Joyo BantAmp 12AX7 Tube 18V DC Power Supply Schematic

Started by bushidov, March 30, 2020, 08:49:20 AM

Previous topic - Next topic

bushidov

Hi All,

So, got on of those Joyo BantAmp Zombie's last year and I have to say, for the price, and pairing it with a Harley Benton 2x12 Celestion Vintage 30's cabinet, it rocks pretty hard. It states it puts out 20 Watts, which is pretty dang loud as far as my bedroom is concerned, as well as small gigs.

I decided to tear it apart, just to figure out what kind of circuitry it is using to drive it's 12AX7 section and then make a PCB to fiddle with it so I don't have to keep cracking open my Zombie. Below is the schematic I drew up and tested successfully.



So the gist of their setup is that the Joyo BantAmps have two boards: A base board that does the power supply work, and then a "daughter" board that houses the tube, and all the analog "emulation" to duplicate whatever sound they are trying to duplicate. Also, the amp comes with a 18V, 2A rated DC power supply (looks like a laptop brick supply)

Internally, their circuits are generally -> op-amp the signal -> feed into 12AX7 tube -> do some tone shaping -> go into solid state power amp, which in the case of the Zombie is a TPA3118.

So, how are they getting the voltages to drive a 12AX7? A nice SOIC-8 boost chip: XL6007E1. I made a PCB based around it that is driving a 12AX7 where the only surface mount component is this chip. All the other boost chips and circuits I tried kept creating a lot of noise that went into the circuit. This chip doesn't. I am guessing it is because of its higher than normal switching frequency for that level of boost.

On the Joyo, it takes that 18V and pushes it up to a hair over 50V. However, on my PCB, I was able to drive it at 12V and still get it up to 50V while keeping the same sound. It only drew around 200mA of current, so to drive this alone, the 2A supply is a bit overkill.

As of the heater element, that was the one that baffled me a bit. Joyo takes the 18V incoming signal and uses a 15V linear regulator (DPAK on theirs). Then they take what appears to be a 1Watt resistor rated at 12 ohms (1%) and that drops the voltage down between 12-13V. The linear regulator gets warm, but not "hot", but that resistor gets a lot warmer than I'd like. I noticed that the voltage was hovering around 13V volts, which I thought would burn out a 12AX7 heater (I thought the range was 6.3V to 12V), but I guess not.

Also to note from the schematic the R11 and R13 resistors. They are also 1 Watt resistors. All the rest are 1/4 Watt. As of capacitors, make sure C7, C9 - C12, C17, C19, and C20 are rated for high enough voltages to handle that 50V rail.
"A designer knows he has achieved perfection not when there is nothing left to add, but when there is nothing left to take away."

- Antoine de Saint-Exupéry

PRR

> a 12AX7 heater (I thought the range was 6.3V to 12V), but I guess not.

Or you could read the sheet. Not "TO" but "OR". 6.3V wired one way. 12.6V wired the other way.

Wired for "12" as it surely is, current is nominal 0.15A. In 12 ohm resistor, 1.8V drop. Assuming 15V supply, that puts 13.2V on the heater, 5% high. The RCA sheet allows +/-10%. The real goal was car-power which could hold 14.4V at cruise. Power in resistor is 0.27W. In a 1 Watt resistor that may be warmer than I'd like to lick, but I don't see an engineering objection. (Especially since Joyo does not have a warranty??)
  • SUPPORTER

phasetrans

Thanks for the layout. I find it funny how similar this is to what I would have guessed. Right down to the xlsemi boost converter.

Does the rest of the circuit have any other hard clipping elements in it?
  • SUPPORTER

bushidov

Thanks Paul for those clarification. I'd knew one of you folks would have the answer for "why" in regards to that.

QuoteDoes the rest of the circuit have any other hard clipping elements in it?
From just the pure elements of the power supply setup, no. But there did appear to be a set of diodes in there, but without really going into it hard, I couldn't tell if it were hard clipping of soft clipping. There appeared to be several TL072 surface mount opamps, and the diodes were close to them.




Here's the base board where the power supply portions are at:

"A designer knows he has achieved perfection not when there is nothing left to add, but when there is nothing left to take away."

- Antoine de Saint-Exupéry

phasetrans

#4
Quote from: bushidov on March 31, 2020, 06:15:37 AM
From just the pure elements of the power supply setup, no. But there did appear to be a set of diodes in there, but without really going into it hard, I couldn't tell if it were hard clipping of soft clipping. There appeared to be several TL072 surface mount opamps, and the diodes were close to them.

Personally, I would try and chase down everything around D2/D5 and Q2/Q3.

Do you have a 12AU7 lying around by any chance for a swap? I've been wanting to do something like this Bantamp for several years, with a moderate plate voltage, and the tube driven arbitrarily hard by an opamp at the beginning. The 12AU7 is known to do well with low plate voltages.
  • SUPPORTER

bushidov

QuoteDo you have a 12AU7 lying around by any chance for a swap?
Actually, I do not. However, if memory serves, and I could be wrong (PRR Paul, help?), the 12AU7 would like that 50V plate voltage.
"A designer knows he has achieved perfection not when there is nothing left to add, but when there is nothing left to take away."

- Antoine de Saint-Exupéry

amptramp

I'm a little at a loss with the power supply schematic.  R4 is 150K in series with the switching diode but the power supply is loaded with 11K at the output.from R5, R6 and R7 meaning only 6.8 percent of the generated voltage shows up at the output with no load and load doesn't pull it any higher.  Am I missing something here?

bushidov

I am not saying I designed it, just that is how Joyo did it.
"A designer knows he has achieved perfection not when there is nothing left to add, but when there is nothing left to take away."

- Antoine de Saint-Exupéry

Vicc

Greetings Bushidov!

I too happen to have bought the JOYO BanTamP Zombie last year! Finest device I've acquired in quite some time... Nothing's perfect though as JOYO doesn't include a comprehensive manual with schematic and all; that's how I found this forum... I'm your typical newbie at the "breadboading" phase of my very first 'DIY pedal' and although I've double-checked everything, neither gain nor distortion is present... Thing is, someone told me at another forum that the problem may lie in the fact that my "preamp" isn't enough to drive my amp so,

Would you happen to know the Zombie's actual input impedance?

Also, reading what you posted before, I couldn't find 1/4-watt resistors for my circuit and ended up using 1/2-watt ones... Could that be affecting the circuit's purpose overrall?

bushidov

Actually, I noticed a flub in my schematic and made me a JLCPCB version of, so I know it works now. Here is the corrected schematic


I drew the 50V rail on the wrong part. This schematic has it corrected and I built a PCB to verify it.
"A designer knows he has achieved perfection not when there is nothing left to add, but when there is nothing left to take away."

- Antoine de Saint-Exupéry

jma38

Quote from: bushidov on March 30, 2020, 08:49:20 AM
Hi All,

So, got on of those Joyo BantAmp Zombie's last year and I have to say, for the price, and pairing it with a Harley Benton 2x12 Celestion Vintage 30's cabinet, it rocks pretty hard. It states it puts out 20 Watts, which is pretty dang loud as far as my bedroom is concerned, as well as small gigs.

I decided to tear it apart, just to figure out what kind of circuitry it is using to drive it's 12AX7 section and then make a PCB to fiddle with it so I don't have to keep cracking open my Zombie. Below is the schematic I drew up and tested successfully.



Thanks for doin the schematic, I have one too and I was fiddling around with the idea of doing a version with a subminiature tube preamp section. Might take a while to design though.

So the gist of their setup is that the Joyo BantAmps have two boards: A base board that does the power supply work, and then a "daughter" board that houses the tube, and all the analog "emulation" to duplicate whatever sound they are trying to duplicate. Also, the amp comes with a 18V, 2A rated DC power supply (looks like a laptop brick supply)

Internally, their circuits are generally -> op-amp the signal -> feed into 12AX7 tube -> do some tone shaping -> go into solid state power amp, which in the case of the Zombie is a TPA3118.

So, how are they getting the voltages to drive a 12AX7? A nice SOIC-8 boost chip: XL6007E1. I made a PCB based around it that is driving a 12AX7 where the only surface mount component is this chip. All the other boost chips and circuits I tried kept creating a lot of noise that went into the circuit. This chip doesn't. I am guessing it is because of its higher than normal switching frequency for that level of boost.

On the Joyo, it takes that 18V and pushes it up to a hair over 50V. However, on my PCB, I was able to drive it at 12V and still get it up to 50V while keeping the same sound. It only drew around 200mA of current, so to drive this alone, the 2A supply is a bit overkill.

As of the heater element, that was the one that baffled me a bit. Joyo takes the 18V incoming signal and uses a 15V linear regulator (DPAK on theirs). Then they take what appears to be a 1Watt resistor rated at 12 ohms (1%) and that drops the voltage down between 12-13V. The linear regulator gets warm, but not "hot", but that resistor gets a lot warmer than I'd like. I noticed that the voltage was hovering around 13V volts, which I thought would burn out a 12AX7 heater (I thought the range was 6.3V to 12V), but I guess not.

Also to note from the schematic the R11 and R13 resistors. They are also 1 Watt resistors. All the rest are 1/4 Watt. As of capacitors, make sure C7, C9 - C12, C17, C19, and C20 are rated for high enough voltages to handle that 50V rail.

anotherjim

The opamps may have feedback diodes, but with a series resistor so they don't limit so hard but merely drop the gain. Or, some prefer head-to-head Zeners for a high voltage swing without the opamp itself clipping. If I went to the trouble of using a tube, I wouldn't want its distortion character overshadowed by diodes or opamp distortion.
Also, if the tube runs at a higher plate voltage than the opamp supply or its safe input voltage range, input protection diodes for the opamps would be wise and many products add the protection anyway if an opamp takes the input signal first.


puretube

@bushidov: your corrected schematic labels: "L2 = 27uF".
Make that: "L2 = 27uH". :icon_wink:
(before s.o. starts to wonder ...)

bushidov

Quote@bushidov: your corrected schematic labels: "L2 = 27uF".

I'm not seeing an L2?
"A designer knows he has achieved perfection not when there is nothing left to add, but when there is nothing left to take away."

- Antoine de Saint-Exupéry

puretube

Quote from: bushidov on July 02, 2022, 07:56:52 AM
Quote@bushidov: your corrected schematic labels: "L2 = 27uF".

I'm not seeing an L2?
I marked it here in your schemo of reply#9:



Owenbrah

Just want to revive this topic...
Like a complete rookie I plugged my bantamp into a 4ohm cab and fried it.
See attached pic. It works as a pre amp into an interface or headphones
Any ideas on fixing it? Or am just dreaming?


antonis

"I'm getting older while being taught all the time" Solon the Athenian..
"I don't mind  being taught all the time but I do mind a lot getting old" Antonis the Thessalonian..

FiveseveN

Dang, do these things not have any sort of thermal protection? :o
By the looks of it you'd be better off building a separate poweramp (or buying a preassembled module). Put it in the same box if it fits but that crispy PCB seems like quite the chore to repair.
Quote from: R.G. on July 31, 2018, 10:34:30 PMDoes the circuit sound better when oriented to magnetic north under a pyramid?

tonyharker


FiveseveN

TPA3118
You can get modules super cheap (though I'm using AliExpress as an example, not a recommendation): https://www.aliexpress.com/i/32685893614.html
Quote from: R.G. on July 31, 2018, 10:34:30 PMDoes the circuit sound better when oriented to magnetic north under a pyramid?