Hammond / Everett AO-50 conversion to Vox AC15

[beatnik]

Hello everybody,

I got this tube amp chassis unit, model AO-50-3C. I found information that it was an Hammond produced amplifier mounted in an Everett organ. I couldn't find any schematic for the original circuit, but I could see many similarities with the Vox ac15 circuit. ez81 as rectifier and a pair of el84 for the power stage, a 12ax7 for the phase inverter and a combination of ef86 and 12ax7 for the preamp stages. I had been really missing a vox sounding amp in my setup so I decided this amp would have been perfect as a conversion experiment.

I started by carefully studying the vox schematic and comparing to the circuit, taking out some components that weren't needed and then wiring up the power supply circuit with new filter capacitors. The original circuit didn't have a filter choke so I have added a nice vintage partridge uk made

After testing the psu section for correct operation I moved on to the preamp stage, starting with the ef86 channel of the vox ac15, the so called normal channel. The original circuit used pretty much the same capacitors and resistors values so i didn't need to change much here.

Then moved on the phase inverter stage. This also had basically the same resistors values as the vox, but wired somehow differently so had to do more work

Before working on the power stage I wanted to do some voltages tests with and without the preamp tubes. I found out to be quite different from the vox schematic i am referencing to.

Without the tubes in circuit I have measured around 420V on the preamp tubes anodes

With the ef86 in circuit i get the following readings

plate (pin6) 50V
cathodes (pins 3 and 8) 3.5V
screen grid (pin 1) 68V

With the phase inverter in circuit I measure around 300V on its plates (pins 1 and 6) and 68V on the cathodes (pins 3 and 8)

The Vox schematic I am referencing to tells the ef86 plate should be at 90V and the phase inverter plate at 220V. The b+ voltage is also different, the schematic calls for 315V whereas my voltage is about 388V, although i only have the first two tubes in

I would really like to get some feedback on these voltage readings before moving on to the power stage. I am not really a tube expert, I have some building skills but my theory and understanding of circuit operation is quite limited. Some advice would be helpful in order to get to the final result without damaging components, especially the transformers. I have attached the Vox AC15 schematic I am following

[beatnik]

i had forgot to mention that the b+ voltage without any tubes in (besides the rectifier) is around 420V. I am wondering if this would be fine for the ac15 circuit or I should do some changes on the plate resistors ?

[DRH1958]

Just in case you want the service manual for this organ, here's the link: https://archive.org/details/Hammond2...nServiceManual It has the AO-48 which is almost the exact same chassis as the AO-50. It sounds like you've gotten a good start but maybe this will help some more. Pretty sure you want the PDF file as there are several to chose from.

Also, I think you want the screen grid lower than the plate on the EF 86. The organ schematic shows 80 volts on the plate with a 100k load resistor and a 390k screen grid resistor with a 250 volt supply but doesn't give the screen voltage. If you look at this data sheet, it gives a typical condition as 250 volts plate and 140v screen. See page one. http://www.mif.pg.gda.pl/homepages/f...030/e/EF86.pdf Of course you can run the screen at many different voltages but from what I know, small signal pentodes usually run the screen around half to 2/3 of the plate voltage. It depends on what you want to accomplish. Here's a good primer on small signal pentodes: http://www.valvewizard.co.uk/pentode.html One example he gives is 250 volts on the plate and 100 on the screen.

[beatnik]

Thanks for your reply

I believe the AO-48 is different in the ef86 stage , as the resistor values are different from what i had in my amplifier, which basically had the same resistor and capacitor values as the vox ac15, minus the cathode bypass capacitor, 100uF. Plate load and screen grid resistors were definitely 220k and 1m

I double checked the wiring and measured voltages again. The screen grid is about 20V higher than the plate , and you totally make a point stating that should be the other way around.

I've also noticed another thing, if i ground the input the voltages go up to 130V for the screen grid and 105V for the plate. Is this normal at all ?

What could i try to get the right voltages on the ef86 ? should i try changing the plate load resistor to 100K ?

[DRH1958]

Do you have the grid referenced to ground right now like the schematic shows? Any common cathode gain stage needs the grid connected to ground by some resistance. So I'm guessing that this is changing how the tube is conducting when you ground it out. Notice in the Vox schematic how the input grid is connected to ground via the 68 grid stoppers and that 1 Meg grid leak resistor. I can't say for sure why the screen grid is higher than the plate.

I would keep the resistances on the EF86 as they are if you want a Vox. Also, concerning your question about 420 volts on the preamp tube plate pins without tubes installed; your voltages will come down as you load the amp with tubes, especially the power tubes. I see that the AO-48 has a 320 volt first node in the power supply and the Vox is 325, so that should be close enough.

One other thing, I haven't done anything to my AO-48 yet, so I can't speak to experience with it. I'm saving this one for later. Too many other projects right now to finish.

[DRH1958]

I found an interesting item in one of my books when doing more research on the EF86. It had the exact input of the Vox AC 15 as the example and it showed a supply of 255 volts, plate resistor of 220k, a screen grid dropping resistor of 1 Meg., everything the same. It showed the plate at 81 volts and screen at 99 volts. It goes on to say that the 1 Meg resistor provides the necessary voltage drop for a suitable working voltage(since DC screen current flows in this resistor). When the amp is at idle, it's called DC idle(static) condition since there is no signal applied. Then when signal is applied, then it's considered dynamic AC condition which is different than idle. So my thinking is this is only at DC idle condition that the screen is higher than the plate and then under load, the screen will drop to below the plate as the data sheet and Valve Wizard site shows. I must admit I don't have much experience with small signal pentodes, just power pentodes where the screen is just a few volts lower than the plate even at idle. So don't worry about the screen being higher than the plate. If anyone has more insight to add, it would be appreciated.

As an added item, that book about preamps on the Valve Wizard site is a very good book in plain, easy to understand English. Very easy to understand with minimal math. Highly recommended.

[beatnik]

I found out i was missing the 1m grid leak resistor, hence the different readings with input shorted to ground. But with the correct input configuration i still have the screen grid at 130v and the plate at 105v

I have done some experiments with different plate and screen grid resistors

With 110k instead of 220k plate resistor the voltage on the plate goes up to 222v with the screen grid still at 130v

Went back to the stock 220k and changed the 1m screen grid resistor to 2m. the screen grid voltage drops to 80v but the plate rises to 180v

I am thinking that i am not getting the optimal voltage ratio between plate and screen grid because the power tubes are not inserted yet. I suspect as soon as the power stage is working the voltages on the ef86 will change again

I think that I will go ahead and wire up the power stage. As you confirmed I have more or less the right supply voltages so I hope I won't be burning anything

[beatnik]

I finally got some time to work on this project. Today have been tidying up the wiring and doing some more checking, then fired up the amp

Passed the smoke test and the voltages around the tubes looked fine. Done some guitar playing and I am very happy with the sound, really much better than expected ! When cranked the distortion is very smooth and thick, definitely what you would expect from a vox. Very good bass response as well. The EF86 is quite microphonic but i have a handful of these and will try them out to find a good sounding one without too much noises

I am now thinking about the next step. I have four more noval sockets available on the chassis and I would really like to expand the amp adding some more flexibility

Right now I am thinking about two things : an fx loop (i use a lot of effects) and a second channel, maybe with 12ax7 instead of ef86, and i would like this channel to be even more gainy and with more preamp distortion

I would really appreciate if someone pops in and contributes some ideas. I am studying more circuits, mainly early mesa boogie stuff, but i don't really know if these preamp stages could match well this vox-like circuit. Any suggestion would be highly appreciated

[beatnik]

Here's a couple of pictures of the amp as it is right now. Still quite messy but i definitely feel it has more potential to grow

[DRH1958]

I would think your Mesa Boogie preamp would work fine if it has two 12AX7 gain stages that are cascaded. The original schematic you posted has this configuration for the second channel. If you need more distortion, you could lower the plate resistor value. Also you could lower the cathode resistor to bias it hotter and possibly add a bypass cap to the cathode resistor if there are any that are unbypassed. You could then mix the channels with 220k-330k resistors like the attached Fender schematic with them highlighted in yellow.

Fender_Bassman-AA165-schematic.pdf

EDIT: not sure why the top 220k resistor isn't highlighted as it is on my schematic but it's the one just before the PI input.

[olddawg]

On the old 18 watt forum a lot of people were adding EF86 preamp tubes at one time. A common thing they did because of inherent microphonics with an EF86 was to use a suspended/ isolated tube socket and/or put a high temp O ring on the tube.

[beatnik]

thanks for your comments. i have found an ef86 which has an acceptable micro phonics but i guess it will have to be used for a while to see if doesn't drift

the ef86 channel sounds very nice but i am looking forward to do some more experimenting on this amp

as you can see from the pictures, i got an empty tube socket just next to the ef86 preamp tube. what if i use this one to add a gain stage or a tone control to the ef86 channel ?

then i have three more sockets and these will be used for a second channel 12ax7 based. would be really nice to add an fx loop to the amp, do you have any circuit to suggest ?

[danielcoakley2112yyz@gmail.com]

Hello everybody,

I got this tube amp chassis unit, model AO-50-3C. I found information that it was an Hammond produced amplifier mounted in an Everett organ. I couldn't find any schematic for the original circuit, but I could see many similarities with the Vox ac15 circuit. ez81 as rectifier and a pair of el84 for the power stage, a 12ax7 for the phase inverter and a combination of ef86 and 12ax7 for the preamp stages. I had been really missing a vox sounding amp in my setup so I decided this amp would have been perfect as a conversion experiment.

I started by carefully studying the vox schematic and comparing to the circuit, taking out some components that weren't needed and then wiring up the power supply circuit with new filter capacitors. The original circuit didn't have a filter choke so I have added a nice vintage partridge uk made

After testing the psu section for correct operation I moved on to the preamp stage, starting with the ef86 channel of the vox ac15, the so called normal channel. The original circuit used pretty much the same capacitors and resistors values so i didn't need to change much here.

Then moved on the phase inverter stage. This also had basically the same resistors values as the vox, but wired somehow differently so had to do more work

Before working on the power stage I wanted to do some voltages tests with and without the preamp tubes. I found out to be quite different from the vox schematic i am referencing to.

Without the tubes in circuit I have measured around 420V on the preamp tubes anodes

With the ef86 in circuit i get the following readings

plate (pin6) 50V
cathodes (pins 3 and 8) 3.5V
screen grid (pin 1) 68V

With the phase inverter in circuit I measure around 300V on its plates (pins 1 and 6) and 68V on the cathodes (pins 3 and 8)

The Vox schematic I am referencing to tells the ef86 plate should be at 90V and the phase inverter plate at 220V. The b+ voltage is also different, the schematic calls for 315V whereas my voltage is about 388V, although i only have the first two tubes in

I would really like to get some feedback on these voltage readings before moving on to the power stage. I am not really a tube expert, I have some building skills but my theory and understanding of circuit operation is quite limited. Some advice would be helpful in order to get to the final result without damaging components, especially the transformers. I have attached the Vox AC15 schematic I am following

Can you send me the schematic you used? And can you send me some pics of your completed amp? I'm doing the same conversion ànd am just getting started building amp