Vox AC30-6/TB Vibrato Switches

[nevetslab]

Another Vox AC30-6/TB has come in with Rotary Vibrato Switch problems. Lorlin CK-series switches. One of them has locked up (3-Position Speed control) while the other, the 2-position Vibrato/Tremolo switch rotates continuously, regardless of the indexing washer present, which is supposed to serve as the stop limiter. Getting that PCB out of the chassis, with those stupid latching peg standoffs is always a thrill. I ran out of my nylon standoffs, which I've been meaning to install in all of our AC30-6/TB amps, which make it so much easier to remove, just removing pan head screws with insulating washers.

Anyway, I had to unsolder the switch PCB from the main PCB, which just didn't want to give up the solder to the Pace desoldering iron. Thankfully I didn't loose any pads, and got the assembly out without ripping the plate-thru's in the pads.

I've messed with this one switch, and the indexing washer simply won't limit the number of positions. I tried fabricating another indexing washer from a 3/8" dia solder lug, cutting the lug down to fit into the indexing holes, leaving the length much longer. No success with that either. The other one, when I had removed the PCB assembly, the black half had somehow separated, and I sealed it shut again.......now it's locked in one position and won't turn. I haven't tried taking it apart. Last time I tried that, the spring launched one of the ball bearings across the shop, hearing it hit the wall and tinkle into oblivion.

So, I searched for replacements, and found a number of models at Mouser, and ordered four of the CK1059 1-12 position Shorting. The CK1049 Non-Shorting was what I was looking for, but no stock. I figured starting over with fresh un-abused switches was easier, though that's chancing many more solder pads to desolder.

I still haven't figured out which holes to use to set the limit to 2 or 3 positions. Watching the rotor rotate thru the tiny rectangular holes make you THINK you have it right, only to find it just rotates past that post you stuck into the hole. No clues provided in the data sheet either.

Having found the Lorlin pdf file stopped working since yesterday, attempt to fix it in editing doesn't work!

[Jon Snell]

Use your multimeter and cold check the switch before fitting it or look at the data sheet attached. If set to position three it will click to four! Position 2 for three.
0900766b8007a1e8.pdf

[Dave H]

If set to position three it will click to four! Position 2 for three.

I guess it's because there's no point in having a 1 way switch option so the first position is marked 2 way, second position 3 way, third position 4 way etc.

[Jon Snell]

Took me a while to suss, out years ago when I used to manufacture light control modules and selectable amplifier controllers. Nothing seems simple until the penny drops!

[nevetslab]

The replacement Lorlin rotary switches arrived yesterday. Same indexing washer, but at least on the new switches, they DO limit the travel of the rotor inside. Setting the limits for 2-position and 3-position was easy. The shaft length on the Vox switches was 10mm, and these came in with 42mm plastic shafts. So, cut them down to size, scribing the shafts using an adjustable machinist square to set the length, scribed the line with an X-Acto blade, then cut to length with an X-Acto saw.

Getting the broken switches out of the Vox PCB was another matter. As luck would have it, the PCB is double-sided, with some of the traces on top under the body of the switches. Using both a fresh desoldering tip in the Pace, while holding the PCB assy in the bench vise, I fed fresh solder to the joints to aid in getting the solder to flow and get sucked up. Of course, the terminal fit inside the holes was tight enough to NOT allow all of the solder to be removed. Had to then resort to using both solder wick AND the desoldering iron, as well as melting thru the plastic to desolder the terminals at the top side of the PCB. Didn't matter what method I used, they remained soldered into the holes, while looking to be empty of solder. Slowly prying the switch bodies up, while applying further desoldering on joints eventually let the switches come free. AND, just to spite me, about 50% of the plate-thru shafts came out still clinging to the switch terminals! So much for having the right tools and years of experience to be outwitted by this. I fetched my calipers and my hole gauge, to see what the mechanics were. Hole size of the plated holes was 0.051", while the terminals (12 around the circumference and one in the middle) were 0.041" wide. If it was round leads on a resistor, that would be fine. On a rotary switch, the fit is very tight once you push the switch into the holes. The hole size where the plate-thru failed were now 0.055" dia. I would have selected a larger hole for the switch terminals in the fabrication.

As it was, I still had to apply solder to a couple of the solder pads on the top of the board when I soldered the new switches in, as I didn't have continuity between the PCB terminals and the switch terminals. Successful failure, with a restored working PCB assembly. WHAT a PITA!

[The Dude]

I think I would have tried cutting/breaking the hell out of the old switches down to separate pins and desoldered them one at a time.

[nevetslab]

I think I would have tried cutting/breaking the hell out of the old switches down to separate pins and desoldered them one at a time.

After THAT episode, that's EXACTLY what I'll do next time!! It didn't dawn on me to do that this time. Nothing more irritating than seeing the plate-thru copper firmly attached to half the switch terminals as you pry it up and out of the PCB!!!

[Jon Snell]

Plenty of flux and the correct size desolder tool works wonders every time, 0.06" is best for that job. To break the old switch will damage the through hole print and may cause issues when fitting the new switch. Good idea to flow fresh leaded solder as it helps with heat flow.

[nevetslab]

Now that CenterStaging, LLC has ingested all of their inventory from their New York facility for the time being, back here in Burbank, CA, there's plenty of maintenance work again to do. Starting with a Vox AC30-6TP having the Vibrato/Tremolo switch no longer stopping at the second position. The stock 'stop-washer' that is supposed to prevent the switch rotor from rotating past the stop selected with the pegged-washer rolls right past with little opposition. This time, having tired of fighting the solder pads and hole sizes in the PCB that allow the plate-thru's to blow out while removal of a switch, destroying it in the process....I tried a different trick.

Using some saved 3W resistor leads (from metal oxide power resistors), they being around 0.055" dia, I soldered a pair of them side-by-side, just wide enough to slip into the slot where the Stop-Washer peg fits into, That DID prevent the rotor from going past the new peg. I then cut a slot in the plastic rim around the washer & slotted holes, bent the newly fashioned stop over, shortened the peg a touch, then installed the original Stop-washer in place at position 3 and tightened down the mounting nut. That did the trick, without having to remove the switch. Cut the excess lead wires off, then put the switch PCB assembly back onto the main PCB assembly. Still, just removing that PCB using the Pace SX70 Desoldering Iron, as careful as I was, Vox leaves their hole sizes for those headers too small, and I lost three plated holes in the process. Had to solder on the component side to restore the circuits.

Next, now that I've stocked up 3/4" #4-40 Nylon standoffs, along with #4 nylon shoulder washers, I removed their stock snap-fit locking tab standoffs from the chassis, now using the nylon standoffs mounted in their place, and from the PCB side, use the #4 nylon shoulder washers to pass the #4-40 PHMS's thru, still insulated. Worlds easier to remove the PCB assembly now!

Then, looking to see where Vox had placed the power supply wire on the Rectifier tube socket. They again had it connected to Pin 2....the Heater. NOT Pin 8, which is the Cathode & Heater wire as it's supposed to be. Moved that. Added thin ITL washers onto the Power and Standby Switches, which I've been doing now for years, having found a huge stash of those at Luky's Surplus Hardware for dirt cheap. Those work great to prevent fuse holders from loosening up on Fender rear panel Combo amp's fuse holders, as well as on power switches so they don't loosen up.

Now, reviewing my methods of modifying the Stand-By Switch. I had been switching the power tubes' Cathode circuit to Ground, but, that does introduce a sizeable transient (that we don't hear), but does show up looking at it on the scope. Vox's stock method is switching the Plate AC voltage leads, which in time cause the 5AR4 tubes to fail. The last time I revised the Stand-By circuit, I hard-wired the plate voltage wires to the plates, left the Cathodes grounded, and shorted the push-pull grid wires together, muting the drive. As I recall, wasn't a total mute, but close to it. We often see Master Volume controls in that location, as well as the usual Tone Cut circuit that's present in this amp.

[nevetslab]

I found the Thread mentioned above on the various Standby Switch options: https://music-electronics-forum.com/...-repair/49154- I opted to place the Standby Switch across R50 & R53, where each of those resistors connect to the 1.5k grid resistors of the upper and lower push-pull tubes. Had to remember to rotate the Standby Switch so in Standby, it mutes. Removed the Plate wires between the Power Xfmr and the rectifier tube plates, joining each pair of wires with in-line insulated crimp terminals, bundled back within the wiring. For some reason, THIS particular amp's Standby / ON Lamp isn't lighting up. Only the Power lamp lights. That lamp is connected across the first Power Supply Filter cap C45. There is a resistor R81 in series with the lamp (82k) that I'll have to check. Neon Lamp assembly. And, as luck would have it, having installed the chassis back into the cabinet before powering it up on the bench, there's at least one VERY NOISY preamp tube to swap out. Can't reach in and find it without burning the dickens out of your hand!

[pdf64]

Counter to my usual view, the AC30 idles so hot that some type of lower dissipation standby mode may be handy. Not for powering up, don’t want to interfere with the GZ34’s slow ramp up; more for those ‘take 5’ breaks, where it’s hardly worth powering down, but it would be nice to save the EL84s operational life / allow the hot air in the cab to vent out.
Maybe switch in a higher value cathode resistor (to keep the bypass cap charged) and also mute the audio path somehow? Unfortunately that would seem to require a dpdt switch

[Helmholtz]

Maybe switch in a higher value cathode resistor (to keep the bypass cap charged) and also mute the audio path somehow?

Maybe disconnect the screen supply?

[nevetslab]

Maybe disconnect the screen supply?

While I like this method the most, implementing it is the challenge. If it weren't for the fact that Vox laid the PCB out with that Screen Supply buss on the back side of the board, and, to add to the difficulty, all of the other tube plate circuits are picking off of that buss, which originates from C46 Filter cap which follows (according to the schematic) CH1 (presumed to be a choke, but none exists in the chassis). So, C45 AND C46 appear to be in parallel, with just the Standby Amber lamp's source resistor R81 between them. Here's the schematics, followed by some photos of the layout:

AC3093PA-1.pdf

AC3093Pre-1.pdf





The Screen resistors are R73, R75, R77 & R79, spread across the PCB to be spaced to mate up with the wiring for the four EL84's on the chassis. As the PCB is double-sided, Vox drills the holes on the tight side, AND, they use Lead-free Solder, plus all of the other tube circuits are pulling their voltage off of this power supply buss, I'd either have to go pull up this PCB, cut and hack the buss to isolate the screen buss from that feeding the rest of the power supply decoupling resistors & their associated filter caps to the other tubes....or remove the buss end of R73, R75, R77 and R79 and create a floating buss above the surface of the PCB on top. That would need to be made mechanically strong, so either using strategic blobs of RTV to support the sleeved buss between the resistors, or some other more elegant mechanical support for this screen buss.

I could for trial's sake, unsolder those four resistors and buss them together, and wire up the S/B switch to turn it on/off.

Meanwhile, having restored the mechanics of the Vibrato/Tremolo switch, and discovered it's Grounding C16 in the Vibrato/Tremolo circuit that is the CCW-most position for Vibrato, and Tremolo, CW position has C16 open/ungrounded. I'm getting virtually NO Tremolo action from the circuit. Tweaking the Preset pot RV1 can't cure it. I've swapped out V9 ECC82 and V10 ECC83 with no difference found. I've checked the resistors in the circuit...at least those that are isolated and not loaded down by parallel networks. Can't check the caps, as all are radial lead parts, which again require lifting the board out again.



So, I still have to start probing with the scope to try and figure out how to get the Tremolo circuit to behave as best it can....as traditionally (as far as all of our AC30-6TB amps go), it's the weaker of the two circuits. The Vibrato circuit seems to work just fine. Seems in the past, I've had to tweak RV1 to increase the effect of the Tremolo at the expense of the strength of the Vibrato circuit. Any help in this from those who have these circuits well mastered over the years would be appreciated.

[pdf64]

Maybe disconnect the screen supply?

Wouldn’t that cause a (common mode) cathode current surge in the EL84s when standby is flipped to operate mode, due to the cathode bypass cap needing to charge up?
Normal power up, relying on the GZ34’s slow ramp up of HT, would tend to mitigate for that.

[Helmholtz]

Wouldn’t that cause a (common mode) cathode current surge in the EL84s when standby is flipped to operate mode, due to the cathode bypass cap needing to charge up?
Normal power up, relying on the GZ34’s slow ramp up of HT, would tend to mitigate for that.

Bugger (), you're right.
Didn't think of the discharged cathode cap.