(Slow) Build log for Vox JMI-era AC15 OA-031 circuit

[dchang0]

WOW, the light bulb current limiter absolutely does cut down the maximum volume of the amp, by a lot!

AND, it also apparently screws up the tremolo channel for some reason. The strange quietness went away completely as soon as I powered up without the current limiter--no other changes. And it is LUSSSSSSSSHHHHHHHH!

So, the amp is working perfectly on first power-up with no changes. I have not even measured voltages yet.

I can confirm the tremolo pots (Weber mods) are wired correctly as shown in the photos. Clockwise increases the speed and depth.

The only control that needs to be rewired from counter-clockwise to clockwise is the Brilliance pot (the bass-cut shelving control from figure 10.7b on page 233 of Merlin's book on preamp design). Easy to do.

---

With the tremolo footswitch jumpered and the push-pull on/off pot pulled up for off, there is a bit of tremolo that comes through in the sound. In other words, it's not a completely dry signal, like the normal channel. There is still tremolo in the mix. Is this normal? Remember that I have added the Weber mods to the tremolo controls--they may be responsible for the problem, but maybe the original Vox design also had the same problem.

---

I'll post up voltages soon.

[dchang0]

Here's the first power up clip.

WATCH OUT FOR THE LOUD SECTIONS AT 1:30 and 2:14.

The amp starts with every knob at half. I am on the tremolo channel. I just start fiddling with these knobs in no particular order, going from extreme left to extreme right in most cases.

Tremolo:
Depth
Speed
Cut
Volume

Normal (around 1:25):

You can hear the tinking of the borderline microphonic EF86 when I pull the plug and move it from jack to jack.

More fiddling in no particular order.
Volume
Brilliance (not much of a range from extreme to extreme)
Cut

It is kinda weird having an amp with two cut controls.

Tremolo (around 2:00):
You can hear me clicking on and off the tremolo function, but there is still some tremolo wet signal in the off position. Depending on the depth, there is more or less wet signal when it ideally would be dry.

---

Guitar is a Les Paul Custom with EMG 81 and 60 with both volume knobs at 8/10 and both tone knobs at 10/10. Strings are 3-year old D'Addario Chromes flatwound 10s (mellow on purpose). Mic is Kel Song Sparrow with gain turned very low. Speaker is a 10" Weber Signature Alnico 10S in a Champ 5F1 cabinet.

I only used the right-hand jacks (I forget whether these are high or low impedance, LOL). Question: on an original 1960s AC15, which jacks are supposed to be the high impedance ones, the ones on the right or left when facing the rear of the amp (jacks are at left end, power is at right)?

Tubes are all Electro Harmonix excepting a JJ 12AU7, JJ EZ81, and Tung Sol EF86.

AC15FirstPowerUp.mp3

Overall, I am very happy with the results.

[dchang0]

Lower chassis tube pin voltages:

V8
pin 3: 341.8V

V4
pin 2: 162.5V
pin 3: 11.25V
pin 7: 314.9V
pin 9: 316.4V

V3
pin 2: 149.3V
pin 3: 11.25V
pin 7: 314.7V
pin 9: 315.1V

Measuring voltages on the upper chassis tubes turned out to be impossible due to the oscillator moving faster than my multimeter could keep up. I verified the static voltages marked on the OA/031 schematic (like 190V at pin 6 on V5) and called it a wrap.

---

The microphonics on the EF86 were mostly solved by putting the Fender-style spring-loaded metal tube shield onto the socket and a bit of heat shrink tubing on the top half of the tube. Dramatic reduction in noise. However, it is still microphonic enough that if I turn the normal channel's volume up above about 7.5/10, it starts to feed-back, even without any guitar plugged in to the amp. Guess I gotta buy a new EF86. Rather than buy a bunch of EF86s and try them all and thus waste money, I probably need to specifically buy a guaranteed, tested non-microphonic EF86.

Reversed direction of Brilliance pot so that fully clockwise has the most bass cut.



Now, everything works as it should (except perhaps the slight tremolo wet signal that should not be present when the tremolo is off).

The amp has very, very low hum and noise when nothing is plugged in, so I definitely lucked out with the JMI-era grounding scheme and heater wires layout. They probably should not have ended up so quiet.

Next: ordering small parts to mount the chassis properly into the combo cabinet, including guitar jack plates for mounting the footswitch jack and speaker jack through the slider board. Could be a while before all the parts come in.

[dchang0]

Thanks to this excellent video, I realized that the vib/trem rotary switch (not the vib/trem footswitch) does not behave as I had expected. It switches between high and low amounts of wet signal.

https://www.youtube.com/watch?v=igZFOcbNtmU

I say wet signal because I am not exactly sure how the vib/trem circuit works. I have heard that the JMI-era AC15 has a true pitch-shifting vibrato combined with a volume-modulating tremolo. Sluckey's version of the AC15 refers to the TREM ON/OFF rotary switch as VIB OR TREM. I don't understand the theory well enough to know what's going on with this switch, but I do know that the signal on this channel is never completely dry.

So my amp's vibrato/tremolo channel is working properly. Turns out the Ted Weber mods are not very significant--he only replaces the Speed 3-position switch (330K, 560K, 1M) with a 1MA pot and a 220K resistor. Thus, the only difference is the range of available speeds (a wider range from 220K to 1.22M). It would not and should not affect the amount of wet signal in the tremolo channel.

So, basically, I am good to go as soon as I fix the EF86 microphonics by decoupling the tube socket from the chassis. That is the only remaining problem.

After that it is a matter of finishing installing the hardware in the combo cabinet.

[pdf64]

...Brilliance (not much of a range from extreme to extreme)

It is kinda weird having an amp with two cut controls...

...Note the Brilliance switch is replaced by a bass-cut shelving control from figure 10.7b on page 233 of Merlin's book on preamp design. I have no idea whether this will sound good or not; I just didn't want to go with 6-discrete-positions switching...

A switch to select different values of coupling cap would be way more effective than that bass slope control, which, assuming 1M slope and 500k volume control values, only has ~9dB range.
A friend has a commercial clone that just has a straight on/ff switch to short out a series 470pF cap, leaving the 10nF only in circuit. But a 3 or more position switch should give a much better control resolution, which I find helpful when changing between differing guitars.

Are the controls are wired so that something increases as they're turned clockwise? That may help avoid them being perceived as 'cut' controls.
If the amp only seems bright enough with the (LTP treble) cut control on max resistance, consider adding a bright cap across the volume control, 220pF between terminals 2 and 3.

[dchang0]

A switch to select different value of coupling cap would be way more effective than that bass slope control. A friend has a commercial clone that just has a straight on/ff switch to short out a series 470pF cap, leaving the 10nF only in circuit.

Are the controls are wired so that something increases as they're turned clockwise? That may help avoid them being perceived as 'cut' controls.
If the amp only seems bright enough with the (LTP treble) cut control on max resistance, consider adding a bright cap across the volume control, 220pF between terminals 2 and 3.

Re: "something increases as they're turned clockwise? That may help avoid them being perceived as 'cut' controls"

Hmm, good point. I hadn't thought of it that way. Previously, I had accidentally wired in the Brilliance bass-shelving pot backwards, and when turned clockwise, it would increase the bass and made the tone feel full and fat. Now, it's wired "correctly" where it is bright and thin when turned fully clockwise, somewhat meeting the expectation of "increasing" Brilliance as it is turned. It still sounds like the treble stays flat and the bass is removed--a cut control, in other words.

I did buy all the parts necessary to reproduce Blind Lemon Chitlin's 6-position rotary Brilliance switch in his BLC 15 amp that selects between different capacitor values. It was just in case Merlin's design didn't work out. Right now, the bass-shelving control is borderline acceptable--I merely wish the range from 0 to 10 were wider. Actually, what I would really like a perceivable increase in brilliance without removing bass--but that is probably a pipe dream without going to a Top Boost add-on.

I'll give your suggestions a try later on and let you know how it goes. I will have to order some capacitors first. Thanks!

[pdf64]

...I had accidentally wired in the Brilliance bass-shelving pot backwards, and when turned clockwise, it would increase the bass and made the tone feel full and fat. Now, it's wired "correctly" where it is bright and thin when turned fully clockwise, somewhat meeting the expectation of "increasing" Brilliance as it is turned. It still sounds like the treble stays flat and the bass is removed--a cut control, in other words...

I think that the 'Brilliance' label is just to indicate that it's mimicking the brilliance channel when selected, ie with a lot of bass cut.
The brilliance channel is only bright in the sense of it having a lack of bass when compared to a normal channel!
As a control, I suggest that you wired it up the best way 1st time, and that it would be best to think of it as a bass control.
The slope control makes it a halfway house, as even at max resistance, there's only a -9dB bass cut shelf, rather than a true high pass -6dB/octave filter, as per an original brilliance channel.

...I did buy all the parts necessary to reproduce Blind Lemon Chitlin's 6-position rotary Brilliance switch in his BLC 15 amp that selects between different capacitor values. It was just in case Merlin's design didn't work out. Right now, the bass-shelving control is borderline acceptable--I merely wish the range from 0 to 10 were wider. Actually, what I would really like a perceivable increase in brilliance without removing bass--but that is probably a pipe dream without going to a Top Boost add-on.

Yes, to actually get more zing on the top end without losing bottom end, a bright cap may definitely help.
A multi-position switch can be used to great effect for a bright switch, as it can be used engage various different bright cap values, so as to home in on the best value for the particular guitar, lead, volume setting, speaker, room, to achieve the desired effect.
For cap values, as a starting point, consider 0, 47pF, 100pF, 220pF, 330pF, 470pF.

[dchang0]

Thanks for the explanation of the underlying theory!

I have these values in silver mica or ceramic bright caps:

220pF
500pF
1500pF
2200pF
4700pF

These are quite different than the values you recommended. I can't return them, so I guess I'll try them out and see how they work.

EDIT: Turns out these 6 values are from sluckey's AC15, not BLC's BLC 15: http://sluckeyamps.com/VAC15/Vox_AC15_mods.pdf

Given that sluckey already uses these values and recommends them, I fully expect them to work.

[dchang0]

I noticed while building Sluckey's 6-position bright capacitor switch that while turning the switch, the signal cuts out between positions and there is a loud pop when the switch reaches the next position. This basically means the Brilliance switch cannot be turned while the Normal volume is above zero. Not a problem if one is switching guitars, but definitely a problem if one is playing the same guitar and just wants to change the tone instantly.

Would this problem be solved by putting another cap in parallel across the switch's poles? I figure at least the volume would not cut out, but the pops probably would still occur. The end result may not be desirable or useful, though, depending on the value of the capacitor bridging the poles.

Also, out of curiosity, did the original design of a 250pF cap in parallel with a on/off SPST switch also pop when the switch was flipped from one state to the other?

EDIT: Found this discussion about the issue. http://www.diystompboxes.com/smfforu...?topic=63135.0

The little circuit described by Mark Hammer and drawn by moro may work, although I don't know if it is possible to pull off one capacitor in series with my 6-position switch selecting between six capacitors, when the switch is still non-shorting and completely breaks the series circuit in between positions. Will take some thought to figure out a solution...

Maybe it's as simple as buying a shorting rotary switch equivalent...

EDIT: I just noticed that Sluckey's instructions specifically say to use a MBB (Make Before Break, i.e. shorting) switch, so that question is settled for certain.

[pdf64]

...I have these values in silver mica or ceramic bright caps:

220pF
500pF
1500pF
2200pF
4700pF

These are quite different than the values you recommended. I can't return them, so I guess I'll try them out and see how they work.

EDIT: Turns out these 6 values are from sluckey's AC15, not BLC's BLC 15: http://sluckeyamps.com/VAC15/Vox_AC15_mods.pdf

Given that sluckey already uses these values and recommends them, I fully expect them to work.

Sorry, the cap values I was referring to was in regard of a bright switch, ie placing cap across terminals 3 and 2 (input signal and wiper) of the volume control, as per many of the larger Fender tube amps.
Such an arrangement will retain a full bottom end whilst adding a selectable degree of top end boost.

The brilliance switch / sluckey switch mod selects coupling caps to progressively roll off the signal's bottom end, such that as bass is removed, the tone becomes, by comparison, more brilliant

Hope that makes sense?

The switching pop can be eliminated by adding a 10M resistor to 0V to each of the caps, thereby maintaining a stable dc reference for both sides of all caps at all times.

The use of a switch with shorting contacts would eliminate the signal cutting out momentarily when it was actuated; I think that the dc referencing resistors would still be required to prevent switching pops though.

[dchang0]

Thanks for the clarification and explanation of the switching pops--I'll add more parts to my shopping list.

[pdf64]

The 10M resistors can be tiny low power type, MF probably best to keep noise at bay, though I've never found that CF are a problem.
The pot track is likely a bigger noise source.

[dchang0]

Cool--thanks. I'll reinstall Merlin's bass-shelving cut control for now while I build up my tiny shopping list (shipping is probably going to cost more than the components). I'd also like to sit down and study the underlying theory of your suggestions.

[dchang0]

The square-profile silicone o-rings arrived today from McMaster Carr. The 7/8in I.D., 1in O.D. sits perfectly under the ceramic tube socket. It is slightly loose (does not wrap around the ceramic socket snugly). I am glad I spent more to buy the square profile ones, as the round one might not stay in place when the socket is screwed down. I figure round ones would be okay if they were stretched under tension around the ceramic socket (smaller I.D.), but there is still a chance that a round-profile o-ring would get squeezed through the chassis hole



I am waiting for the 4-40 nylon-insert locking nuts to arrive (the photo shows Keps nuts that will definitely work loose under vibration). You can see that the underside rubber o-rings will be pressing up against each other when the screws are installed--it will make installation a bit tricky. I may have to omit the two underside small o-rings, keeping only the ones on top.

Also, this approach did not require that I de-solder anything, and the pins do not need any heat shrink for insulation because there is no chance of them shorting out on the chassis. I may end up having to de-solder some things to get the screws installed, though.

[dchang0]

The switching pop can be eliminated by adding a 10M resistor to 0V to each of the caps, thereby maintaining a stable dc reference for both sides of all caps at all times.

About "both sides of all caps"--I'm studying your recommendation by reading this article here: What really causes switch pop | Mr. Black

The third diagram http://cdn.shopify.com/s/files/1/017...downs.png?1324 is somewhat like my 2-pole, 6-position switch--pretend the Circuit Block black box is a single capacitor and the coupling cap right after the Circuit Block is missing. Does this mean I will need 12x 10M resistors, one for each side of the 6 capacitors on the 6-way rotary switch?

Or, can I pull it off with just 6x 10M resistors, with one at each of the output end of the 6 capacitors like in the fourth diagram? http://cdn.shopify.com/s/files/1/017...ypass.png?1326

I figure the latter will work.

I guess I will order 12+1 10M resistors just to be sure I have enough on hand.

EDIT: Due to Sluckey's design, because the "to board" side of all the capacitors are tied together, a single 10M resistor could work on that side, so the total to order would be 6+1 10M resistors, rendering the above question mostly academic.

The current limiting resistors (1M in the article) pose a challenge. If I understand it correctly, I would need 1+6 1M resistors, meaning the rotary switch will be crawling with point-to-point-wired components*. But I probably don't want the current-limiting resistors anyway, since this is an amp and not a stomp box...

Here is a schematic of what I envision the "Sluckey's AC15/AC30 6-Position Brilliance Switch with Added Pull-Down Resistors" should look like:



Thoughts?

---
*Another option is to step down from 6 positions to fewer. This would provide more space for fewer point-to-point components. Perhaps a 2-pole, 4-position switch would be much smarter. Here is a 3-pole, 4-position, shorting solid shaft switch (just ignore the 3rd pole): https://www.tubesandmore.com/product...les-4-position