to the side lug on the volume pot that connects to the wiper of the treble pot.
But then...
Listening carefully on the volume wiper I could here the clipped signal (I only had the volume at 2-3) as I turned up the volume I could hear the clipping very clearly. Here's where I get confused, I disconnected the wire going from the wiper to the grid and suddenly I have a very nice loud clear signal at the wiper.
Seems the next step is to replace the wire and verify the signal on the other end before connecting it to the tube. Maybe the wire got nicked or something. Despite needing to help a friend hang a set of french doors I should be able to get to that today. It sure would be nice to get at least one problem ironed out this weekend.
thanks again for everything!
-
messed around some more with the volume pot wiper/pin 7 connection, replaced the wire, no change. It's really weird (to me anyways) that when listening to the signal at the wiper with the lead disconnected from the tube it's clear but as soon as I connect it to the tube there's clipping.
It doesn't seem like there's a whole lot going on in the circuit at that point...
-I'm 100% sure the tube is fine.
-I rechecked the voltages at all of the pins, everything looks good. Meaning the two halves of the 12ax7 match and I checked the measurements against a few other members charts
-Unless I'm really really failing to read some component values, I'm positive things are set up correctly
-I think I'll pull the tube and check for continuity in the tube socket. Maybe it got damaged or was faulty, not outside the realm of possibility.
-My only other idea at the moment is to disconnect pin 6 just to try and isolate things further.
-
Any chance you mixed up pins when wiring the tube socket? Lord knows I've done it!
Chip
-
Hey Chip, how's it going? I really think the wiring is correct. I mean, obviously something is wrong and nothing is off the table as far as potential mistakes.
Disconnecting pin 6 didn't get me anywhere. If anything things sound weirder without it hooked up. What do you think about jumpering around in the amp? IE sending the signal from the normal channel over to the reverb section and vice versa. I'm just trying to think of different ways to isolate problems.
-
Ugggh
I tried an interesting experiment tonight, I connected the output of the normal preamp to the coupling cap that goes to the reverb circuit. I had good sound going into the reverb driver (where as before it was clipping right out of the vib channel preamp) but I had clipping coming out of the reverb driver. Which has got to be a separate problem than the first.
I reconnected the vib channel pre but this time connected pin 8 to the cathode resistor that normally goes with the normal channel (of course disconnecting it from the norm channel's tube). No change, same old crap.
I am way burnt out on this. I think I need to start over. I feel motivated to find the problem with the vib channel pre, but knowing that there's another problem just like at the next tube is pretty depressing. Also, the overall construction and lead dress was less than awesome to begin with and having been messing with it, everything feels pretty wild and messy.
I'll let it sit for a awhile, maybe I'll have some new ideas tomorrow.
-
Walk away from it for a week. Play guitar, go sailing/hiking/biking, whatever.
You don't want to hear this, but get a new copy of your schematic. Assume NOTHING. Start at the input jack and highlight each and every component and connection on your way to the speaker jack. Use your DMM to verify, for example, that the Volume pot wiper goes to the next tube's grid pin. On your way, test the value of every component that you can. If you can't get an accurate reading with the component in the circuit, lift one leg and get an accurate reading. You may have fried a component, more likely with a cap than a resistor, while soldering and/or re-soldering.
Did I mention ASSUME NOTHING? Verify the pin numbers on each and every tube socket.
OH, and get a roll of desoldering braid. Use it to get rid of as much old solder as possible before re-flowing any solder joints. Adding new solder to an already bad joint often doesn't help at all in my experience.
You won't figure out what is wrong if you actually start over.
Remember Hoffman's Rule: if it were wired properly, it would work properly (or something like that).
I know this already has been a painfully frustrating experience. It's a complex circuit and there's a LOT that can go wrong. BTW did you use color-coded wiring or is everything "code blue" (or yellow, or whatever)? I'm sure the old hands think I'm compulsive, but I know that blue wires are plates, yellow are cathodes, orange (or brown) are grids, red is B+, etc. It helps me a lot.
Cheers & good luck!
Chip
-
Insert Quote
Walk away from it for a week. Play guitar, go sailing/hiking/biking, whatever.
Yeah, you're so right, I can't remember the last time I sat around for a few hours playing guitar on a saturday.