Megatron Electrostatic Headphone Amplifier

[jose]

I'm not very good at describing sounds but you can imagine. Megatron has tubes in his pre-stage and this conditions su sound.

[mwl168]

My Grounded Grid sounds very similar to my Blue Hawaii. The Megatron sounds quite different from those two. I agree with Jose that the all-tube front end contributes a lot to the different sound signature. The Megatron is also capacitor-coupled rather than DC-coupled. 

[jose]

Between the GG and the BH there are differences of "nuances", but both sound very similar. I can´t do a direct comparison with Megatron. They are very different animals.

[Oneguy]

Thank you for the quick replies guys. My curiosity got the better of me so I figured I’d reach out to the community for help. Unfortunately there’s not a way to demo DIY amps before you buy them unless you know someone that has them. 

[eggil]

build one of each

[mwl168]

I spent most of this past weekend working with JimL and Kevin tweaking my Megatron.( I was just the coolie equipped with a soldering iron and a multi meter and Jim and Kevin supplied all the brain power. ) Kevin suggested that I share the information here.

My goal was to raise the standing current to the plate of the driver tube (12AX7). It's set at 0.5mA in Kevin's original design. Kevin has mentioned in his June 28, 2012 post that the only thing stable at 0.5mA is the 2SA1968 which has since been obsolete. My build uses KSA1156 in place of the 2SA1968 so this tweak may be more applicable to similar builds. 

This tweak, at the end, turned out to be the few relatively simple steps below:

1.       Reduce R6 and R7 (2.2K) to about 1.3K which sets the plate CCS at 0.85mA. I used 3K to parallel with the 2.2K to get to this value. This higher current also raises the standing plate voltage of the 12AX7 from around 245VDC to about 275VDC (this will vary some with the tubes used).    

2.       Reduce R11, R23 cathode resistor (3.3K) to about 2K so that it pulls slightly more current and brings the plate voltage down to around 245VDC.  I used 4.7K to parallel the 3.3K to get to 1.99K. We need minimum 50V differential between the plate and the B+ (300VDC) to allow for the driver voltage swing without clipping.

3.       Because the resulting higher current draw of the front end, you need to adjust R19, the drop resistor, to maintain the 300VDC for front end B+.

I like my Megatron’s sound better with this tweak. YMMV.

It may be possible to raise the driver current even higher subject to the plate dissipation and CCS transistor limit. I am happy and content with the outcome now but may experiment further in the future. 

I am also hoping someone may be interested in running a SPICE model and/or observe the outcome with a scope to validate.

Edited August 13, 2018 by mwl168

[jose]

I have homework for the weekend… ??

[JimL]

So, just a little technical explanation if anyone is interested. In the Megatron design, the driver 12AX7 and EL34 outputs operate within a global feedback loop with the gain set by  the ratio of R21 (33) to R11+R12 (23+24). The operating conditions for the 12AX7 are set by the constant current loads Q1 and Q2 and the grid-to-cathode voltage which is set by R11.

 

If you look at the tube characteristics, for a set grid-to-cathode voltage, as the plate current increases the plate voltage will also increase. However, because the plate current runs through R11, this makes the grid-to-cathode voltage more negative, which further increase the plate voltage. The result is that the plate voltage gets closer to B+, which can result in premature clipping of the driver stage. So in order to increase the plate current without increasing the plate voltage, we must decrease the grid-to-cathode voltage. This is done by decreasing R11. Note that this will also increase the overall gain and decrease the closed loop bandwidth, however, since R11 is only 1/3rd the value of R12, a 10% change in R11 will only result in a 2.5% change in gain and bandwidth. If we wish to maintain the overall gain and bandwidth, then we need to increase R12 so that the sum total of R11 and R12 are unchanged, .e.g if we decrease R11 to 2k, R12 should increase to ll.3k.

[mwl168]

It may worth mentioning that, based on the plate characteristic chart of the 12AX7, this higher plate current also puts the tube in the more linear operation zone. Whether the effect is audible is debatable as always. 

 

[jose]

I made the mod.

Modify R6 and R7 at the same time as R11and R23 with the values indicated by Mwl.

My data are now:

12AX7 volt plate: 270vDC

Voltage Before R11: 12.7vDC and between R11 and R12: 11.7vDC

I have not modified the resistance that manages B + 300vDC but now it is at 290vDC. Should I change the value of the resistance to get to 300vDC or can I leave it like this?

I need some time to assess everything but in principle everything looks good.

Edited August 12, 2018 by jose

[kevin gilmore]

something is wrong, with .85ma and 300v, should be 240 volts

that is only 20 volts now to B+, the stage is likely to clip

[mwl168]

13 minutes ago, kevin gilmore said:

something is wrong, with .85ma and 300v, should be 240 volts

that is only 20 volts now to B+, the stage is likely to clip

Jose's mod sets the CCS at 1mA instead of .85mA. This is likely the reason why the 12X7 plate voltage is quite a bit higher than the 245VDC I am seeing on my amp. 

[kevin gilmore]

then you need to lower the resistance of r11 even further.

[mwl168]

4 hours ago, kevin gilmore said:

then you need to lower the resistance of r11 even further.

Yes, and, as JimL stated, may also want to increase R12 to maintain (R11 + R12) close to 13.3K to maintain the gain and bandwidth as originally designed .  

EDIT: attached Megatron schema for reference.

megatron.PDF

Edited August 13, 2018 by mwl168

[JimL]

Jose,

If you are using R11 of 2.0k, that is why your plate voltage is too high. I suggest dropping R11 to 1.5 or 1.6k. Then if you want, you can increase R12 so that the sum of the two is 13.3k. However, R12 shouldn't affect the plate voltage to any significant degree, it is R11 that is the prime determinant. You want to aim for a plate voltage around 240 volts.

Edited August 13, 2018 by JimL

[jose]

Thank you. All the tips are well received of course.

Okay, last night I was also thinking about lowering R11even more but I don´t know until what value I can lower it. Maybe 1K ?... I have right now is 1.95K so I'm going to try to lower this value up to 1.5K easily.

I'm thinking about Jim's comment about the total value of R11 + R12. Do I have to compensate R12 obligatorily? On the other hand, change R12 you say that should not significantly affect the plate voltage.

Edited August 13, 2018 by jose

[JimL]

I would start with changing R11, so that the plate voltage is around 240 volts. This will ensure that the driver stage does not clip. If you decrease R11 alone, that will increase the gain, as it is part of the feedback network. Increasing R12 to keep the sum of R11 and R12 the same will maintain the same closed loop gain, but it will only change the plate voltage by the amount of voltage change in that resistor alone. In other words, if you have 1 mA plate current, and you change R12 from 10k to 11.5k, that will only change the plate voltage by 1.5 volts.

[jose]

Thanks Jim for explanation 

Changed R11 for 1.65K and I had around 245~250vDC in all plates. I was not happy and I tried with 1.5K, now I have 235~240vDC in all plates with 1mA plate current. So I guess that now is more correct. Right?

Now I need to change R21 because I have 290vDC and I would like to have all correct voltaje. 

[JimL]

!.5k for R11 is fine. R21 is the feedback resistor. If you want the front end voltage to be around 300 volts you should decrease R19, which is the drop resistor.

[jose]

1 minute ago, JimL said:

!.5k for R11 is fine. R21 is the feedback resistor. If you want the front end voltage to be around 300 volts you should decrease R19, which is the drop resistor.

Sorry Jim I means R21K 

[mwl168]

2 hours ago, jose said:

Sorry Jim I means R21K 

My bad. I gave Jose bad information. The drop resistor is R19 which is marked 21K on the PCB. I got those two numbers mixed up...

[JimL]

No problem. Dropping R19 to around 19k should do the trick. Also, if R12 is increased to 11.8k, the gain structure will remain unchanged.

Edited August 14, 2018 by JimL

[jose]

On ‎8‎/‎7‎/‎2018 at 8:25 PM, mwl168 said:

I like my Megatron’s sound better with this tweak. YMMV.

It's been a little over a week since the mod finished and I agree with MWL comments. I think it's worth doing or trying it before making a final configuration. 

Now I am testing some valves for the first stage. The process is still very interesting (as always). I think I like to have valves with a higher Mu close to that of an ECC81

Thanks Kevin, MWL and Jim 

[steefdebruijn]

Hi all,

After a l.o.n.g time first start of my Megatron, fully bare version for now. Scary as hell but no cracks, fumes, funny smells etc. Pfew. I did check PSUs with variac first. Did measure filament AC on trafos first and DC out on the PSUs unloaded. Did them loaded with some load resistors a long time ago also.

Amp board full power on first launch with all tubes in (after triple checking the wire mess, that is ? )

Some operational parameters

• Balanced & matched 12AU7EH and 12AX7EH tubes
• Matched octet 6CA7EH tubes
• Bias +570VDC (on testpoint)
• HV +-457VDC after delay of about a minute after filament power on
• 300V line is about 265-270VDC
• all 4 leds are glowing
• all filaments too ? 
• LV +-12VDC
• Shorted inputs gives a steady 19VDC on one output and a jumping 20-25VDC on the other output
• HV regulator heatsink L-bar not warm at all (about 15 minutes uptime)
• Bias regulator heatsink also not warm at all
• LV regulator heatsink about 60 degrees Celsius (seems a bit hot to me)
• AC input on LV section 2x15VAC

Questions

• Is bias voltage OK (a bit lower than 580)
• Is 300V line OK (or should I lower the 21K resistor value to get closer to 300VDC)
• Is +-457VDC OK (is a bit higher than 450)
• are the output DC offsets OK (especially the jumping one)
• should I do something about the LV section (maybe a larger heatsink although I have a high 60mm version, or get me a lower secondary transformer)?
• should I check some other parameters...
• ... or should I just plug in a Stax and listen ?

Thanks so far,

Steef

[jose]

15 minutes ago, steefdebruijn said:

Questions

• Is bias voltage OK (a bit lower than 580)
• Is 300V line OK (or should I lower the 21K resistor value to get closer to 300VDC)
• Is +-457VDC OK (is a bit higher than 450)
• are the output DC offsets OK (especially the jumping one)
• should I do something about the LV section (maybe a larger heatsink although I have a high 60mm version, or get me a lower secondary transformer)?
• should I check some other parameters...
• ... or should I just plug in a Stax and listen ?

 

I answer based on the little knowledge I have.

1. Yes 

2. You need lower resistor a bit 

3. Yes

4. offset is measured with each channel with respect to ground. I'm not sure what you're doing (I think you're measuring the balance)

5. I use a large heatsink. maybe it would be convenient for you to think about that when you do your chasis. 

First check the offset, if it is low and the Bias is correct you should not have problems connecting your stax.

 

Edited November 13, 2018 by jose