Scratch build of Mullard / Dyna A420 UL tube amp

BuzzK

AK Member
Subscriber
I have a bunch of parts lying about -- including:

1. A Heyboer replacement power transformer for an Eico HF 87;

2. A pair of Magnequest Dynaco A420 UL OPTs;

3. A chassis that is partially punched;

4. Vintage Mullard EL34 & EF86 tubes; and

5. A PCB for Dave Gillespie's EFB™ bias control

I'm going to cobble together a 30 WPC tube amp. I'll provide pics along the way. Stay tuned if you are interested . . .
 
Register to hide this ad
Here is the schematic.

AK's Dave Gillespie designed the EFB™ section and provided invaluable help and counsel on the concept & schematic.

I prefer fixed bias to cathode bias. And Dave's circuit will help to optimized operation by decreasing distortion and maximizing power and tube life.
 

Attachments

This is the raw chassis. It consists of 1/8" aluminum top and bottom panels, and a 24 gauge steel frame -- all powder coated with PIM nuts for the 8-32 attaching hardware.

Next I'll lay out the top panel for drilling and lettering. I make "footprints" of the various components out of paper, so I can move them around and determine optimum placement.

After I have the layout, I have an ancient CAD program for the Mac that I'll use to create a drill template. Also a lettering template for the trophy shop, who will laser etch the indelible lettering into the sheet metal.

This time I'll have a fabrication shop cut the holes in the top panel with a laser. There will be some square cut-outs, which are always a PITA to do by hand.


Raw chassis overview.jpg Chassis rear.jpg Chassis interior.jpg Graph paper.jpg
 
Last edited:
I was thinking about the best way to measure & display bias. I've often done it with test probe jacks for each tube's cathode (across a 10 ohm resistor), and a ground terminal. You connect a VOM to adjust for bias and balance.

However, I came across these little voltmeters on eBay -- 5 of them for about $10. They measure hundredths of a volt, which makes them suitable for displaying bias voltages:

https://www.ebay.com/itm/281478533527

I thought about using just one of them, but then I'd still need test probe jacks for balance -- they don't show negative voltages. But at this price, I could use one per EL34 tube, and show bias voltages in real time. It would be interesting too, in a nerdy way . . .
 
I was thinking about the best way to measure & display bias. I've often done it with test probe jacks for each tube's cathode (across a 10 ohm resistor), and a ground terminal. You connect a VOM to adjust for bias and balance.

However, I came across these little voltmeters on eBay -- 5 of them for about $10. They measure hundredths of a volt, which makes them suitable for displaying bias voltages:

https://www.ebay.com/itm/281478533527

I thought about using just one of them, but then I'd still need test probe jacks for balance -- they don't show negative voltages. But at this price, I could use one per EL34 tube, and show bias voltages in real time. It would be interesting too, in a nerdy way . . .

Can these mini meters be powered by measured voltage or do they need an isolated power supply?
 
Can these mini meters be powered by measured voltage or do they need an isolated power supply?

They come with three wires -- two are for power supply and one is for measurement. You can put the measurement and positive lead together and it will report the supply voltage (I tried this with a lithium cell). If you are measuring cathode voltage than a separate DC power supply is needed.
 
nice, some of those cheap meters power from what they measure, which makes them rather less useful at low volts.
 
Here is a rough parts placement layout. A couple of pieces are missing, but this is the general idea.

I "rediscovered" a Velleman K4307 digital RMS wattmeter among my stuff, so I decided to use it. Of course I had to buy another one for stereo. I don't need all of the LEDs, nor do I want to use the plastic labels provided in the kit. But at least I have a template.

There are going to be a lot of holes in this top cover plate, but since I am going to use a laser cutter (instead of hand tools), I'm going to go for broke.

There is also going to be a lot of "eye candy" on this amp -- bias voltage pots; bias current displays; wattmeters, A/C balance controls, etc. All to be labelled by laser etching. I'm going to include a switch for disabling the digital bias meters. I'll install probe test points for each tube in case a meter fails or calibration needs to be confirmed. A hum balance pot will be mounted on the back of the frame.

The green EFB™ circuit board show will be tucked away below deck, out of sight.

Laying all of this out on the CAD program will be time-consuming, but when you figure that the same drawing will be used for cut-outs and lettering, it's time well invested.

Parts layout.jpg
 
But hey, for $2 each, who's complaining? :)

well, what I mean is the sort that power from what they measure may not be usable for a bias current display. With an external supply input that limitation goes away.

and it is sort of amazing what you can get for basically nothing now. Once upon a time digital anything was voodoo magic big dollars, now if you want an analog meter you're going to pay considerably more just because the mechanical stuff is more expensive to make than IC chips are.
 
This is what the laser cut-outs look like on the top panel. The panel will have to be powder coated again. Looks like that laser got it pretty hot.

I'll have a trophy shop with a lower-powered laser etch through the paint to make silver lettering.


Lasered plate.jpg
 
Back
Top Bottom