my first build - Vox AC30

[Dimitree]

hello everyone,
I almost finished my first build, that actually started many years ago (2010 I think), but due to time/money/other DIY projects, it took ages to complete.
It's a Vox AC30 Top Boost with minor modifications, mainly no vibrato, added EF86 channel (circuit for it still not fitted, will do someday) and Master volume.
I built the cabinet too, from scratch, using birch plywood, and then covered it with green tolex.
I still haven't powered it on, I'm waiting for a GZ34, and I'm still trying to understand what to do to avoid issues on the first power on, so I'm checking twice the connections, reading how to debug, and so on..also, the board layout I used is drawn by me so probably not correct 
Of course since this is my first build, I'll be lucky if it just works.
Also, I want to thank forum member and friend 'archaos' for the help and the materials he provided me.
What do you think?
Dimitri

[pdf64]

Looks nice, well done 👍
FYI it not s good idea to use tube socket terminals for tube pins noted as ‘IC’, ie 1 and 8 on a EL84. Different tube manufacturers use them differently, eg may be open circuit, maybe connected same as another pin, may be used as an internal tie point.
So it may be fine with one tube brand but cause problems with another.

[sluckey]

Here's an example... This 6BQ5 (same as EL84) has pin 1 internally connected to pin 2. If this tube were plugged into your socket the grid stopper resistor would be shorted out. In this case, you would probably never know it.   

[kagliostro]

I would modify that part of the Power Tube connection

after that

Very Nice Build 

Franco

[Dimitree]

thank you! I wasn't aware of the fact that I shouldn't use those terminals, I'm going to fix that for sure

[Jyvoipabo]

Good evening,

How to avoid first power-on problems?

It's easy.
Install a 100W filament lamp serial on the mains power (or in place of the fuse).

Then, remove all the tubes and switch power on.
Watch the lamp.
If there is no short circuit, the lamp must not lit.

Then add the lamps 1 by 1 starting by the rectifier.
Monitor the color of the lamp.

It should blush more and more, ending dark red color.
But in no case should it light up very strongly. That will indicate a short circuit somewere.

Don't forget to load the output transformer with a speaker or something like that.

JP

[nandrewjackson]

Jyvoipabo +1


Build a lightbulb current limiter, and start your powering up trials with no tubes installed. 


Take note, the lightbulb in the limiter you're making is in series with the mains hot wire. Normally lightbulbs are wired in parallel,  not in this case.


In theory, all of the amp's current will be THROUGH the lightbulb filament.  The lightbulb filament will give you a visual indicator of how much current your amplifier is pulling.


Maybe start with a 50 watt bulb, possibly 75watt. A 50 watt bulb should be able to handle the ac30 fully loaded, and the bulb may give you better visualization of how much current is flowing. 


As jyvoipabo said, if the amplifier were to have an internal short, the lightbulb will glow brightly.

[nandrewjackson]

Oh, and


NICE BUILD 

[Dimitree]

many thanks!
I just buit this light bulb tester, and I will try it tomorrow on the amp and let you know.

[Dimitree]

so I tried to power up the amp with no tubes and a 45W bulb in series with the mains as suggested. Looks ok, the bulb doesn't turn on, but the lamp indicator that I installed on the amp (across 6.3V) is on.
Then I added the GZ34, this time the bulb turned on quite bright for just one second, then stayed barely lit.
The GZ34 looks on, I measured +320V DC after the choke, but I measured AC voltages across the heaters and is not 6.3V but 5.3V

So I started checking heater wirings across all the sockets to understand if there was a short somewhere. But the heater wiring is fine, since once I disconnected the two wires coming from the PT, I read 200Ohm between them, and 100Ohm to ground on each side (I used a pair a 100Ohm to create a center tap). To be sure I also removed the two 100Ohm resistors, and there was infinite resistance bewteen the two sides of the heaters.
So there should be a problem on the PT? I checked the voltage across the 6.3V output (now disconnected from the rest of the circuit), and it is 5.7V. Resistance between the two green wires is something like 0.4Ohm

what do you think?

[shooter]

heaters and is not 6.3V but 5.3V

quite possible it's from being on the lamp limiter.   If you're 120AC is now 110AC (from lamp) the filaments will also be low

[HotBluePlates]

... FYI it not s good idea to use tube socket terminals for tube pins noted as ‘IC’, ie 1 and 8 on a EL84. Different tube manufacturers use them differently, eg may be open circuit, maybe connected same as another pin, may be used as an internal tie point.
So it may be fine with one tube brand but cause problems with another.

Here's an example... This 6BQ5 (same as EL84) has pin 1 internally connected to pin 2. If this tube were plugged into your socket the grid stopper resistor would be shorted out. In this case, you would probably never know it.   

While we may be in a habit to use pins as a tie-point with some output tubes, IMO it is imperative to avoid using Pins 1, 6, and 8 on EL84/6BQ5 because of the above.

EL84 data sheet

I recently acquired a bunch of EL84/6BQ5 tubes.  Just going through the RCA ones (that were 100% made-by-RCA), I found the following variations:

• Pin 1 connected to Pin 2 (G1)
• Pin 1 almost touches mica near Plate tab (arcing risk)
• Pin 8 bent near Pin 9 (G2)
• Pins 6 & 8 are short, straight stubs
• Combinations of the above

This was after going through just 14 6BQ5s and 7189s from only RCA.  For this brand, using Pin 1 appears to always be a bad idea, while Pin 8 increases the risk of arcing to the screen.  Sometimes the "unused pins" are left long and touch the lower mica to support it, and wind up very near some other tube element (sometimes the plate, sometimes the beam-forming electrode's tabs).

Unrelated: It was also interesting to see that even within RCA alone, I could identify some tubes that were true pentodes (with a suppressor grid) while others had beam-forming electrodes.  That was unexpected, since RCA's data sheets for both 6BQ5 and 7189 always describe them as pentodes.

Got a bunch of other American & European EL84/6BQ5 to document & test.  Curious to see what I'll find...

[HotBluePlates]

... added the GZ34, this time the bulb turned on quite bright for just one second, then stayed barely lit.
The GZ34 looks on, I measured +320V DC after the choke ...

The momentary flash of the bulb-limiter was in-rush current to charge the filter caps.  That's why the bulb brightness quickly faded (which indicates good, non-shorted operation).

You can assume no short-circuit is detected, and can either re-check with all tubes installed, or plug the amp into the regular wall outlet.  If there was an actual short-circuit, the bulb-limiter would remain at full brightness.

... I measured AC voltages across the heaters and is not 6.3V but 5.3V ...

That's normal, because the bulb-limiter is reducing the voltage applied to the PT primary.  Your B+ voltage "appears right" only because there are no EL84s in the amp to load it down (it will drop "below-normal" if you re-test with the EL84s installed and using the bulb-limiter).

Your heaters are/were fine.

[pdf64]

...While we may be in a habit to use pins as a tie-point with some output tubes, IMO it is imperative to avoid using Pins 1, 6, and 8 on EL84/6BQ5 because of the above...

Have you checked your AC30/6? The power tube sockets on mine and other JMI built ones I've seen had 1 and 2 lugs linked together.
Dang, I forgot to remove that link on the last one I serviced!

[Dimitree]

thank you guys!
so after some more tests with the bulb limiter, I concluded there were no shorts so I plugged the guitar and connected the amp directly to mains..and..it works!

I had to fix some wiring mistakes on the audio path (I couldn't hear the brilliant channel but only the normal channel), so now almost everything seems perfect..
Almost!

those are my current issues (or that looks like issues)

1) huge FM radio picking on the Brilliant channel (even with no guitar chord attached) that gets equalized by Treble/Bass controls and fades out with Brilliant volume. That doesn't happen on the Normal channel. I was going to enjoy a football match on my AC30..! 
2) crosstalk between Normal and Brilliant channel. What I mean is that if I plug the guitar in the Normal channel, and I keep the Normal volume at min, but the Brilliant volume at max, I can hear the guitar (not loud). The same is true in the opposite situation, using the Brilliant channel. I don't know if it is a problem or it is inherent to the amp design.
3) Normal channel is really dark, but probably is intended to be like that (I never played an actual AC30 through Normal channel).
4) As you can see from my schematic (attached below), I put a EF86 channel volume potentiometer going to the phase inverter tube. That channel is still not built, so the potentiometer is now just shorted to ground. But moving the potentiometer, the noise can increase/decrease. Do you think it's a ground loop issue, and I shoul just tie the unused input of the PI to ground, until I build the EF86 channel?
5) voltages are low compared to what one would expect (for example, filament voltage is 5.5V). The reason seems to be due to my house's mains, instead of 240V AC I'm getting 210V AC!! At the moment I don't know how to fix that..btw yesterday mains were 220V so looks like a network problem in my house or neighborhood.

I noted the voltages on the schematic. Those written in red are the mods that I did to the original AC30/6 schematic.
thank you for your support! 

[HotBluePlates]

...While we may be in a habit to use pins as a tie-point with some output tubes, IMO it is imperative to avoid using Pins 1, 6, and 8 on EL84/6BQ5 because of the above...

Have you checked your AC30/6? The power tube sockets on mine and other JMI built ones I've seen had 1 and 2 lugs linked together. ...

The '65 AC10 and ~65 AC30 Super Twin have nothing wired to pins 1, 6, or 8.  Grid stoppers are routed right to Pin 2.

I don't think I've got photos of the '63 AC30/6.  Might be a few days before I get a chance to open it up.

[archaos]

Yep, vintage Voxen indeed used 8/10 lug terminal strips for mounting grids. No resistors weren't mounted on pins 1, 6 & 8. I have pics in another HDD I'll post later.

[pdf64]

...I don't think I've got photos of the '63 AC30/6.  Might be a few days before I get a chance to open it up.

Here's a recent one 'as found' https://www.instagram.com/p/B9kii5UF6Qc/?igshid=14gzej4jpwfpd
The link loop wires mentioned are just about visible, especially on the far LH base.
Yes the 100ohm screen grids are replacements, but the 1k5 grid stoppers (and link loop wires) are original.

[archaos]

AC30 TB 1966

[archaos]

AC30 TB 1964

[orangehead72]

Very nice looking and great craftsmanship Dimitree. Sweet !

[pdf64]

AC30 TB 1966

This one looks to have terminals 1 & 2 linked, can't see enough detail on the 1964 to tell.
What brand / type are the red signal caps in the '66?
Was the '64 a 'front panel' TB type, like the '66?