Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: liddokun on April 26, 2020, 02:39:45 am
-
Hello,
I've finished an AC15 build, based on Sluckey's layout.
Most things have gone smoothly up until now.
The last kink I'm trying to iron out is two fold:
First, the volume pot on the normal EF86 channel. It passes signal even when the pot is set to zero. I thought it was a faulty pot at first, but tried several other pots and same issue.
Second, when both the top cut and volume (ef86 normal channel) are set to zero, there is a buzz, almost sounds like 120hz. As soon as either the volume or top cut pot is turned up, the buzz goes away immediately. I'm not sure if this is related to the first problem.
I've taken some voltages at the tubes
V1 (12ax7 trem/vib preamp)
1-67v
3-3v
6-190v
8-3v
V2 (ef86 normal channel preamp)
1-111v
6-107v
V4 (12au7 modulator)
1-67v
3-3v
6-67v
8-3v
V5 (phase inverter)
1-251v
3-63v
8-63v
6-238v
V6 (12ax7 PI/oscillator)
1-128v
6-190v
V7 (el84)
3-11v
7-341v
9-342v
V8 (el84)
save as V8
I was using Sluckey's schematic as reference, with his listed voltage measurements, I can see it's off, but not by much. I realize there may be sevaral issues at play here.
-
Check solder to ground
Show schematic or it is like any Vox AC15 ?
You do a nice job.
-
I used Sluckey's layout/schematic, which can be found here
https://sluckeyamps.com/VAC15/Vox_AC15.pdf (https://sluckeyamps.com/VAC15/Vox_AC15.pdf)
-
Thank you ?
Update; low pot value= low output power. It is not your issue.
How much chan Normal volume lets signal passed ? 1/10, 6/10 ?
Reheat some solders.
-
I suggest to try revising connection of the 0V terminal of both vol controls to the 0V turret of the LTP tail resistor R38.
That arrangement seems to better conform to Merlin’s ‘multiple star ground’ system, see Fig 15.14 http://www.valvewizard.co.uk/Grounding.pdf
-
I'll give that a shot and see how it goes
-
I suggest to try revising connection of the 0V terminal of both vol controls to the 0V turret of the LTP tail resistor R38.
That arrangement seems to better conform to Merlin’s ‘multiple star ground’ system, see Fig 15.14 http://www.valvewizard.co.uk/Grounding.pdf
I gave that a shot, it doesn't make any discernible difference to the initial issue of the normal channel passing signal still.
The vibrato/tremolo channel does behave as expected and passes no signal at zero.
-
Use a gator clip lead to ground the point in the pic. Does that totally kill the normal channel?
If so, ground the wiper of the pot with the gator clip lead. Does that totally kill the normal channel?
Only thing between the pot and that turret is a wire.
-
Use a gator clip lead to ground the point in the pic. Does that totally kill the normal channel?
If so, ground the wiper of the pot with the gator clip lead. Does that totally kill the normal channel?
Only thing between the pot and that turret is a wire.
Grounding that point in the picture does not totally kill the normal channel. There's no preceivable difference when I aligator clip that point to ground. Grounding the wiper of the pot comes up with the same result.
-
Grounding that point in the picture does not totally kill the normal channel.
Hmm. OK... Disconnect that white wire from the turret in the pic. Does that kill the normal channel?
-
Grounding that point in the picture does not totally kill the normal channel.
Hmm. OK... Disconnect that white wire from the turret in the pic. Does that kill the normal channel?
When I disconnect it, I get a louder signal and a ton of hum
-
OK, with the white wire still disconnected, ground the turret in the pic. Does this kill the normal channel?
-
Disconnecting the wire and grounding the turret does not kill the normal channel. I get the same amount of volume as before (as described in my oringal post).
-
Beginning to look like one or both of the 8µF filter caps are either not grounded or are just bad.
First use your gator clip to ground the negative lead of each cap, one at a time. Any joy?
Second use two gator clip leads to connect another filter cap parallel to a 8µF cap, one at a time. Any joy?
edit... solder joints on the negative leads of both those 8µF caps look suspicious.
-
Beginning to look like one or both of the 8µF filter caps are either not grounded or are just bad.
First use your gator clip to ground the negative lead of each cap, one at a time. Any joy?
Second use two gator clip leads to connect another filter cap parallel to a 8µF cap, one at a time. Any joy?
edit... solder joints on the negative leads of both those 8µF caps look suspicious.
Should I be doing this with the white wire disconnected?
-
You can reconnect the white wire.
-
Neither grounding each 8uF negative lead, nor strapping another filter capacitor parallel makes a difference.
edit: I also went ahead and reflowed some solder onto those suspicious 8uf negative joints.
I'll try to source some brand new 8uF caps tomorrow, and see if swapping those out makes a difference.
-
I suggest to try revising connection of the 0V terminal of both vol controls to the 0V turret of the LTP tail resistor R38.
That arrangement seems to better conform to Merlin’s ‘multiple star ground’ system, see Fig 15.14 http://www.valvewizard.co.uk/Grounding.pdf
I gave that a shot, it doesn't make any discernible difference to the initial issue of the normal channel passing signal still.
The vibrato/tremolo channel does behave as expected and passes no signal at zero.
OK, it was worth a try :think1:
Did it bring any improvement to Second, when both the top cut and volume (ef86 normal channel) are set to zero, there is a buzz, almost sounds like 120hz
-
I suggest to try revising connection of the 0V terminal of both vol controls to the 0V turret of the LTP tail resistor R38.
That arrangement seems to better conform to Merlin’s ‘multiple star ground’ system, see Fig 15.14 http://www.valvewizard.co.uk/Grounding.pdf
I gave that a shot, it doesn't make any discernible difference to the initial issue of the normal channel passing signal still.
The vibrato/tremolo channel does behave as expected and passes no signal at zero.
OK, it was worth a try :think1:
Did it bring any improvement to Second, when both the top cut and volume (ef86 normal channel) are set to zero, there is a buzz, almost sounds like 120hz
No, unfortunately not. I'm going to try replacing the 8uf filter caps, and reflow the solder on all my wiring.
There's a massive build thread of this same circuit that I found here https://el34world.com/Forum/index.php?topic=24337.0 (https://el34world.com/Forum/index.php?topic=24337.0) . This user seemed to have had some similar issues to myself as well, I'm going to scour through and see what I can find here as well.
-
If the amp is passing audio signal when volume pot is turned to "zero", it is likely the filter cap for the preamp is effectively "not there" (bad, cap, bad solder, bad wiring, etc...)
especially if as you turn the knob up, audio decreases slightly and possibly goes silent and then as you continue to turn it up, audio reappears and gets louder like you would expect.
What happens is the pot is doing it's job., no signal reaches the next stage via the next stage grid input, BUT it does reach the plate of the next stage by "escaping" up past V1's plate resistor and finding its way to V2's plate via V2's plate resistor.
the escaping V1 audio signal is out of phase with what the V2 plate produces. That is why it goes silent around 0.5-1 volume when the two signals completely cancel each other out.
Also, that escaping V1 audio is attenuated by the voltage divider that is the V1 plate resistor and the last dropping resistor in the B+ chain, so if the plate R is 100K and the dropping resistor is 10K then is about a ~90% signal voltage cut. That is why you get that phase cancellation notch of near silence around 5-10% volume pot throw.
If you have three or more triodes in a series (signal wise) all sharing the same B+ node, you can get some weird feedback as later stages feed each stages. Also a CF stage will put audio signal on its plate b/c that dropping resistor is not effectively the CF stage plate resistor.
-
If the amp is passing audio signal when volume pot is turned to "zero", it is likely the filter cap for the preamp is effectively "not there" (bad, cap, bad solder, bad wiring, etc...)
especially if as you turn the knob up, audio decreases slightly and possibly goes silent and then as you continue to turn it up, audio reappears and gets louder like you would expect.
What happens is the pot is doing it's job., no signal reaches the next stage via the next stage grid input, BUT it does reach the plate of the next stage by "escaping" up past V1's plate resistor and finding its way to V2's plate via V2's plate resistor.
the escaping V1 audio signal is out of phase with what the V2 plate produces. That is why it goes silent around 0.5-1 volume when the two signals completely cancel each other out.
Also, that escaping V1 audio is attenuated by the voltage divider that is the V1 plate resistor and the last dropping resistor in the B+ chain, so if the plate R is 100K and the dropping resistor is 10K then is about a ~90% signal voltage cut. That is why you get that phase cancellation notch of near silence around 5-10% volume pot throw.
If you have three or more triodes in a series (signal wise) all sharing the same B+ node, you can get some weird feedback as later stages feed each stages. Also a CF stage will put audio signal on its plate b/c that dropping resistor is not effectively the CF stage plate resistor.
Ah, got it. So I replaced the filter cap with a new one, I'm still getting this issue. Reflowed all the solder joints, in addition to the new filter cap. Any other suggestions as to where I need to look to address this?
-
I'm guess you don't have an o-scope....
double check to make sure you have 0-ohm between filter cap (-), the ground on the pot, and the ground under your 1st reservoir/filter cap.
beyond that, I'd suspect a wiring error somewhere...
-
You have eliminated all the logical gremlins. I too expect a wiring error.
-
I'm guess you don't have an o-scope....
double check to make sure you have 0-ohm between filter cap (-), the ground on the pot, and the ground under your 1st reservoir/filter cap.
beyond that, I'd suspect a wiring error somewhere...
Unfortunately I have no access to a scope.
My meter reads .02 when I measure between those 3 points.
-
I too expect a wiring error.
I agree,
liddokum, double check wiring.
If you build or fix amps, a scope is a must. You can find good one for $150 second hand, a audio signal generator too.
Most difficult is to know how to use them properly.
Good luck,
-
I noticed that the schematic shows very small filter cap values.
I’d start by increasing those, especially for the preamp, to e.g. 22uF.
You could test by temporarily parallelling bigger caps to the preamp filter stages with alligator clips.
-
I noticed that the schematic shows very small filter cap values.
I’d start by increasing those, especially for the preamp, to e.g. 22uF.
You could test by temporarily parallelling bigger caps to the preamp filter stages with alligator clips.
Original AC15 use 16 mfd.
OP follow Sluckey layout, I'm sure Sluckey built this amp and it work with low filter caps
-
I noticed that the schematic shows very small filter cap values.
I’d start by increasing those, especially for the preamp, to e.g. 22uF.
You could test by temporarily parallelling bigger caps to the preamp filter stages with alligator clips.
Original AC15 use 16 mfd.
OP follow Sluckey layout, I'm sure Sluckey built this amp and it work with low filter caps
No, 8µF is the original value for this amp. Small size is NOT the problem. There have been at least a dozen successful builds of this amp using this layout. Probably other builds that I don't know about.
-
I noticed that the schematic shows very small filter cap values.
I’d start by increasing those, especially for the preamp, to e.g. 22uF.
You could test by temporarily parallelling bigger caps to the preamp filter stages with alligator clips.
Original AC15 use 16 mfd.
OP follow Sluckey layout, I'm sure Sluckey built this amp and it work with low filter caps
No, 8µF is the original value for this amp. Small size is NOT the problem. There have been at least a dozen successful builds of this amp using this layout. Probably other builds that I don't know about.
I see those values what it look a old schematic on your link
https://sluckeyamps.com/VAC15/Vox_AC15.pdf
-
How to double check wiring;
https://el34world.com/Forum/index.php?topic=17701.msg178630#msg178630
-
I see those values what it look a old schematic on your link
https://sluckeyamps.com/VAC15/Vox_AC15.pdf
If you look closely you will see five filter caps on that old schematic... two 16s, two 8s and one 32. They are scattered about, but not too difficult to find.
-
You are right :BangHead:
-
Thanks for all the suggestions so far.
As mentioned, lots of successful builds with this layout, I probably made a wiring issue somewhere along the line.
Going to go through and double check everything.
-
120hz hum, sound af zero volume..
I would have expected a faulty preamp filter cap (c26). That being said, the 500k volume pot in conjuction with the 0.1uF(c27) capacitor to the PI, could be acting as a parasitic antenna, picking up stray capacitance from the nearby area, when the impedance is high enough (500k).
You could try to lift the cap from the turrets, and move it around (carefully), to see if the sound changed.
Also, just for good measure (pun intended), do you see a voltage on pin 1 of the volume pot, when it's turned down? If so, c12 might be leaking dc.
While you have the multi meter pointed at the pot, make sure the pot reads zero ohm across pin 2 and 3 when the volume is turned down, otherwise sound will creep through.
Hope you sort it out
Kind regards, Christian
-
120hz hum, sound af zero volume..
I would have expected a faulty preamp filter cap (c26). That being said, the 500k volume pot in conjuction with the 0.1uF(c27) capacitor to the PI, could be acting as a parasitic antenna, picking up stray capacitance from the nearby area, when the impedance is high enough (500k).
You could try to lift the cap from the turrets, and move it around (carefully), to see if the sound changed.
Also, just for good measure (pun intended), do you see a voltage on pin 1 of the volume pot, when it's turned down? If so, c12 might be leaking dc.
While you have the multi meter pointed at the pot, make sure the pot reads zero ohm across pin 2 and 3 when the volume is turned down, otherwise sound will creep through.
Hope you sort it out
Kind regards, Christian
Sluckey suspected a faulty filter cap too, and suggested this, and I replaced the cap with brand new ones, to no avail. I've also tried moving the cap around with longer leads.
Lastly, 0 volts on all 3 pins, unfortunately. I was hoping it would have been something like a leaky cap.
Alas, I will have to go through all my wiring. The amp isn't that neatly wired so maybe it's a good thing, anyways that I'm going through it all.
-
I built a lot of electronic gear. Each time they did not work on first power ON, is was one of my mistake in wiring.
Good luck.
-
Sooooo.....
I left this alone for a week or two and came back to it.
Reworked all the solder joints, went through the layout with a highlighter and double checked everthing. I damn near redid every off board wire.
I've managed to solve the main issue of signal passing through when the volume is at zero.
I'm just about at a point where I'm happy with this, but I haven't been able to resolve the "buzz when both top cut and volume at zero" problem. If both volume pots AND the top cut are at zero, I get a pretty loud hum, I think it's 120hz. When I turn any of those controls up, the there's a faint pop and the hum goes away.
Very peculiarly, if I take an aligator clip and clip it to one of the turrets circled in red in the attached photo, the hum disappears. The aligator lead is not connected to anything on the other end. Clipping the aligator lead to the other end (potentiometer terminal) does not have the same behaviour.
-
Solder the alligator clip there and put electrical tape on other end AhAh !
Where is my red arrow, this is the turrets you are talking about ?
What are the 2 black wires ? Power transformer ? Il yes, IMO these black wire are to close from the 2 white wires from the turret.
I'll put the 2 whites wire under the black ones, close to the chassis.
I don't like your lead dress. All whites wires I see must be run close to the chassis.
(https://live.staticflickr.com/65535/49908148886_bb16305c6d_b.jpg) (https://flic.kr/p/2j3diTC)Capture d’écran, le 2020-05-18 à 03.43.35 (https://flic.kr/p/2j3diTC) by Louis LAFLEUR (https://www.flickr.com/photos/137325788@N05/), sur Flickr
-
Those two black wires are the choke wires.
I've relocated those wires though, they no longer come into the chassis there.
I'll snap a few photos of my redone wiring and post.
-
All blue wires must cross other blues wire in 90 degrees angle and close to chassis, never close in parallel to each other
-
(https://el34world.com/Forum/index.php?action=dlattach;topic=25894.0;attach=83831;image)
Can you try something like this shield (from Larry Amps) ?
(http://spinoo.free.fr/projetG5/Images%20Amplis/Larry/British%20Purist/brit.pur103.jpg)
(https://el34world.com/Forum/index.php?action=dlattach;topic=13066.0;attach=27933;image)
Franco
-
Franco,
What would the shield do? Also, where would I be placing it?
Here's some updated photos of my wiring, after I reworked a bunch of it to be cleaner/address the initial problem I posted about.
I've circled the turrets in red that can be aligator clipped to make the noise problem go away.
-
try moving the white wire closer to my red trace
also try and move the blue ef86 wire further away from v5s green wire
you have tube shields on?
-
Connect the PI ground to the power ground. Any better?
-
What would the shield do?
What the shield do ? He shields ....
Larry used that trick probably because a nightmare incomed on his circuit/layout
Near for sure the shield was placed there empirically, with a try and repeat procedure, something like what you did with the gators
I was suggesting you to give a try to solve the problem putting on the layout a shield and moving it in a position that will give you positive feedback (if you are so lucky to find one)
Of course, before give a try to what Sluckey is suggesting, may be you can solve the problem without any shield (shield that isn't a for sure solution)
Franco
-
@shooter
Moved the white wire, as well as the blue wire on the ef86. Had no changes to lowering the noise. Did notice that if the white wire is closer to the board, the hum gets louder.
All tubes are shielded.
@sluckey
Relocated the PI ground to the power ground. I left the wire a little longer than usual, I noticed that the hum gets quieter depending on the position of the ground wire. But even when the ground wire is moved to a position where the hum is "barely audible", I still get the popping noise when I turn either of the top cut or volume controls up.
As I mentioned, clipping an aligator clip to any of the turrets I mentioned makes this hum go away completely. I'm curious as to why this is.
-
I'm curious as to why this is.
might be acting as an RF shunt, same principle as only soldering one end of the shielding.
what is you're bias set at, how much current ?
-
I'm curious as to why this is.
might be acting as an RF shunt, same principle as only soldering one end of the shielding.
what is you're bias set at, how much current ?
What's the best way to measure this on a cathode biased circuit? (please excuse my newbiness)
-
I use these bias calculator websites
http://www.tedweber.com/webervst/tubes1/calcbias.htm
Or
https://robrobinette.com/Tube_Bias_Calculator.htm
Take your voltage readings, plug in the numbers.
-
I remembered a small trick, something I've seen in some old amps
this can be a way to apply "Larry's" shield to your problem
is not a difficult thing to be tried
rolls a thin rigid wire tightly, without spaces, around the wires that go to those turrets, say for 5 - 10 cm
then join these thin rigid wires together (only one end) and connect this junction to ground
this will act as a small capacitor/shield which should discharge the disturbances picked up by the wires towards ground
does the noise decreases?
Franco
p.s.: This isn't the good example I would like to show you (the link is dead), however you can see what I mean
(https://i.imgur.com/G083fsT.jpg)
-
If you followed Sluckeys layout there are 2 1 ohm resistors from el84s cathode (pin 3)
to R60.
measure milliVOLTS across the 1 ohm R. That ='s tube current. post those values
-
If you followed Sluckeys layout there are 2 1 ohm resistors from el84s cathode (pin 3)
to R60.
measure milliVOLTS across the 1 ohm R. That ='s tube current. post those values
I followed Sluckey's layout but I missed this part. The 1 ohm resistors are in the schematic but do not show up in the layout.
I'll add these and get to measuring.
-
So the cathode current is 43mA
The hum is very very quiet now. But there is still a "pop" as either of the volume controls or top cut is turned up. The pop occurs about 30 degrees of rotation for the pot.
-
a "pop"
indicates a mechanical "switch", so something is going "open"/"Closed" in the signal path. should be able to "see" a flat spot in a pot, by just dialing real slow watching the ohm's
-
a "pop"
indicates a mechanical "switch", so something is going "open"/"Closed" in the signal path. should be able to "see" a flat spot in a pot, by just dialing real slow watching the ohm's
I noticed that moving certain wires around made the pop more or less quiet. Attaching the aligator clip to the turrets made the pop go away completely.
Gave the "twisted wire around the lead" suggested by Franco a few posts back. It seems that did the trick in eking out the last of the pop.
Super satisfied I've finally been able to sort this out, this amp's been through several rebuilds and debuggings.
I have attached a photo of it in the head cabinet I had made for it. Thanks all for the patience with me and the help.
-
The "pop" is a stage(s) going into supersonic oscillation. A voltmeter on cathodes may show a sharp change at the same time, perhaps from normal 1.5V to a starved 0.3V. OR poking a voltmeter in there may spoil the oscillation.
Another tip is to tease it to the "pop" and then use a lead pencil to move wires around. The osillation is usually sneakage from one wire to another. Just like a mike on a PA system, moving things changes the howl. Usually for worse, it may take a while to make it better, and in all control settings.
-
The "pop" is a stage(s) going into supersonic oscillation. A voltmeter on cathodes may show a sharp change at the same time, perhaps from normal 1.5V to a starved 0.3V. OR poking a voltmeter in there may spoil the oscillation.
Another tip is to tease it to the "pop" and then use a lead pencil to move wires around. The osillation is usually sneakage from one wire to another. Just like a mike on a PA system, moving things changes the howl. Usually for worse, it may take a while to make it better, and in all control settings.
Hi,
I was able to solve this by poking around with a chopstick and moving wires around. It took a lot of prodding but somehow I figured it out.
-
So I've returned to this thread because I don't want to start a new one just for the same amp.
Don't typically get the chance to crank my amps because I don't want to bother my family, but for a bit the entire family left the house to pick up some food so I got a chance to crank the volume.
There is a 60hz hum is not really audible at bedroom levels but gets pretty audible as the volume is turned up. I hadn't noticed this until I had the volume wide open. I'm trying to narrow the source of the hum down. It's only present on the the EF86 channel, the 12ax7 channel has no hum with the volume wide open. Hum is present as well when there is no input.
I'm *maybe* thinking it's the EF86, but I pulled the tube, and the noise is still present, albeit at a lower volume. I'm thinking the hum is occuring before the preamp tube, and it's getting amplified once it goes through preamp. Could it be a grouding issue? Any suggestions?
-
60hz hum
If it's 60hz, it's from the filaments (tubes, or wires and the way they run)
120hz is a different thing with different "fixes"
-
60hz hum
If it's 60hz, it's from the filaments (tubes, or wires and the way they run)
120hz is a different thing with different "fixes"
I just might be looking at 120hz afterall, I recorded the noise and put it through a spectrum analyzer, looks like the loudest frequencies present in the hum are around 120hz. What kind of fixes should I be looking at?
-
Maybe you can reduce hum by paralleling a filter capacitor to existing filter capacitors.