Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: wittyjeff on February 18, 2021, 06:44:15 pm
-
I'm finishing up an AC30 type build that incorporates 3 channels: 1. A single-tube EF86 Channel with a Triode/Pentode Switch, a "Bass Mode" Switch (taken from VOX's recent Handwired model), and a Brilliance Switch; 2. A "Top Boost" Channel with a 12AX7 Parallel'd and a cascaded second 12AX7, including Treble and Bass tone controls, Vol, and a Master Volume; and 3. A Vibrato Channel (based on one drawn up by Mr. Luckey). I also added a post Phase Inverter Master Volume (also keeping the typical "Top Cut" control). I based the power section and power supply roughly on Kevin O'Connor's write-up in T.U.T. 3. I added a Triode/Pentode switch for the power tubes in the hopes that I could lower the output volume (bedroom practice vs gigging) with a switch. I've also added a relay based switching system to facilitate switching/blending the channels and controlling the main "voicing" switches from an 8-switch footswitch.
Here's where I'm at (and issues):
1. The EF86 Channel and all three of its switches work fine but there's a huge difference in gain/vol when switching between the Triode and Pentode modes of operation using the "Mode" switch. Is that normal? (obviously this is my first time running an EF86 and doing Triode/Pentode switching).
2. The "Top Boost" Channel works but my bass and treble controls are acting weird. Maybe I got something wrong in the schematic or the way I wired them up? Turning the Bass control does seem to be affecting the tone, but not to the degree I'm familiar with on other amps. The Treble, kinda the same thing but at 80% of turning it up it suddenly cuts the treble instead of "adding" more.
3. When blending (i.e. engaging both channels simultaneously with the same input signal) the first two channels (EF86 + "Top Boost") it results in a weird anemic sound (slightly less volume). In other words, it seems that they are out of phase..? I'm assuming this problem results from either the way I've connected them together or that I'm not understanding how each gain stage is inverting the signal phase. Thoughts?
4. My third channel, the "Vibrato" channel is barely audible even with all Volumes turned all the way up. I'm thinking it could be that: a. I got something wrong in copying down the design schematic, b. I copied the layout wrong, c. I got the first two correct, but wired something up wrong, or d. the way I have designed the connection to the phase invertor while also connecting a negative feedback loop to the same side of the phase invertor is shunting (or otherwise jacking with) my signal somehow. Thoughts on this one would be appreciated as well.
5. The "Power" Switch that changes the power tubes' operation between Triode and Pentode modes seems to be working, but I was expecting a large difference (in volume) between the two. It was billed as a "half-power" switch on a VOX "hand-wired" circuit I found from a few years ago (the model just prior to the current VOX AC30 Hand-Wired).
6. I was expecting more volume from this design than I'm currently getting but I've never worked with a 4-tube EL84 "30W" design before. Seems to me that my 2-tube EL84 "20W" "Marshall TMB" amp is nearly as loud. Shouldn't an AC30 design be a bit painful to the ear when turned all the way up?
6. The post phase invertor Master Volume, Top Cut Control, Switchable Negative Feedback, and the 8-switch relay board all seem to be working as expected.
I know this is a lot, but since some of the issues might be related to one or more of the others, I figured I'd throw it all out there for your valuable guidance and insight. I've attached the Schematic and the Layout I used. Thank you in advance for your help.
-
Very ambitious project! I can't look at that layout, so I'll comment referring to the schematic.
Here are a few things for you to fix.
1. Add cap between J8 and V7 grid to prevent FX loop from interfering with bootstrap bias on V7. (see pic)
2. Delete C57 and all of that NFB circuit. That's what's killing the VIB channel. (see pic)
3. R47 should be 1M. Not sure where you got 10K from.
4. You cannot blend the EF86 channel and the TB channel because the EF86 channel has only one inverter stage but the TB channel has two inverter stages. This means the signals will be out of phase and partially cancel each other, causing the anemic sound.
That's enough for now.
-
Steve, Thanks! My response to each of your recommendations:
1. Great catch! Thanks and will do.
2. Makes sense that your suggestion will solve the problem with the VIB channel... I kinda figured my ignorant attempt to add negative feedback would mess something up. I'll remove that Cap and then I'm thinking that if I switch the negative feedback to the off position it will essentially eliminate the Neg Feedback circuit I put in there (right?). If that fixes things, I'll take it out permanently. Is there another way to incorporate negative feedback elsewhere? It seems quite handy in cleaning things up when looking for certain sounds/tone (and for other instruments) so it'd be ideal if it could be done.
3. Again, excellent catch... I apparently just made a typo error when I drew up the schematic based on your AC30 model.
4. Agreed. I guess I don't understand which gains stages invert and which don't (as I suspected). Could I get the phases to match if I "un-parallel" that 12AX7 (V2) and use the extra side of that tube as an additional gain stage in the first channel after the EF86 (seems to need more gain anyway)...?
Thanks again for your essential input.
-
Half power is just -3dB, which is audible, but a very minor difference in volume.
To get what subjectively is percieved as half volume, you need to get down to about 1/10 power. I.e. 5W is about half as loud as 50W.
The volume difference between 20W and 30W should be barely audible.
To be able to crank a tube amp at ”bedroom” volumes, you need to get down to something like 0.1W (depending on the speaker).
About the EL86, small signal pentodes have way more gain than a regular triode. So the preamp pentode/switch should result in a huge difference in gain.
-
Assuming a decent speaker cab, I can’t envisage how a properly working 2xEL84 wouldn’t (in a domestic setting) be painfully loud, nevermind an AC30?
In scenarios like this, a sig gen, scope and dummy load are invaluable for facilitating the gathering of quantifiable data.
-
I'll remove that Cap and then I'm thinking that if I switch the negative feedback to the off position it will essentially eliminate the Neg Feedback circuit I put in there (right?).
That's correct.
Is there another way to incorporate negative feedback elsewhere?
Not many options in that amp. I don't consider NFB a part of the AC-30 sound. But if you must have it, your schematic shows the best way. You could always move the VIB channel to the other side of the PI and use mixing resistors to mix it in with the normal and TB channels. Which reminds me... you need mixing resistors (220K work fine) for the normal and TB channels. Your normal volume control will kill the TB channel when set to zero. Mixing resistors will minimize the affect of one channel on the other. And if you decide to move the Vib channel to the other side of the PI, you will need a third 220K mixing resistor.
I don't understand which gains stages invert
If the signal goes in on the grid and goes out on the plate, it will be inverted. But if the signal goes in the grid and goes out on the cathode (as in a cathode follower) there is no inversion.
As for jumping/blending two channels together, it's not important how many gain stages are involved. The important thing is the two channels must ***BOTH*** have either an even number of inversions or an odd number of inversions. For example, if #1 has 1 inversion then #2 can have 1, 3, 5, etc. inversions. But if #1 has 1 inversion then #2 cannot have 2, 4, 6, etc. inversions.
I'm gonna repeat in case it got lost in my rambling... YOU NEED TWO 220K MIXING RESISTORS ON THE OUTPUTS OF THE NORMAL AND TB CHANNELS.
-
IME the half power switching should be called: almost equally loud tone sucking switch.
If you want to tame the amp you'd better look into attenuators or VVR power scaling that B+. But since you already have a liftime supply of components packed in one amp the VVR is not recommended.
-
Also.. I wanna share this:
1/2 or 1/4 power switch... people expect same tone but half the volume. But for the ones who think about adding it anyway .. here's my description of how I think it sounds like:
We all know that part of the sweep on the volume pot between no sound and where it only starts to sound nice and thick.. wel it's that area you're winding up. A thin bland tone
-
Steve, thanks for the catch of the Mixing Resistors. Will add them right away. I feel kinda dumb not knowing that, but embarrassment is an effective learning tool I suppose.
I also will try moving the VIB Channel over to the other side of the PI and see what happens with the NFB just for kicks.
-
IME the half power switching should be called: almost equally loud tone sucking switch.
If you want to tame the amp you'd better look into attenuators or VVR power scaling that B+. But since you already have a liftime supply of components packed in one amp the VVR is not recommended.
Well said my friend. In a way, it appears then that the "1/2 Power" switch is working... it's just that they suck (both tone and otherwise). ...and you're right about the conflagration of components that is this amp. It's been a challenge to cram them in there (yes I know the downsides of component proximity and over-complicating things).
What are your thoughts about the idea that with split pairs of cathode resistor/capacitors one can pull a pair of power tubes to achieve "1/2 Power?" I've used Kevin O'Connor's design with this in mind.
-
Half power is just -3dB, which is audible, but a very minor difference in volume.
To get what subjectively is percieved as half volume, you need to get down to about 1/10 power. I.e. 5W is about half as loud as 50W.
The volume difference between 20W and 30W should be barely audible.
To be able to crank a tube amp at ”bedroom” volumes, you need to get down to something like 0.1W (depending on the speaker).
About the EL86, small signal pentodes have way more gain than a regular triode. So the preamp pentode/switch should result in a huge difference in gain.
d95err: Thanks. This explains what I'm perceiving. I just threw in that triode/pentode switch since VOX did at one point... I figured it did something useful and was hoping that it would reduce the volume more without jacking up the tone. Apparently that's not the case and it appears that section is working as it should... just my expectations were out of whack. I also feel better about the huge difference in the EL86, so thanks for that as well.
-
Another quick question... does signal phase matter if an input is on the other side of the Phase Invertor? For example if one channel in an inverted phase (i.e. one inverted gain stage) is on one side and another channel in a non-inverted phase (i.e. and even number of inverted gain stages) is connected to the other side of the Phase Invertor... will they still tend to interfere with each other? It seems like they would, but if they would not, it might simplify my idea to get these channels to cooperate with each other. I just want to be sure before I start tearing into this mess.
-
IME the half power switching should be called: almost equally loud tone sucking switch.
If you want to tame the amp you'd better look into attenuators or VVR power scaling that B+. But since you already have a liftime supply of components packed in one amp the VVR is not recommended.
Well said my friend. In a way, it appears then that the "1/2 Power" switch is working... it's just that they suck (both tone and otherwise). ...and you're right about the conflagration of components that is this amp. It's been a challenge to cram them in there (yes I know the downsides of component proximity and over-complicating things).
What are your thoughts about the idea that with split pairs of cathode resistor/capacitors one can pull a pair of power tubes to achieve "1/2 Power?" I've used Kevin O'Connor's design with this in mind.
Like mentioned earlier.. to get to a volume perceived as half power you'd have to get the power to 10%.. so 3W in your case. That's only possible with power scaling the B+ or a really good attenuator. The more logic way would be you buying earplugs for your family and really enjoy them full 30W !!!
I ignore my family when playing. 😃😃😃
-
does signal phase matter if an input is on the other side of the Phase Invertor?
NO
-
if you move that trem channel and use mixing resistors, i have found the nfb switch on a vox-ish pentode preamp 6v6 build to be super useful... while no nfb is part of the "vox sound" it can get a bit tubby in the bass when really cranked. nfb tightens that up while smoothing out the highs. makes for a pretty versatile amp in my opinion
-
I'm more a simple amp guy. Few features... reverb and tremolo + natural overdrive. I found out that whrn I have an amp that's supposedly versatile I end up using 1 setting of that amp 99% of the time.
In fact, I'm more a multiple amp kinda guy. 😄
-
One final issue... both the Treble and Bass controls are basically doing very little and certainly not responding as one would expect even though they do change the tone as they are turned, the tone changes make very little sense and don't seem to be linear. What I'm thinking happened is that the genius Master Volume control that I added (which actually is functioning as expected) might be messing up the way that the two tone controls are supposed to function. See the schematic section attached. I've circled what I've added, the rest is verbatim from the VOX AC30H2/AC30H2L Schematic I copied. I've re-checked my wiring and verified that what's supposed to be grounded is actually grounded and things are wired as in the schematic. Any thoughts on if that Volume control is indeed what's causing the aberrant behavior? If so, are there any good simple solutions that will still allow this Master Volume to work?
-
General update: After all of your useful suggestions, I have routed all three channels to the same side of the PI (with the mixing resistors now) and they work as intended and don't seem to be interacting with each other. Well, when patched together, they do interact but that's because one channel is out of phase with the other two (although it doesn't sound terrible to me)... and the trem/vib does sound different depending on which channel is patched together with the VIB channel. All good there in my opinion.
I left the NFB in place on the other side of the PI and it seems to be functioning. It was very subtle when I had the 47K resistor in there so I changed it to a 33K. Now I'm thinking it might be a bit too much. ...will have to play with it awhile to settle on which is better.
This amp really makes some great sounds and each channel is distinct and interesting. ...just two issues left. The tone issue I mentioned above, and the one in the next post.
-
Ok, one additional "final" issue: While the amp seems to be nearly functional and sounding great. It is still lacking volume in general. To quantify/verify, I used a cheap audio decibel meter and played the three amps I have on hand (one of which was this AC30) as loud as possible (neighbors, cat, and wife... you're welcome). My little Marshall 15W TMB (Hammond organ AO39 conversion) was actually the loudest at 98db. My 6L6 push-pull AB763 circuit amp (but kinda under-voltaged, long story) produced 96db. This "AC30" could only produce 95db no matter what settings I tried and nearly breaking the strings! I would think that it should produce at least a decibel or two more than my little beat-up two EL84 tube amp. ...right? Just to check to make sure all four EL84s in this amp were working, I pulled a pair and tested. That produced 92-93db... about 2 or 3 db less... which seems like what one would expect. Same with the other pair removed.
Is there something that could be attenuating the overall volume somehow? Maybe the safety resistor across the 16 Ohm speaker output is knocking out 3db? Shouldn't be, right? Any thoughts?
-
NFB is the most likely culprit.
-
NFB is the most likely culprit.
The NFB I have rigged up is switchable. When switched "off" it should be just going to ground and ignoring the NFB, correct? On the other hand, your suggestion totally makes sense and I trust your judgment. It's easy enough to disconnect entirely, which I will do at your suggestion. I may have something wired incorrectly and it's shunting power somehow. I'll report back as to the results.
-
Any thoughts on the tone control issue 5 posts up?
-
Never seen or heard a TB circuit. Don't know what to expect from it.
-
Do you think the way I have moved the "Master" volume control in the attached schematic would work? I'm thinking maybe this will fix my Treb & Bass Tone Control issues...?
-
Worth a try.
-
I was able to remove the NFB connection completely and also the 250 Ohm safety resistor just for kicks. It had no effect on the volume level whatsoever. I wonder, are my voltages in the correct range or are they on the low side? Not sure what else to consider regarding why this 30W design is putting out less than my 15W. Ideas?
-
> are my voltages in the correct range
Did you post them? Somewhere? Can't see your voltmeter from here.
-
A vox type tone stack has a wider than normal range of control, as the bass control introduces an extreme mid scoop when set to max.
Do you think the way I have moved the "Master" volume control in the attached schematic would work? I'm thinking maybe this will fix my Treb & Bass Tone Control issues...?
There's no reason why a master vol after the tone stack shouldn't work.
Also no reason it would affect your tone stack issue.
There's no obvious need for a potential divider between the tone stack and master vol.
With the build's complexity and layout (all those signal wires running to and fro - the relays would be better placed where they're needed rather than bunched together off to one side), perhaps the treble control issue is due to oscillation?
Perhaps the low power output issue is due to a mistake somewhere in all those power switching options, eg perhaps it never actually arranges all 4 EL84 in pentode mode?
Only the HT VDC measurements seem to be supplied in the schematic of post #1; they seem rather high for an AC30.
96dB - what speaker are you using, how far away from it are you?
Whatever, an actual 'VAC measurement across a resistive load' would be much better.
FYI whilst it's beneficial when using a directly heated rectifier, in the case of an indirectly heated rectifier such as a GZ34, the use of a CT on the 5V winding for the rectified DC output might act to add to the ripple. Take the DC output directly from pin 8 of the GZ34.
https://tubedata.altanatubes.com.br/sheets/010/g/GZ34.pdf
And to avoid ripple current contaminating the power section's star 0V point, Merlin points out that the HT winding's CT should connect directly to the reservoir cap terminal, see 15.3 of his document.
http://www.valvewizard.co.uk/Grounding.pdf
-
> are my voltages in the correct range
Did you post them? Somewhere? Can't see your voltmeter from here.
They are listed above the power supply section on Schematic attached in the first post. Here they are again for convenience: The "A" (where the first HT off the first filter cap usually is the OT positive supply) is 356~362, the "B" (Power Tube Screen Supply) is 350~356, the "C" (Phase Invertor Supply) is 267~271V, the "D" (Supply for Channels 1 & 2 preamp tubes) is 213~217, and the "E" (Supply for the Vibrato channel tubes) is 175~178V. I'll take some additional voltages directly from the appropriate pins of the various tubes and such later and post them.
-
With the build's complexity and layout (all those signal wires running to and fro - the relays would be better placed where they're needed rather than bunched together off to one side), perhaps the treble control issue is due to oscillation?
For sure. The relays were just added as an off-the-shelf relay board for "convenience" and mostly my lack of electronic skills... however, in retrospect, it would have been more convenient and wouldn't have involved any more or less "skills" to do as you said... put the relays closer to where they were needed. Although that eyelet board stretches the entire length of the chassis and is stuffed with more components than should be on there, so there would still be the issue of space. On future builds of this particular amp, I would definitely eliminate one of the channels and that would make enough room for a more efficient layout for sure.
Perhaps the low power output issue is due to a mistake somewhere in all those power switching options, eg perhaps it never actually arranges all 4 EL84 in pentode mode?
I'd be surprised if there wasn't a mistake somewhere in all of this mess. In fact, I'm surprised this thing works as well as it does! I did pull the power tubes in pairs and test (since they have separate bias) and the two pairs compared equally to each other. Is it possible for an individual tube within a pair to be in pentode vs triode mode? If it is, then one tube in each pair could be in the wrong mode... would push-pull even work in that case? Would it make sound? If so, this could easily be the case and I'll look into this possibility.
Only the HT VDC measurements seem to be supplied in the schematic of post #1; they seem rather high for an AC30.
Right, and sorry. I didn't report any other voltage measurements. I'll take some useful ones and report a bit later. If the HTs are high then it's likely that my suspicion of low voltage isn't the problem with the volume I'm guessing.
96dB - what speaker are you using, how far away from it are you?
Whatever, an actual 'VAC measurement across a resistive load' would be much better.
For useful quantitative diagnostic results... that for sure would be better. I just wanted to do a quicky measurement to compare it with my 2-EL84 tube 15W amp to see which was louder. Seems like the small 15W amp being 3db louder... means this new build isn't putting out what it should be capable of regardless.
FYI whilst it's beneficial when using a directly heated rectifier, in the case of an indirectly heated rectifier such as a GZ34, the use of a CT on the 5V winding for the rectified DC output might act to add to the ripple. Take the DC output directly from pin 8 of the GZ34.
https://tubedata.altanatubes.com.br/sheets/010/g/GZ34.pdf
Agreed and that's how I would usually do it. Kevin O'Conner claims this method has a "humdinger" effect and in his T.U.T.3 he lays it out exactly as I've constructed it. Thought I'd try it. The amp is really quiet so it appears that it's at least a functional option. Thoughts?
And to avoid ripple current contaminating the power section's star 0V point, Merlin points out that the HT winding's CT should connect directly to the reservoir cap terminal, see 15.3 of his document.
http://www.valvewizard.co.uk/Grounding.pdf
Agreed. Just brain-damage on my part when putting this to paper and soldering. I should have known to do that. Thanks for catching that for me. I'll just move that wire over to the first cap ground terminal. Still, this amp is quieter than most (except when my footswitch unit's LEDs are powered with a cheap 9V power supply... then there's some buzz).
-
...
Perhaps the low power output issue is due to a mistake somewhere in all those power switching options, eg perhaps it never actually arranges all 4 EL84 in pentode mode?
I'd be surprised if there wasn't a mistake somewhere in all of this mess. In fact, I'm surprised this thing works as well as it does! I did pull the power tubes in pairs and test (since they have separate bias) and the two pairs compared equally to each other. Is it possible for an individual tube within a pair to be in pentode vs triode mode? If it is, then one tube in each pair could be in the wrong mode... would push-pull even work in that case? Would it make sound? If so, this could easily be the case and I'll look into this possibility.
Yes, as long as they're vaguely functional, although some degree of non-linearity / assymmetry will be introduced, different valve types or configurations can function in push-pull. Also consider the Boogie 'simu-class' arrangement, in which the outer pair are wired as triodes, the inner pair as pentodes.
So yes, it would be a good idea to verify everything that will affect power output.
FYI whilst it's beneficial when using a directly heated rectifier, in the case of an indirectly heated rectifier such as a GZ34, the use of a CT on the 5V winding for the rectified DC output might act to add to the ripple. Take the DC output directly from pin 8 of the GZ34.
https://tubedata.altanatubes.com.br/sheets/010/g/GZ34.pdf
Agreed and that's how I would usually do it. Kevin O'Conner claims this method has a "humdinger" effect and in his T.U.T.3 he lays it out exactly as I've constructed it. Thought I'd try it. The amp is really quiet so it appears that it's at least a functional option. Thoughts?
...
To my shame, I never got around to buying any of the TUT series, so :dontknow:
I just can't see how with an indirectly heated rectifier, provided the DC is taken from pin8 rather than 2, there's any heater hum to ding :icon_biggrin:
Rather that as the cathode's only connection to its heater is at pin8, by introducing a connection to pin2 to the DC output, some degree of the heater voltage will be added to it, maybe 2.5VAC?
-
To my shame, I never got around to buying any of the TUT series, so :dontknow:
I just can't see how with an indirectly heated rectifier, provided the DC is taken from pin8 rather than 2, there's any heater hum to ding :icon_biggrin:
Rather that as the cathode's only connection to its heater is at pin8, by introducing a connection to pin2 to the DC output, some degree of the heater voltage will be added to it, maybe 2.5VAC?
:dontknow: I've attached the small paragraph that he uses to explain it. I just figured I'd try it just to see what happened. Again, seems to work.
-
Yes, as long as they're vaguely functional, although some degree of non-linearity / assymmetry will be introduced, different valve types or configurations can function in push-pull. Also consider the Boogie 'simu-class' arrangement, in which the outer pair are wired as triodes, the inner pair as pentodes.
So yes, it would be a good idea to verify everything that will affect power output.
Got it. This is a definite possibility and would account exactly for the symptoms I am observing. There easily could be a wiring issue with the "2nd" tube of each power pair. I'll dig into it and report back.
-
Here's a gutshot (finally) of what this mess looks like in the real world...
-
The Iron Maiden song 'Can I Play With Madness' comes to mind :icon_biggrin:
-
Tone Control Problem continued...
I bypassed the channel Master Volume to see if that somehow made a difference in the symptoms I have been experiencing with the Treb and Bass controls of the "Top Boost" channel. Bypassing made no change in the symptoms. To clarify, the symptoms are: Treble: It does seem to change the tone from full-but-muted-high-frequencies (acceptable) to very high frequencies accentuated (acceptable) but a thinner tone (seems odd compared to other amps... like it's "subtracting" some lower frequencies). Bass: doesn't seem to be affecting the tone at all until 8 to 10 where it abruptly attenuates the overall volume somewhat and sucks the "fullness" from the sound.
I've attached the following jpegs: 1. Schematic of how I constructed it noting my addition to the circuit I copied from VOX. 2. How I bypassed my MV yielding no improvement or change in symptoms. 3. Proposed change if I bypassed that voltage divider and eliminated the MV. and 4. A proposed schematic eliminating the voltage divider stuff and keeping a MV.
Any ideas why schematic #1 wouldn't work? Would schematic #4 solve the problem? I do know that the best answer is usually, "try it and see if it works or not." In this case, if there's a way to have some reasonable expectation of success it will be worth the pain in the butt to tear into this mess and redo this... so thought I'd ask the community before I dived in.
Thanks in advance for your help and advice.
-
Totally disconnect the other two channels so that you only have the TB channel connected to the PI. Any better?
-
Are the slope resistor R25, the 1M pots R27 and R28, and the treble cap C20, all definitely good?
That symptom set seems to require 2 things to be wonky :w2:
Though an open circuit mid cap (C19) can cause both treb and bass controls to act weird.
-
Tone control issue continued...
1. I have disconnected the other two channels completely and no change in symptoms result.
2. I have also tried bypassing that voltage divider section completely and that didn't result in any changes in the symptoms. I'll probably leave it like that unless anyone really knows what those two 220K resistors and the 22pF cap even do.
All of my components are brand new, decent quality, and purchased from a reputable source. ...not that that guarantees anything. I took a meter to the .022uF caps and they have the same resistance and capacitance (realizing that these readings might be skewed when done "in situ") the point being; I'm certain that the C19 and C21 caps are not open.
To restate the symptoms, let's say that the Treble pot is performing adequately, but the Bass pot seems to not be doing anything as it's turned until it is turned up past 80% of it's rotation where it suddenly attenuates the signal in general.
I'm kinda down to thinking that I have a bad pot (even though it's a brand new Alpha and I'm pretty careful to make sure that it's rotated all the way prior to soldering)... or something in the way I have it wired in the layout (see attachment to first original post) that I'm just not getting.
Thoughts?
-
rb29 and r30 are a voltage divider that attenuates half the signal. c29 bypasses r30 for frequencies above about 32khz
try tack soldering a bigger capacitor (10nf? 22nf? whatever) in parallel with c16. bass control work better now? if so, then you're not amplifying enough low frequencies to make that control useful and should play with with value of c16... 4n7? 2n2? who knows
-
The bass control is wired up back to front, ie the 0V connection needs swapping to the other track end.
The resolution will then be much improved, as the audio taper will be working as intended.
-
rb29 and r30 are a voltage divider that attenuates half the signal. c29 bypasses r30 for frequencies above about 32khz
try tack soldering a bigger capacitor (10nf? 22nf? whatever) in parallel with c16. bass control work better now? if so, then you're not amplifying enough low frequencies to make that control useful and should play with with value of c16... 4n7? 2n2? who knows
Thanks, that explains exactly what the voltage divider is/was for. Good to know on the C16 idea. It seems to have enough bass in general even if the bass control didn't seem to be doing much. Turns out I had it wired backwards (see below) so it was messing with my perceptions apparently. I'll mess with it awhile and see if that cap needs changing.
It does seem to be a bit louder (which is good). Will there be any drawbacks to just leaving out that voltage divider and cap?
-
The bass control is wired up back to front, is the 0V connection needs swapping to the other track end.
The resolution will then be much improved, as the audio taper will be working right.
Well... that must have been the dumb thing I figured I did. After re-wiring that pot as you suggested, it seems to make much more sense now. In fact, while it doesn't seem to have the same bass adjustment effect that I'm accustomed to on Fenders and Marshalls... there is now definitely attenuated bass when turned to the left and definitely more bass when turned to the right and seems linear enough between those two points. There's no longer an abrupt cut-off on either end... so I'm happy with it.
So, thanks a million! This amp now is fully functional despite the person who built it.
-
Conclusion... with all of your help, it now appears that this amplifier is fully functional. It really does sound glorious. Lot's of VOX chime, 3 distinct channels that can be blended together, a fantastic trem/vib system (thanks S. Luckey!) that blends superbly with the other channels, an 8-option footswitch unit, and some built in power reduction options. ...not too shabby.
This was a large and complex project for a newbie like myself. I began building amps for the very first time exactly one year ago (well reading some books on electronics and then later in the spring began amp building and then proceeding from there). This is my 7th build. If the COVID pandemic has been good for something, it's given me an opportunity to have the time to pursue the amp-building idea I had always wanted to try. I see now that it can be a challenging (i.e. sometimes frustrating) yet very rewarding endeavor (and rather expensive too).
Thanks again for all of your help. I'll post a pict soon when I get it mounted in it's beautiful cabinet.
-
if it were my amp, whether to keep or chuck the voltage divider would depend on how well the channels blend. if the top boost channel overpowers the others, then attenuating that signal might allow it to play nice with the others
since pdf figured it out for you, I'd leave that 500pf coupling cap as is to accentuate the top boost
-
if it were my amp, whether to keep or chuck the voltage divider would depend on how well the channels blend. if the top boost channel overpowers the others, then attenuating that signal might allow it to play nice with the others
Since my ears are not as tuned to the nuances as a seasoned guitarist's... the next stop for this amp is to be shipped to my cousin who was a touring guitarist and still does quite a bit of studio work (and weekend fun gigs of a wide variety of genres). Guitarists can really hear things that ordinary people can't. I'm sure he will come up with a list of things that need to be adjusted/tweek'd. That's why your suggestion is quite helpful since I'll know exactly what to do if he feels that channel is more overpowering than the others or needs more bass for example. Thanks again.
-
So here's the pict of what the amp looks like in it's "Ode to VOX" cabinet (it's just a re-covered cheap rack cabinet).
-
...
This was a large and complex project for a newbie like myself...
I suspect it would be a substantial, significant project for anyone, so well done, kudos to you!
Regarding the TB circuit, I suggest to remove the cathode bypass C18 from the CC section of the DCCF. Typically everything seems happier and more balanced without it.
-
I've finally gotten around to taking more appropriate supply voltage readings. I've attached an updated schematic to this post with the Voltage readings in purple next to the places where they were taken. In designing this amp, I just looked at several schematics and kinda guesstimated what the voltages should be based on the higher ones I saw. So... I'm assuming there's a good chance that they are not exactly what they should be, even though the amp seems to be working well. I will say the the Power Transformer does get and stay on the warm side (around 135 to <150F at the hottest point on the exterior top) and it's not due to radiated heat from the power tubes (I tested with a reflective heat shield in place). From what I read, the PT can handle that amount of heat and my cabinet is built to handle and dissipate it., but just thought I'd throw that out there so all the info is in the mix.
Any thoughts on what the voltages should be? Any real problems if they stay where they are at? Any problems/consequences/benefits if I lower them in general (i.e. I don't want to risk lowering the output volume or lose headroom, etc.)?
Thanks again for your help.
-
Here's what the back of the amp in the cabinet looks like. I put a reflective aluminum shield at the top where there is about a 1/2in air space between it and the tolex covered insides (kinda stupid to cover the inside with tolex, I know) plus the tolex under the shield is covered with metal foil tape and then stapled. I haven't smelled any hot tolex (yet) after leaving the amp on for two hours even though the hottest part of that shield is getting over 170F (although that's reading with an IR device which could be reading reflections from the tubes so who knows). It's got good airflow out the back and out the front vents, so should be able to stay under meltdown temps easy enough. With an open back it's pretty easy to just aim a fan toward it to help out on a hot day.
-
...any thoughts on the voltage question above?
-
Don't touch!
:l2:
-
The HT to the power stage seems a bit higher than ideal, 320V seems a happier place to me.
What’s the cathode voltage and cathode current?
And the HT to the vib/trem preamp seems bit low.
-
The HT to the power stage seems a bit higher than ideal, 320V seems a happier place to me.
I can knock it down a bit easily enough but it will pull down everything downstream a bit.
What’s the cathode voltage and cathode current?
I'm assuming you mean the cathodes off the power tubes... I will check and report back.
And the HT to the vib/trem preamp seems bit low.
This is the "E" HT tap on my schematic. From what I could see in some other schematics that had voltages listed (although I could only find two) they seemed to think that 165V was adequate. I'm running at 177V. The whole VIB channel and all the oscillators and such are rather complicated and I have no idea what voltages really should be. Steve Luckey built at least one of these (probably more) and I used his schematic for this channel, but he didn't include measured voltage readings on the one I used. He seems to think I'm within a reasonable range. If there's a benefit to increased voltage, I'm all for it and that particular voltage reducing resistor is easy to access and change. Just curious to know your thoughts and further details before I dive in.
-
Ok, I put my multimeter in series with one of the power tube cathodes and got a consistent current reading of 0.041~0.043 A. I got a voltage reading (as compared to chassis ground) of 10.8~10.82 VDC.
How does that look? Remember that these are tied to two sets of resistor/bypass cap units by pairs (so that one pair of power tubes can be pulled and the remaining pair will be biased correctly (so half values I believe).
Does this look ok? I'm not 100% sure what to expect for readings.
-
..any final thoughts on voltages before I consider this amp "finished and ready for testing?"
-
Your EL84 seem to be idling a bit cooler than I expected, given the HT voltage. But that's fine, it's not a concern, I suppose that the 1k screen grid resistors are helping out a bit there (what's the VDC across them?).
Here's a schematic with some voltages https://el34world.com/charts/Schematics/files/Vox/Vox_ac30volt_factory.jpg
Not sure how valid it is or what era it's from but suspect it's the late 60s, perhaps toward the end of the Birch-Stolec ownership; the VAC noted for the PT HT winding (180-0-180) must be in error.
Whatever, the voltages noted there for the Vib/Trem channel seem a bit more in line with what I'd expect than those of your amp. But if your amp is playing fine and you're not getting nasty early overdrive on that channel (?) then it's a moot point.
My concern was that as the vol control for that channel is right at the end of it, it can be susceptible to harsh overdrive if the input signal is much above 'vintage single coil' level, eg using my R8 Les Paul with stock vintage type 'Burstbuckers', when plugged into that channel on a typical old AC30, I have to use the 'low' input or turn the guitar vol down a bit.
-
I'm surprised you didn't use the same power supply as the AC-30.4 used. If you had, you would be able to easily change the voltage for any node (C, D, and E) by changing a single 22K resistor, without affecting the voltage of the other nodes. I suspect your tube voltages would be a bit different if you had followed the original schematic.
-
But if your amp is playing fine and you're not getting nasty early overdrive on that channel (?) then it's a moot point.
I haven't experienced any early overdrive on that channel, but I only have a cheap guitar with single coils to test with. I do notice that the VIB Channel is not as loud as the other two channels, but other than that it sounds ok. Sounds real nice when blended with the other channels though, so I'd say it's acceptable. On the other hand, those voltages in the schematic you sent make sense and maybe I should get mine just a bit closer to those and see what happens.
-
I'm surprised you didn't use the same power supply as the AC-30.4 used. If you had, you would be able to easily change the voltage for any node (C, D, and E) by changing a single 22K resistor, without affecting the voltage of the other nodes. I suspect your tube voltages would be a bit different if you had followed the original schematic.
A LOT of things would have been much simpler if I had followed your original schematic, that's for sure. In this case I wanted to implement some of the Kevin O'Conner ideas (distributed filtering for one) in the power supply just to see if there's a benefit to it. I will say that the amp is very quiet (low noise floor). In any future AC30 style builds, I'm definitely going to make things a LOT simpler. the ability to fine-tune the voltages on each node in your design makes a whole lot of sense and will definitely be used in any future iterations. Thanks again for publishing that and for your help.
-
That's not my design. I just copied the original Vox schematic, but untangled it so that it's easier to see the structure. The original Vox schematic has those filter caps scattered all over the drawing. Troubleshooting the Vox schematic is like an easter egg hunt! :icon_biggrin:
-
That's not my design. I just copied the original Vox schematic, but untangled it so that it's easier to see the structure. The original Vox schematic has those filter caps scattered all over the drawing. Troubleshooting the Vox schematic is like an easter egg hunt! :icon_biggrin:
Your schematics are fantastic. I've studied (well at least looked at anyway) most of the ones on your website. They have been invaluable in my learning process. I've built a couple of Marshall 18W designs, and the All American Duo, and used parts of others (I especially like the "Warbler" concept). It's amazing how simple or complex (headache inducing) the same schematic can be depending on who is drawing it. We all owe a lot to how well you communicate your depth of experience in amplification. I, for one, appreciate it greatly.
-
With thanks to pdf64 for the VOX AC30 schematic with voltages notated (see above) I've made some adjustments to fine-tune the voltages somewhat. Please take a look and let me know if you see any problems with where my voltages ended up after swapping out the various main power supply resistors (see chart attached). Thanks in advance, any comments are welcome.
-
Seems good 1st look over :thumbsup:
With the screen grids up over 350V, I’m still confused why the EL84 aren’t drawing more current :w2:
-
Seems good 1st look over :thumbsup:
With the screen grids up over 350V, I’m still confused why the EL84 aren’t drawing more current :w2:
Thanks! I have no idea either, but I'm confident in my measurement (i.e. I disconnected the cathode wire and ran my multimeter in series with it to get the amp reading, amp was at idle, but I think that's ok).
Everything seems to be functioning quite well. The VIB Channel is still about 25% less volume than the other two channels but I'm scared to even go down that path. The channel sounds really nice when blended with the other two so I think that will allow the VIB channel to have a quality life and sense of purpose.
I think this amp is going to be considered, dare I say... complete! Unless anyone else has something to add... there's still time.
-
The VIB Channel is still about 25% less volume than the other two channels
That's just the way it is. Nothing to do here. The AC-15 acts the same way.
-
... The VIB Channel is still about 25% less volume than the other two channels ...
The VIB Channel is still about 25% less volume than the other two channels
That's just the way it is. Nothing to do here. ...
The Vib/Trem channel is what I play through 95% of the time on my vintage AC30.
I never bothered to pay attention to whether I have to turn the knob a few degrees more. As long as I don't turn the Volume control full-up and still say, "It's not loud enough," then there's no problem.
-
Thanks for your help and comments. Looks like this thing is ready to rock!