Magnavox 116 to Guitar Amp

[Cross773]

Hello,

I’ve come into possession of a Magnavox 116 and I’ve put together a circuit to convert it into a guitar amp head unit. I based it off of the Magnavox 142 (which is the newer version of the 116). This is my first crack at an instrument amp but I have years of experience building tube hi-fi stereo units. I’m hoping I could have some advice on this project and all comments are welcome.

I’ve attached all appropriate schematics for your convenience and I will alter my schematic and re-upload as time goes on.

Along with the usual adjustments to the power supply section, coupling caps, and various other adjustments, I’ve also done the following.

Notable Changes

· Two ¼” guitar jacks placed in Parallel

· Old input/phase inverter transformer and circuitry deleted

· Preamp and phase inverter section added from Magnavox 142

· Active tone control section added from here http://www.angelfire.com/electronic/funwithtubes/Amp-Tone-A.html (figure 14)

· Separate gain and volume controls added

· Power and Standby switch added


Concerns

1. I spliced in the tone control, preamp, and phase inverter circuit and I want to make sure I wired it correctly.

2. I am concerned about the voltage rating of the volume pot. It is difficult to find voltage ratings on pots and this part of the circuit will get much higher voltage than the gain pot.

3. I fear that I will have to use a new output transformer. According to the photofact for the Maggie 116 the power output should be 45W. However the output transformer has 3 taps at odd impedances. This was because this console would have had several speakers wired in parallel. Seeing as this is a guitar amp head unit and I want to maximize power output I figured I would need only one winding at 4 ohms.

4. I want to add a Line Out jack and I am wondering where I should splice it in. I figure from right after the tone control section. Also, if I should add a line out function would that mean I would have to add a switch that disconnects it from the rest of the circuit?

5. The power tubes are self-biased via the cathode (which I prefer) and I left it connected stock (albeit with a bumped up electrolytic). I usually have a separate bias network for each side of the push-pull power section but over the years I haven’t seen much value in doing that. If someone has good reason to think I should then I’ll separate the two sections.

6. I want to delete the two small inductors between the 6V6s in the power section as well as the 220mmf cap after the preamp tube. They don’t seem to do much and I would like to keep this build clean.

7. I would prefer to use 5AR4s in the power supply section rather than 5Y3s. They are basically drop in replacements albeit with much higher available current. The risk would be that something would burn out with all the extra power draw.



Thank you for your time and have a great day.

 

[Cross773]

For larger pictures I've uploaded the new schematic and the Maggie 142 to Photobucket.

Guitar Amp Schematic
http://i1036.photobucket.com/albums/a446/cross773/Maggie-Cross Guitar Amp V1_zpswqvoqlew.png

Magnavox 142
http://i1036.photobucket.com/albums/a446/cross773/Magnavox 142_zpsgtgrdcq2.jpg

 

[thorpej]

4. I want to add a Line Out jack and I am wondering where I should splice it in. I figure from right after the tone control section. Also, if I should add a line out function would that mean I would have to add a switch that disconnects it from the rest of the circuit?

You could easily tap off without having to disconnect other stuff, but you'll want to add an op amp or MOSFET buffer for it (in addition to the low output impedance needed for the line-out, you'll also want the very high input impedance of an op amp or MOSFET that will allow you to "tee" the signal off to the line-out without having to disconnect it from the rest of the amp.)

 

[gadget73]

1 you have the voltage amp at the wrong end of things. It should have a gain stage for the guitar, a gain stage going into the tone controls, a gain stage after to recover the loss, a gain stage at the start of the power amp section that feeds into the phase inverter, and the two output tubes. The volume control would go between the tone amp and the voltage amp at the front of the power amp section. If you don't need the extra gain after the tone controls, make it a cathode follower to feed into the volume control instead. You also don't want the feedback loop to include the guitar pickup stage or the tone controls. Basically this ought to look like the AMP-142 with things added in front of it.

2 No worries about volume control voltage rating. It sees nothing but a few volts of audio signal. No DC plate voltage is on it, and you would never want there to be anyway.

3 horribly unlikely it makes that kind of power. 20-25 watts is more realistic. Stabbing at it, your primary impedance is probably in the 5-6k range, assuming the schematic markings are right you should have a usable 8 and 16 ohm tap, plus one wierdo one. When in doubt, work out the transformer ratios and see what makes sense.

http://www.radioremembered.org/outimp.htm

short version: feed some amount of AC from plate lead to plate lead. Measure output voltage. Divide input by output, thats your turns ratio. Square that number and multiply it by the speaker load and thats what the plates will see. You can use any AC voltage you have handy, it doesn't really matter as long as you can accurately measure both the input and output voltages. I usually use my variac and the isolation transformer and wind it up until I get about 1 volt on the output but if you've got some random transformer laying around that does at least 10 volts or so, it'll work.

4 what thorpej said. op-amp as a buffer also works fine.

5 with tubes in parallel you do not want to individually bias them. Get a matched quad, or just install 2 6L6 tubes. It will work pretty much the same as the four 6V6

6 Don't. They're necessary for stability, unless you want to make it a 6L6 amp, in which case the inductors can go. You should still ring this out to make sure it doesn't oscillate or do anything else that is generally stupid since you have changed it considerably.

7 this will bump the voltages an easy 40v. Not wise unless you compensate for it on the output stage by re-biasing as needed. If you plan to drop the voltage with a big resistor, there is no point in changing the rectifier because you're throwing away the capacity. Honestly its not needed, a pair of 5Y3's can handle enough current to feed this.

 

[Cross773]

Thank you for your suggestions, I'm glad that I asked before I began the build.

You could easily tap off without having to disconnect other stuff, but you'll want to add an op amp or MOSFET buffer for it (in addition to the low output impedance needed for the line-out, you'll also want the very high input impedance of an op amp or MOSFET that will allow you to "tee" the signal off to the line-out without having to disconnect it from the rest of the amp.)

I am trying my best to make this an all tube build, and an op-amp from the pre-amp might be a bit complex. I was able to find this simple and fast addition that comes off the output transformer. Seems like it would be less sophisticated than the buffer route but much simpler. http://www.dougcircuits.com/lineout.html

1 you have the voltage amp at the wrong end of things. It should have a gain stage for the guitar, a gain stage going into the tone controls, a gain stage after to recover the loss, a gain stage at the start of the power amp section that feeds into the phase inverter, and the two output tubes. The volume control would go between the tone amp and the voltage amp at the front of the power amp section. If you don't need the extra gain after the tone controls, make it a cathode follower to feed into the volume control instead. You also don't want the feedback loop to include the guitar pickup stage or the tone controls. Basically this ought to look like the AMP-142 with things added in front of it.

gadget73, I am glad that you noticed this thread. I've taken your advice many times before over the years putting together old maggies and I have much respect for you.

From what I understand, it sounds like you are suggesting the following order - Input- gain stage - gain stage - tone control - gain stage - gain stage - power amp - out. I've taken cues from other guitar amp circuits and only made one gain stage that feeds into the tone control stage. I also thought that with an active tone control stage (using a 12AX7) there is little power loss. If I am mistaken or you think I need more gain stages (more tubes) please correct me.

Please excuse my confusion. I'm not sure what you mean by "AMP-142 with things added in front of it" and "you have the voltage amp at the wrong end of things". I thought the basic order is "Input - Preamp - Tone Control - Power (with inverter) - Out" which is what I tried to put together.

I'll make adjustments to the feedback loop.

Thank you so much for your response to each of my concerns, they are greatly appreciated.

 

[gadget73]

Just the way its drawn shows basically an amp142 with a tone control circuit between the voltage amp and the phase inverter. You really don't want that. Another reason for keeping what amounts to a 142 with the tone controls upstream is to make the negative feedback work properly. Without it, you'd have to feed it back in one stage ahead of the phase inverter, and the way its drawn that would end up on the other side of the volume control.

If you're using this for your tone control, according to the site its about -1db, so lets just call it unity for S&G purposes.

Stock input sensitivity on an AMP-142 is probably somewhere around a half volt. Adding the tone control in front of a 142 still means you need to feed about a half volt into the tone stack to get full power output. I don't pretend to know guitars but I suspect you'd need at least one more stage of amplification between the guitar and the tone control input to get there.

so all told, I'm thinking one gain stage for the guitar. One feeding into the tone amp, one recovering out of it. Output from tone amp to volume control. Output of volume control to the voltage amp stage of the power amplifier. Output from that point to one power tube and to the phase inverter. Output from the PI to the other power tube. Looks like a pair of 12AX7's and maybe a 6AV6 (its half a 12AX7) would do the job for you.

It may end up too clean for guitar use though. Honestly instrument amps baffle me. They aren't necessarily meant to just make a clean signal and I don't know what the right kind of wrong is. its possible you may want a level control between the output of the guitar gain stage and the input of the tone amp. turn that up and turn down the master volume to overdrive the tone stack without blowing your eardrums out.

 

[Cross773]

I'm reluctant to add any more gain stages for two reasons. For one, I have looked at old schematics for both Fenders and Bogens, and they both have only one gain stage. Other designs have two stages before going into a tone control circuit (either passive or active). Instrument amplification is baffling to me too as my background is in stereo music amplification. The second reason for my reluctance is that I am already running two extra tubes off the power transformer of the Maggie 116. The Maggie 116 does not have any 12AX7 tubes (just the two 5Y3s and quad 6V6s), and I am already betting that the power tranny can handle the extra two tubes that I'll need for the preamp, tone control, and PI. I have doubts that the tranny has enough juice to power the heaters of four or five extra tubes. If it comes down to it I can always add another transformer but I'd rather not.

 

[knockbill]

Since you mentioned Fenders and Bogens, and there are a lot more guitar amps out there... you may be best off looking at some well known guitar/instrument amp schematics, that could fit with the tube sockets you already have on the Maggy chassis,,,
I highlighted "well known", as I just finished building a small 6AQ5 PP "generic" guitar amp, that doesn't have enough gain for a guitar,,, it is also way to clean, can't get it to distort at any level, even without FB...
Turns out, a little researching on some guitar forums, explained how it was a nice little mono audio amp (which it really is), as it doesn't have enough gain stages (1/2 12AT7) suitable for guitar... It seems tried and true is the way to go, unless you are well versed in guitar amp design...
I've built a few clones, following the schems from various makers that sound absolutely amazing...

 

[gadget73]

Some of it may be related to feedback levels too. I don't know what the input sensitivity of just the power amp section on a proper guitar amp is, but it may well be greater than the Magnavox circuit is. Less feedback makes for a more sensitive power amp section. Power amp in this case meaning the part from the end of the tone controls through the speaker.

Did the 116 power a tuner externally? if so it probably has a spare heater winding you can repurpose for extra stuff. The five small triodes I suspect you'd need (two 12ax7 and a 6av6) will need 750ma.

Cloning an existing design is a good suggestion.

 

[Leland47]

Since you mentioned Fenders and Bogens, and there are a lot more guitar amps out there... you may be best off looking at some well known guitar/instrument amp schematics, that could fit with the tube sockets you already have on the Maggy chassis,,,
I highlighted "well known", as I just finished building a small 6AQ5 PP "generic" guitar amp, that doesn't have enough gain for a guitar,,, it is also way to clean, can't get it to distort at any level, even without FB...
Turns out, a little researching on some guitar forums, explained how it was a nice little mono audio amp (which it really is), as it doesn't have enough gain stages (1/2 12AT7) suitable for guitar... It seems tried and true is the way to go, unless you are well versed in guitar amp design...
I've built a few clones, following the schems from various makers that sound absolutely amazing...

Don't know squat about guitar amps....but was talking with a friend at work that builds little tube Amps out of tube radios. He said the "new" trend is to build "clean" sounding amps and do all effects from pedal (boxes)....

 

[Cross773]

Since you mentioned Fenders and Bogens, and there are a lot more guitar amps out there... you may be best off looking at some well known guitar/instrument amp schematics, that could fit with the tube sockets you already have on the Maggy chassis

I took inspiration from the old Fenders because they appear to be the classic kind of old style tube amp and their schematics are easy to find. I also have respect for the old Bogen PA systems - specifically the 100W Bogen CHB100 (please see schematic below), for which I've worked on. The CHB100 is designed for mic level amplification (which is in the same ballpark as guitar level) so I figured one gain stage would do the trick. These old styles tend to have one pre-amp section but then again just because it is old doesn't mean it's the best. I'm willing to add more stages to make this amp a success. In terms of some well known guitar amps the biggest name that comes to mind for tube amps is the venerable Mesa-Boogie. However, they make extensive use of solid state components to make their circuits work and their schematics are hand drawn which makes them difficult for me to follow (see example below). I'd be very grateful if you could recommend a good guitar forum so I can dive in and try to find a suitable clone.

Did the 116 power a tuner externally? if so it probably has a spare heater winding you can repurpose for extra stuff

The Maggie 116 did have a power umbilical (as seen in the above attached schematic) and the power tranny is the largest I've ever seen on any tube stereo (including my Fisher 500C which used to be my daily listener). So I'm rolling the dice that it can accommodate the extra load. It's a shame that you can't find the datasheet on any Maggie transformer. I also have an old Radio Shack 6.3V tranny which would be perfect for the heaters in the new additions and luckily there is plenty of space on the chassis as well (if it is needed that is).

I also had a thought earlier. If the sound is 'too clean' would I be able to cheat and simply put a pot on the cathode bias rather than a fixed resistor? Taking away headspace will cause the amp to distort earlier (however this will shorten the life of the tubes and will not address my possible sensitivity and loudness issues).

I've begun the process of tearing down the Maggie 116 and parts have been arriving for the new build. Hopefully I'll find some leads to make an update to this schematic.

 

[gadget73]

you really don't want to mess with the output tube bias. Its likely to roast the tubes if you try and bias it up too much. driving a small signal tube harder will get you there, or reducing the negative feedback may as well.

The difference in the CHB-100 is probably the tone controls. They may not be as lossy, or the amp may be more sensitive. Input to the front of the tone control seems to be essentially line level, probably about a half volt for full output. I'd expect about the same for the Magnavox power stage with the tone controls added. Don't pay too much mind to the total number of gain stages, its more about the total gain of the circuit. Those two tubes in the tone control you're looking at net out to be slightly less than unity gain when the controls are set flat. For practical purposes you can pretend its not there at all.

That leaves us with one triode gain stage for the guitar input, then the power amp which needs two small triodes to make the phase inverter work. I may be off base here, again, not experienced with instrument amps, but I really think that with the basic setup you've got picked its going to need 5 triodes to work as expected. Pulling the tone controls would put you at 3.

The Mesa has no negative feedback. It will not need nearly as many gain stages as a result. Guessing at it based on other Magnavox stuff, but you've probably got a good 10db of feedback, which is essentially a 10x reduction in amplifier gain.

 

[Cross773]

The Mesa has no negative feedback. It will not need nearly as many gain stages as a result. Guessing at it based on other Magnavox stuff, but you've probably got a good 10db of feedback, which is essentially a 10x reduction in amplifier gain.

I didn't realize the NFB loop takes away so much gain. Then again I never questioned the importance of it with my usual goal in mind being the purest, cleanest sound. Perhaps at first I wont add the NFB components (the .0022uf cap and 2.7K resistor if I understand right) and see if the amp is stable. Otherwise maybe a smaller resistor and cap would do the trick? I could also add a simple slider switch that disconnects the NFB circuit. That will give the user some flexibility for different playing scenarios.

I really think that with the basic setup you've got picked its going to need 5 triodes to work as expected.

To be honest I had always thought that two triodes for the guitar gain would be best. Part of what is holding me back (along with the power requirement concern) is that that would require an additional tube. This might sound trivial but if I do need another triode I would use another 12AX7 (rather than a single triode like a 6AV6). This is because a goal of mine for this build is ease of use for the customer. 12AX7s are plentiful, well known, and some people like to tube roll. With a 6AV6 there is less variety to choose from. Where my reluctance lies is that inserting another 12AX7 will leave one half of the tube wasted (unless I want to feed back the signal again through to the other side I guess).

 

[gadget73]

larger resistor makes for less feedback. Less feedback means more gain. it also tends to make the frequency response less flat but for a guitar amp that may not be a bad thing. Using a variable pot in series with the resistor would let you adjust it to suit.

you could do two inputs and use a 12ax7. half for each input. Or use the other half as a cathode follower to drive into the tone controls. That won't add any extra gain.

 

[Cross773]

Today I reworked the entire schematic for which I've attached below. You can also see the full size version here https://imgur.com/a/OSuRj

This version forgoes the active tone control circuit in favor of the much simpler passive controls from the Fender Bassman Model 5E6-A for which I've attached the schematic below and full sized version can be found here https://imgur.com/a/MtBqh

I spliced in the preamp and cathode follower circuit from the Fender while keeping the phase inverter circuit from the Maggie 142. While I was in there I saw that I forgot to add a wire which has since been corrected.

Using a variable pot in series with the resistor would let you adjust it to suit.

I may still remove the NFB loop but I discovered that they do in fact add a pot to the NFB and call that control "Presence". Not a control feature I'm familiar with but good to know that some designs make that a user controllable option. I might just do that but I want to avoid cluttering up the controls.

Notable Changes
-The active tone control stack from Angelfire was deleted and replaced with the passive controls from the Fender.
-With the now freed 12AX7 an additional gain stage was added along with a cathode follower for the tone stack.
-A simple Line Out function was added to the output transformer.
-A wiring error was corrected in the phase inverter.
-The NFB Loop was moved to the cathode of the phase inverter.

Concerns

1. I am not confident that I placed the Gain and Volume pots in the right locations.
2. I am glad I noticed my wiring error in the PI but if I can have confirmation that it is wired up right I would be grateful

I am much more confident about this build seeing as it is much closer to a proven working amp. I am disappointed that I couldn't keep the active controls but everything is a compromise.

 

[dcgillespie]

Cross -- Gadget has brought you a long way from your beginning concept, with everything moving in the right direction. Some additional comments from your most latest schematic:

1. Your volume control is shown backwards: Signal goes into the top of the control, the signal out is taken from the wiper. As shown will suck out all the gain and bass at low volume settings.

2. Your NFB scheme does not work -- It is only encompassing half of the the push-pull signal. You're missing the top half of the inverter stage, which the volume control should feed, and the NFB should be applied to.

3. You don't have enough gain as shown. With only 2 gain stages, it can work -- but only if you don't use any NFB. So either remove the NFB, or add the top half of the inverter stage as mentioned, and keep the NFB.

4. You're showing a direct connection from the tone stack back to the grid of the second preamp. If you're using the Fender tone stack, it looks like you're missing a 10 M resistor in that connection.

5. Good move on using the Fender tone stack -- hifi tone controls don't work well for musical instruments.

You're getting there!

Dave

 

[Cross773]

Dave,

Thank you for your help, I truly appreciate it. I have updated the schematic which I'll attach below and can be found full size here https://imgur.com/a/HJKbk

The NFB Loop has been removed, the volume was fixed, and the 10M resistor was put into place.

You're missing the top half of the inverter stage, which the volume control should feed

I'm starting to become curious about this phase inverter circuit. I'm familiar with both Cathodyne and LTP variety of inverters but this doesn't look like either. I tried my best to recreate the connections from the Maggie 142 schematic but it doesn't make sense to me how it accomplishes phase inversion. Anyway, if worse comes to worse I'll ditch it for the PI in the Bassman amp and perhaps add the other gain stage while I'm at it (effectively making this build a Fender clone which might not be a terrible idea).

Also, I had a thought earlier concerning my Line Out fix. Can the amp function safely without a speaker load and only the line out attached? I know it's a bad idea to leave a tube amp unloaded, and this circuit is particularly sketchy with no NFB Loop. Perhaps I can have a slider switch with an 8ohm resistor for the user to flick over when using the Line out function?

Other than that if everything looks good I think I'll be ready to start this build and we can see how it turns out.

 

[dcgillespie]

Cross -- The phase inverter is of the "Paraphase" type, and works on a most basic principle of vacuum tube operation: Any signal presented to a grid of a tube will be inverted by 180 degrees at the plate. So, from one push-pull signal, a portion of it is tapped off and precisely attenuated, applied to the inverter grid where it is amplified back to its original level at the plate, but inverted in phase, becoming the other side of the push-pull signal.

I would again suggest that you need to install the top section of the inverter circuit that is missing from your schematic for two reasons:

1. As shown, you only have two gain stages, which with low Gm output tubes like 6V6s, will be quite marginal unless the guitar signal is particularly hot. Cathode follower stages do not provide gain, and actually present a slight loss to the circuit.

2. In a push-pull Class AB arrangement, it's never a good idea to drive the output stage directly with a high impedance volume control. Besides the change in circuit response this will create at different settings, it will also accentuate the effects of blocking distortion as the limits of the power stage are reached.

Dave

 

[gadget73]

Paraphase inverters are actually so simple they're confusing. Its nothing more than two stages in series with a bunch of reduction between them so each has equal but opposite output.

http://www.valvewizard.co.uk/paraphase.html

if you'd like a little more reading on it. Its got some flaws that can be partly helped by making it a floating paraphase, or you can use an LTP or a cathodyne inverter if you like.

 

[Cross773]

Cross -- The phase inverter is of the "Paraphase" type

I would again suggest that you need to install the top section of the inverter circuit that is missing from your schematic

Ah it all makes sense now why it was looking so off to me. Thank you for giving me a heads up.

I went ahead and took the plunge by adding a third 12AX7. I completely re-worked the schematic and made some dramatic changes which I hope will make the amp better than ever. I have included the schematic below and the full sized version can be found here https://imgur.com/a/CV6fg

This updated build makes use of the classic Fender Bassman AA-865 for which I will include the schematic below and full sized version can be found here https://imgur.com/a/3dQKS

Notable Changes

- An additional 12AX7 has been added
- The paraphase phase inverter has been deleted and replaced with the LTP inverter from the Bassman AA-865
- With the now freed up triode I made another gain stage in between the tone stack and the phase inverter.
-The NFB loop from the Maggie 142 was reinserted
- I added a switch with a dummy load on the outputs for use when the Line-Out jack is in use. This was done to protect the OPT from burning up.

Concerns

1. Although not called for in the Bassman AA-865 I added a coupling capacitor (.022uf 600V) in between the third gain stage and phase inverter. It just felt right.
2. I am concerned about the arrangement of my volume control with the components around it. I'm also curious about the 500pf cap in the phase inverter circuit.
3. The NFB was placed before the phase inverter which I believe is correct (so it can encompass the full sound wave rather than half like I had it before.
4. I feel like their should be a bypass cap on the cathode of the third gain stage (like in the first and second gain stage).

I have come to realize that this is a sort of Franken-Amp at this point. I've been keeping track and it looks like this

1. The chassis, power transformer, and output transformer of a Magnavox 116
2. The schematic scaffold of a Magnavox 142 using it's NFB loop
3. The first gain stage, second gain stage, cathode follower, and tone stack of a Fender Bassman 5E6-A
4. The third gain stage and phase inverter of a Bassman AA-865

Once again, thank you all for your help and interest on this project.