slow AC-30 clone build.

[Mark]

That is what came to mind too, loaded or unloaded. The Mercury power transformers quote 303-0-303 as there secondary unloaded voltage. I'm not sure what the final B+ will be. I may have to pad it down with a Zener diode, or add the series resistor Merlin Blencowe mentions.

I fell for an old trap with 9 pin sockets it seems. I bought NOS Russian 9 pin ceramic sockets as I think ceramic is more durable than bakelite sockets, which I found carbonised between pins 2 and 3 when using 6L6's etc.
The trap was these are 22mm sockets and the chassis is cut for 18mm sockets. In a perfect world I would drill the chassis out to accept the larger sockets, I would need a drill press, clamps, and a large drill bit. It's easier to get new ceramic sockets. I had thought of using Belton plastic sockets as I've heard they are good too.

[Colossal]

That is what came to mind too, loaded or unloaded. The Mercury power transformers quote 303-0-303 as there secondary unloaded voltage. I'm not sure what the final B+ will be. I may have to pad it down with a Zener diode, or add the series resistor Merlin Blencowe mentions.

Mark, if you have concerns about the correct transformer, email Patrick. He has been very kind and helpful (at least to me). People that crusade that Mercury is somehow out to blow up every amp with 490V B+ transformers likely asked for the wrong transformer for their application. All of Mercury's stated voltages are unloaded. But I digress. For a 303V unloaded secondary and running a GZ34, I go with 1.20 to 1.23 as the rectification factor, so, 303VAC * 1.20 = 364V B+.

By example, one of the Radiospares 18W PTs I use is 301-0-301. That is an unloaded voltage and guess what?...it sags down to 289VAC running 122VAC on the primary with an EZ81, just like all the other 18W and 36W EL84 derivatives listed as 290-0-290, loaded. So it runs exactly where it should and there is no zener diode or other shenanigans needed to "correct" it for your build. The devil is in the details and knowing what an unloaded or loaded voltage is as well as the rectification factors you can expect comes with experience, as you know. For your AC30, there are quite a few PTs that could fit the bill, depending on where you want to end up with your B+. From what you said, that's going to be 320V and lower. Have a look at MMP-V30H-M Haddon PT which is 285-0-285. Rectified, I might expect that to be around 335-340VDC loaded, possibly lower because an AC30 is going to be sucking down 200 of the available 250mA. That is the PT I am considering for my EF86/AC30/4 build. I will likely pad those secondaries with 25-50Ω of series resistance too.

I fell for an old trap with 9 pin sockets it seems. I bought NOS Russian 9 pin ceramic sockets as I think ceramic is more durable than bakelite sockets, which I found carbonised between pins 2 and 3 when using 6L6's etc.
The trap was these are 22mm sockets and the chassis is cut for 18mm sockets. In a perfect world I would drill the chassis out to accept the larger sockets, I would need a drill press, clamps, and a large drill bit. It's easier to get new ceramic sockets. I had thought of using Belton plastic sockets as I've heard they are good too.

Alas, such is the way of things. You might consider using a stepped bit or reamer to enlarge your holes to 22mm. You can do that without a drill press. Or, just go with the Beltons. The Beltons are decent and I've never had any fail. They tolerate EL84 heat well.

[M Fowler]

https://www.angela.com/9pinamtransjapan ... ocket.aspx

If you need ceramic 9 pin these will fit .715 hole

[norburybrook]

That is what came to mind too, loaded or unloaded. The Mercury power transformers quote 303-0-303 as there secondary unloaded voltage. I'm not sure what the final B+ will be. I may have to pad it down with a Zener diode, or add the series resistor Merlin Blencowe mentions.

I fell for an old trap with 9 pin sockets it seems. I bought NOS Russian 9 pin ceramic sockets as I think ceramic is more durable than bakelite sockets, which I found carbonised between pins 2 and 3 when using 6L6's etc.
The trap was these are 22mm sockets and the chassis is cut for 18mm sockets. In a perfect world I would drill the chassis out to accept the larger sockets, I would need a drill press, clamps, and a large drill bit. It's easier to get new ceramic sockets. I had thought of using Belton plastic sockets as I've heard they are good too.

step drill, I do this all the time.Cheap on ebay


M

[M Fowler]

Dave,

Sorry for late reply. I have been using so many different brand PT and OT's lately rather then Heyboer stuff. Pacific, Edcor, etc.

Some I bought off Robert (rjguitars), some from Ampgarage guys, Mojotone (Heyboer), Hammonds, etc.

I try to buy them now with much lower HT because frankly I was cooking those poor EL84's to death. You could smell them and the paint was burnt off.

I also bought some 62R and 68R power resistors to lower the bias. Still on some I really should lower the dropping string rails been using 18K or 22k, 9k1 or 10k and 9k1 or 10k.
Don't have a voltage chart, sorry.

I use the Hammond 158L choke on all. Lately I have been using 5K2 rather than 4K3 primary OT's just because I had them sitting around.

The funny thing is I clean the amp up changing the plate resistor R4 from 220k to 120k or 150k but the owners want them changed back to 220k for more grit.

Just like Andy Fuchs I test all builds with my 1978 The Paul les paul which has some great high out pups but most of my customers use strats or teles.

Mark

[Mark]

I just saw this and am reading it.

https://www.ampbooks.com/mobile/classic ... /vox-ac30/

[Mark]

I have had a few thoughts on this build such as including current limiting resistors to protect the GZ-34.

https://ampgarage.com/forum/viewtopic.php?f=6&t=32131

In this thread, it's determined that 39 ohm resistors would be attached
Between the mains transformer secondary and the anodes of the GZ-34. The point of this mod is to increase the reliability of the GZ-34.

https://ampgarage.com/forum/viewtopic.php?f=6&t=32130

The link above covers adding protective fusing, there are three different opinions in the link.

[M Fowler]

I added diodes to the GZ34 tube socket on a local gigging musician's Dr Z EZG50 because out on the road the rectifier tube would give up the ghost and he came to me with that complaint so out with JJ GZ34 and in with DR Z/Don's recommendation to use Sovtek 5AR4 rectifier tubes.

I advised the player that if that rectifier tube goes you will be running on the diode rectification which will deliver higher B+ so bias can be off.
The amp hasn't been back since. I don't know if the rectifier tube is still good I should have him bring the amp over to check.

[Mark]

Hi Mark, that is good to know that the Sovtek is a better 5AR4 than the JJ version. I wouldn't think that that the diodes on the input to plates of the GZ-34 offer any protection as much as offer the transformer and first filter cap protection. I take it that you need more than diode in series with the plate of the GZ-34.

[Mark]

On to the next issue. The chassis I have is machined like an old chassis. This for the most part isn't a problem as many new parts still fit the chassis, I was lucky as I even managed to get the correct length tag strip, however the except is the can cap mount for the main filter cap, the holes for this clamp are 2" apart. I haven't seen a clamp like this. Drilling new holes is an option as is trying to slot the original holes in the chassis though this would be a lot of work.

I notice that there are 2 X 8uF can caps around, but I haven't seen the can clips Vox used that allowed them to be mounted horizontally.

Any thoughts.

The 9 pin sockets are on the chassis. The octal goes on once I get the right screws.

[M Fowler]

The backup diodes create full wave rectifier incase the rectifier tube fails.

Rectifier_Tube_Backup_Diodes.png

[chief mushroom cloud]

I notice that there are 2 X 8uF can caps around, but I haven't seen the can clips Vox used that allowed them to be mounted horizontally.
Any thoughts.

This will hold a 1in F&T 8x8 450V

https://www.ebay.co.uk/itm/H1-Bhc-Compo ... SwJzpco3Jq

[martin manning]

I wouldn't think that that the diodes on the input to plates of the GZ-34 offer any protection as much as offer the transformer and first filter cap protection.

Adding 1N4007's to the rectifier socket is no different from the practice of using two or three silicon diodes in series... You increase the allowable PIV, and if one diode does fail, you will still have one working.

[Mark]

I've had a few other wild and wonderful thoughts, these being:

1. Wire the cathode follower circuit up so it can also use a 12DW7, this valve has one 12AX7 triode and a 12AU7 triode. From what I've read the 12AX7 takes a real beating in the cathode follower position. Perhaps the compression of this circuit will sound really good with an 12AX7, but I will have the option of using a different valve that won't compromise the gain of the circuit.

2. There is the Merlin Blencowe mod where he puts a diode and resistor from grid to cathode on the cathode follower to prevent arc over. In one circuit there is a 100 ohm resistor between gain stage plate and the followers grid. I looked at his book too and I don't see the significance of that resistor. Any thoughts on the purpose of the 100 ohm resistor?
http://www.valvewizard.co.uk/dccf.html

3. As stated before the power transformer has a 303 - 0 - 303vac secondary, as was stated previously this will result in a rail of 360VDC when coupled with the 50 ohm cathode bias resistor, this is just insane as this will place an enormous load on the output valves. There are two ways to deal with this, one is to simply increase the cathode bias resistor till the valves are at 100% dissipation or less, the other method is to use a back biased Zener to lower the rail to a more reasonable 320VDC. I can see pro's and cons for both solutions. The diode just sitting there burning power does concern me a little particularly if it went short circuit.

4. I like Greg Fryer's idea of having a bleed resistor across the main filter cap, it drain the voltage away when the amp is off. I could also split this resistor to float the centre tap of the heaters at roughly 80VDC, which MIGHT result in a quieter amp.

5. Caps and resistors, I was going to use SoZo nextgen Mustard caps as I've heard they sound good in an AC-30. Any thoughts or experiences with caps in an AC-30?
As far as resistors go, I don't think the mojo is enough from carbon comp resistors to consider using them. I think I will get noise and value drift, which will exceed resistor mojo. Again any thoughts?

Thanks for your time and assistance.

[Mark]

I notice that there are 2 X 8uF can caps around, but I haven't seen the can clips Vox used that allowed them to be mounted horizontally.
Any thoughts.

This will hold a 1in F&T 8x8 450V

https://www.ebay.co.uk/itm/H1-Bhc-Compo ... SwJzpco3Jq

Thanks for the tip, I made the purchase.