Tube Amp Tele - High Voltage Edition

[Jolly]

Hey All!

This build thread is going to be a sequel to my last build. I'm going to build three this time, one left handed and two right handed.

The last build was mostly a proof of concept; I wanted to see how everything would fit together and where the potential issues were. I didn't know what I didn't know about designing a self-amplified guitar. It turned out pretty decent so it's time to take what was learned from the first build and put together a more refined design. This build is going to focus on using top quality and readily available materials, preferably common to the guitar community. The individual specs aren't all worked out yet, but here's what I've got so far:

Amp: I'm going to use a single 12AT7 tube and a Fender reverb transformer for the output. The low voltage tubes sounded great, but I'm a little worried about their availability and they are a little power hungry for the volume you can get from them. I'm shooting for a little more headroom and volume with this build. I've mocked up an amp already but its relatively uninspiring still and needs some major tweaking.

https://www.tdpri.com/threads/battery-powered-single-12at7-build.1081567/

Body/Speaker Cab: That might be misleading because I'm going to try and ditch the speaker. It was one of the heaviest components and the large hole in the pickguard is asking for trouble. I've done some preliminary tests with surface transducers and think I'll be able to make that work instead. This setup will also make room for a neck pickup! The open back can be a little annoying and occasionally catches on things. I'm going to move the battery to the edge of the body and generally alter the construction/layout a bit

Neck: I'm still going to go with a premade neck to save a little time and setup. I'm going to get it unfinished though to give my own finishing process a go. I've been looking at best guitar parts for this but haven't really started buying parts yet.

Progress To Date: Besides lots of planning I've bought some wood already. It was sourced from the Architectural Salvage Warehouse of Detroit, highly recommend walking though if you ever find yourself in the area. These are 2x4's salvaged from a deconstructed home in Detroit, likely pre 1930's based on the dimension of the lumber and surface finish. I don't know the exact species of pine or anything like that, but I rummaged through the pile until I found some exceptionally light examples that had a nice ring. The boards were mostly de-nailed but there were a few stragglers to remove still. I also scored some thick maple stair treads and a bit of mahogany trim for a rainy day.

Wish me luck!
 

[MiKro]

Jolly,

Glad to see you trying the exciter direction. I too am working on a design that uses the exciters mounted to a plate for speaker. This should give you some flexibility as to tune the output some for better bass response based on location on the plate/soundboard. The problem you may run into though is the feedback, so that will need to be addressed. In some cases though it may also act in some way as a sustainer.    

I think I mentioned in your last build that you may want to look at the Cockroft Walton Voltage multiplier circuit. Using diodes and capacitors you can get a lot of forward voltage that may suit your needs as far as upping your voltage requirements in a light small package.

May I suggest, you also look into using Lipo RC type batteries as a power source. Smaller with as much power and less weight maybe. .

MK

BTW Congrats on the GOTM win.

[Jolly]

Thanks!

The benchtop exciter tests have gone pretty well so I'm optimistic they're going to work out. I didn't really keep track of signal phase in the first build but my hunch is that I can use the mechanical "feedback" in the system similar to negative feedback loops common in amplifier design. I'm going to make this a 2 channel amp: a clean one that has negative mechanical feedback, and a lead one that has an additional gain stage switched in. That should also flip my mechanical feedback to positive, ideally adding sustain and controllable feedback, like you'd get standing in front of a cranked amp. 

I'm not exactly sure how it does it but this is the power supply I've been playing with. It can convert 12V to ~220V. My test amp is pretty noisy, I don't think its the power supply but I haven't ruled it out yet.

https://omnixie.com/products/nch6300hv-nixie-hv-power-module

I took a look at some Lipo RC batteries. For now I like the clean packaging and quick swap setup drill batteries allow but I think the Lipo's might be good for some future, more power hungry/space limited projects I have planned.  Thanks for the tip!

[MiKro]

Keep up the work, trial and error are part of the electronics when it comes to being unconventional in purpose. LOL

MK

[curtisa]

Have you tried out the amp side of things yet? 20K effective plate resistor with a 130R cathode R on a 12AT7 at 200-ish volts has the tube pretty much hard on the end stops. I would've thought any overdrive you'd be getting would be pretty 'farty' sounding.

Bear in mind that boosting the plate voltage up using a DC-DC converter still makes it a power-hungry circuit on batteries. All you're effectively doing is selecting 4th gear on the gearbox instead of 1st.

[Jolly]

I've played around with the amp a little bit. It passes sound but it doesn't have as much volume as I'd like and the overdrive is pretty much non existent. The two potentiometers in the picture adjust the cathode resistance of each stage. With that, I've played around with different settings and nothing was close to "it". Even using the built in clean boost from original tele the most I could get was a mild crunch and it was still a little off. 

My next move is going to be a topology change. I'm going to try to run both stages of the tube in parallel for the power amp and use an op amp or two for the pre amp. 

The DC-DC converter certainly isn't the most efficient, but I think its power draw is still pretty mild compared to the heaters. 

[MiKro]

Looking at the specs on that DC buck converter,  I don't see any for frequency.

I might be concerned about the frequency it is running at? The others I have seen that produce that much voltage variance seem to be in the 20Khz range not the usual lower frequencies. Just something to look at for reference and rule it out.

MK

[Jolly]

The information on that converter was sparse for sure, its worked for the proof of concept though.  I've thought about designing my own power supply to get a little more voltage and have control over all the specs but its hard to pass on a plug and play package.

I took a trip to visit my parents last weekend and got all of the wood processed into rough blanks.

I'm still not 100% on the finished specs yet but here's what I'm thinking as of now.

I'm pretty confident in my plan for this one. This board is pine or fir or something similar that's been sitting in my dads garage for ages, I have no idea how exactly how old it is or where it came from but I've been thinking it would make a good guitar for years now. It's going to get a Baltic birch plywood top/back and tweed wrap similar to the last build. I'm thinking it's going to get a p-90 for the neck pickup and a mild to somewhat heavy relic job. I've yet to try and relic anything but I love the look of a beat up tweed amp and I think it will transfer to a guitar well. 

This one is going to be all wood salvaged from Detroit. The body is going to be a glue up of 2x4's and the top is from the maple stair treads. My experience is somewhat limited but this was the hardest maple I've ever worked with. The end goal for this one is going to be a double bound, 3 tone sunburst finish with a standard tele neck pickup.

This one is also going to be all wood salvaged from Detroit. The body is another glue up of 2x4's and the top is also from the maple stair treads. This one is going to be a little more experimental but I'm going to try and keep it a little rougher and really show off the "reclaimed". Ideally I'm going to sand the top and show off all the layers of paint from over the years. There's at least 3 or 4 different colors along with some cool wear, nails holes, etc. The back will be clear coated, possibly even leaving the saw marks and aged wood look. This one will also get a standard telecaster neck pickup.

The tweed one is going to be right handed for sure, I'm not sure which of the Detroit ones is going to be the lefty yet. 

 

[curtisa]

Have a look at the driver circuit used by Fender for their reverb stages (such as in the Fender Twin Reverb). The cathode resistor used in them is much higher, and the two halves of the 12AT7 are in parallel as per your thoughts on possible changes to the topology.

If you're exploring the idea of using both halves of the 12AT7 as the output driver another option is to configure it for push-pull configuration. The Firefly project used as 12AU7 in push-pull for sub-1 watt output. You could probably sub in the 12AT7 and it would be happy enough to work OK.

[Jolly]

I'm going to save push-pull for a later project. All of the off the shelf transformers I've found so far are a little too big. The next iteration will certainly be heavily inspired by the Fender reverb driver circuit.

I've gotten a little more "wood work" done. I used a chisel to scrape off as much glue as possible from the top of the stair treads. That was followed with a little bit of sanding.

Before:

After:

I'll likely end up doing a little more sanding before its done but this is pretty close to the look I'm going for. I wanted to get them roughly cleaned up and "flat" before cleaning up my glue joint.

There's a pretty big difference in thickness between what was the front and back of the stair tread as shown above. There was certainly some variation in my splitting cut but most of that is from wear, these were some well used stairs! I'm going to attempt to get the two top sides (paint side) as flat as possible to each other, then come back and flatten the back relative to that.

Hoping to end up with something like this:

[MiKro]

@Jolly,

I had a thought about this in one of those waking moments. YOU are going for HIGH VOLTAGE on this one. That all sounds cool but I question the safety of it.

Since the player is the only true earth ground and  if you have any short that would feedback voltage it will go through the player to ground. Just as a properly grounded chassis is done to prevent a deadly shock. The player becomes the chassis ground. I would be very hesitant to go this route unless you can some how isolate the high voltage from the player.

Just a thought about safety from my view of things.

MK

[Dave Higham]

On 11/11/2021 at 5:38 PM, Jolly said:

Hoping to end up with something like this:

How about an etched metal pick guard. Copper perhaps?

[Jolly]

On 12/3/2021 at 1:10 PM, MiKro said:

I would be very hesitant to go this route unless you can some how isolate the high voltage from the player.

Safety is certainly a major concern. I know a few individuals that are very knowledgeable about this subject that I can have double-check my final schematics for safety.

I believe I have a few things going for me:

1. The voltage is high but there isn't a lot of current (100ma max). That's certainly enough to do some damage to a person, but would likely require a major and total failure

2. I think the battery will somewhat "self contain" my power. The player may be the only direct link to earth ground, but there is no positive side link to ground to compleat the circuit like a typical amp connected to grid power.

By all means if anyone has further ideas on this or suggestions for safety measures to put in place let me know!

[curtisa]

Design safety (ie the schematic) is only a small component to making the build safe. Arguably the larger contributor will be how you make it. I don't think it's an impossible task, but it will be tricky. 200V @ 100mA at the very least will hurt. 

Battery-powered portable high voltage tube stuff did exist in the past (radios, TVs, test equipment etc). Maybe look into some of the construction techniques used in them for ideas as to what made them safe (ish) to use.

 

2 hours ago, Jolly said:

but would likely require a major and total failure

Plate-to-grid short on the first tube stage could put a hundred volts or so on the pickups. The string ground connection could come loose and touch something carrying HT volts. Admittedly rare failures, but this is also a rare build.

Also consider that while this circuit is ungrounded and floating, it's not this that inherently makes it difficult to make safe. You could have a failure that puts the HT supply on some metallic part of the guitar you're in contact with, but because you're also ungrounded it doesn't become apparent until you reach over and touch something that is grounded and get a zap that way.

[nakedzen]

Just to join in on the choir of caution, 100mA is 10 times the limit for current to contract your muscles involuntarily, meaning once you get zapped, you won't be able to let go of the guitar without help.

Quote

Currents above 10 mA can paralyze or “freeze” muscles. When this “freezing” happens, a person is no longer able to release a tool, wire, or other object...

https://www.elcosh.org/document/1624/888/d000543/section2.html

[Jolly]

On 12/9/2021 at 3:50 PM, curtisa said:

Design safety (ie the schematic) is only a small component to making the build safe. Arguably the larger contributor will be how you make it.

Lucky! I'm slowly getting the electric side of things figured out but I'm much more mechanically savvy.

On 12/9/2021 at 3:50 PM, curtisa said:

Plate-to-grid short on the first tube stage could put a hundred volts or so on the pickups.

I don't have a schematic put together yet but I've been scheming about a topology change. I think it's going to use a TL072 op amp for the preamp and the 12AT7 in parallel for the output. The TL072 should act like a "fuse" for a plate to grid short before any power leaves the amp section.

On 12/16/2021 at 3:45 AM, nakedzen said:

Just to join in on the choir of caution, 100mA is 10 times the limit for current to contract your muscles involuntarily

Another consideration is 100mA is the max the power supply can produce, the circuit itself should only draw 20-30mA. This may still be dangerous, but we're slowly chipping away at what could go wrong. 

I had a brief conversation about this with a close friend that is an electrical engineer. He suggested a fuse between the battery and power supply. He also proposed sacrificing a power supply or two by dead shorting it to make sure my fuse blows as intended. With a little test and tune I should be able to get something that runs the amp fine but limits the worst case scenario current significantly.

[curtisa]

10 hours ago, Jolly said:

I had a brief conversation about this with a close friend that is an electrical engineer. He suggested a fuse between the battery and power supply.

The better solution would be to fuse the output of the power supply. Placing the fuse between battery and power supply protects the battery. Placing the fuse after the power supply protects the power supply and by extension, you too.

Fusing the PSU output doesn't help with the second hypothetical failure I mentioned above though (disconnected string ground or other ungrounded part of the guitar coming in to contact with the HT). Such a failure wouldn't blow the fuse, no matter how sensitive it is. Your only recourse in that situation is to mechanically design the circuit such that the potential for such a failure is minimised in the first place. Look in to things like using heavy duty terminals that positively grip the wires rather than solder connections direct to circuit boards, providing strain relief on any wire leaving the amp or power supply, or using shrouded/insulated terminations on wires that may be unplugged.

I'd be tempted to mount the whole thing inside some kind of non-conductive box within the guitar and consider the contents inside it the 'danger zone'. Put all the nasty bits inside this box, and design it such that anything inside it can be allowed to fail without coming into contact with anything on the other side of the walls of the box. Anything leaving this danger zone to the outside world (eg pickup leads, speaker wires, 12V battery supply etc) can then receive the focus of your attention on how best to make as mechanically robust as possible.

 

[Jolly]

On 12/17/2021 at 10:36 PM, curtisa said:

I'd be tempted to mount the whole thing inside some kind of non-conductive box within the guitar and consider the contents inside it the 'danger zone'. Put all the nasty bits inside this box, and design it such that anything inside it can be allowed to fail without coming into contact with anything on the other side of the walls of the box.

That's the direction I was headed from a physical layout/construction angle anyway! I'd like to be able to assemble the guitar and amp separately and connect them as a final assembly step rather than build the amp in permanently. Along with electronically isolating the amp I'm hoping a mechanical separation from the body will help minimize microphonic tubes.

 

Here's roughly the space I have for a "safety box". It's going to have the tube, transformer, power supply, and various passive components mounted on it. 

I'd like it to look like the front control cavity cover but on the top back side. I'm thinking about Richlite for the insolating box/cover but I haven't really looked into all my options yet. It should also be able to fit the on/off switch and some vent slots so you can see the tube glow. I haven't started doing a detailed plan yet but I think this is the general layout I'm going to try. 

On 12/17/2021 at 10:36 PM, curtisa said:

Fusing the PSU output doesn't help with the second hypothetical failure I mentioned above though (disconnected string ground or other ungrounded part of the guitar coming in to contact with the HT).

This layout has the added bonus of generating and containing all of the high voltage in the upper section of the guitar. This puts gravity on the side of keeping mechanical failures away from high voltage.