Rack preamp redux

[Zozobra]

So I've been dicking aroung with my rack preamp for a while and for reasons I'll never fully understand I've decided to strip it down and rebuild it using a cool little development eyelet/pcb hybrid board that I've acquired. As an additional piece of tomfoolery, I have dispensed with a perfectly good toroidal Px to use a DC-DC boost converter. I'll be powering both this and the heaters from my bench PSU for now so I can get a good handle on the current requirement. Heaters will be wired for 12V series operation and everything will eventually all be powered from a small 1 to 2A SMPS.

The slight method to all this madness is that I want to take my developed designs and then make some PCBs. The whole ethos of these designs is that they will be cheap to build using 12V LED SMPS power supplies and the DC-DC buck converter, and that they will operate from any line voltage because SMPS, much like the honey badger, don't give a shit. The cost of both of those is less than £10 combined on ebay. A 2U rack chassis is £30-40.

Enough babbling, here are some pictures.





Since I intend to run some pretty high gain preamps up in this bitch, I'm going to use a MOSFET gyrator choke to clean the power supply right up as the DC-DC boost converter can sometimes add a little bit of chirp. Something like the circuit below will attenuate PS noise above 100 Hz by ~70 dB for very little voltage drop!



I'm on lockdown from work a the moment due to COVID-19 so as long as I have enough bits in my inventory I should make pretty good progress on this quickly. The final part of the puzzle is the output buffer. Usually I'd opt for a high voltage MOSFET (IRF820 or similar) here, but this time since I have a convenient 12V supply I'm going to use an opamp. I have a bunch of TL074 quad opamps so I intend to use one channel as the buffer, a second 2 channel active baxandal tonestack to simulate presence and depth controls (https://sound-au.com/articles/eq.htm - fig 5). I have 2 left over so one might be used as a mid boost and then the final one as a voltage follower to buffer the master volume to output. This is all low voltage stuff and can be breadboarded outside the chassis while I find what I like.

[Zozobra]

Testing the heaters with my bench supply. 0.44A is right on the money for 3 12ax7s running 12.6V series heaters. I've only ever used regular ass 6.3V parallel heaters so I was keen to test this before going much further.

[jgreenwd]

I think I want to build a good, regulated bench supply this summer.

[Zozobra]

You should totally do that. I've only had mine a month or so and it's mega useful. This is just a cheap one from eBay that is reasonably well reviewed on YouTube and whatnot not. Some of the cheap ones have issues with switching between voltage and current limiting/regulation but this one is fine. 30V and 10A is enough for my needs and I can always use boost converters if I ever need more or want bipolar. The ZVS flyback driver I'm using in the preamp will boost 12V to anywhere between 45 to 380V!

[Zozobra]

I've done a bit more wiring and tested the ZVS driver. It'll quite happily spit out 390V which is plenty. The real cool thing with these is that the output voltage is set on a multiturn trimmer so I can experiment with preamp voltages quickly. I'll post up some clips of a circuit using a reamp to showcase the tonal effect of different voltages as this is quite often overlooked. I also wired up the filter caps onto the main dev board but I want to rework that with the main reservoir cap and the gyrator choke on a mini breakout board to free up some space. Just waiting on some 10R 2W resistors and that'd good to go.

[Zozobra]

Caught an hour or two on this today in between zoom meetings.

ZVS board tested and working.


Gyrator choke breakout board added. Filter caps added to main board and the first tube socket is wired up.

[CrunchBerries]

Hey man, that looks like a very awesome and relevant build for the modern guitarist.

I've been looking into voltage regulation for tube amps, here is something that may be fruitful, hopefully..

http://paulrubyamps.com/PSMDesign.html

I was thinking of using the IRFP450 to regulate the voltage to a specific desired point. Not a huge relative difference from the IFR820 in terms of the application, but it does provide a bit of extra headroom that allows for a wider voltage selection range. That specific build offers different voltage references, but the circuit can be completely simplified by choosing a specific operating voltage, and the second B+ is obviously optional. I just noticed that your design did not include a reference voltage.

Cheers my friend!

[CrunchBerries]

I noticed that you already have that figured out, my comment is therefore maybe useful to others reading this, as you did mention you would use an irfp420 or similar. So yeah I didn't add anything you didn't already know lol...

Anyways, have you had a chance to test the performance of a mosfet regulator on a guitar amp? In terms of regulation performance itself and practical effect on sound?

I found a thread somehwere, a guy showed that the regulation was vwry efficient, solid at 405V for his design, but that there was heavier current on the reservoir cap.. how much of that is a concern?

[Zozobra]

I really need to to do an update on this thread I've built it up and pulled it apart since I last posted.

The choice of MOSFET isn't super important for what I'm doing here. IRF820 or 840 is more than adequate to act as a gyrator choke without a heatsink for the couple of mA a preamp will pull. Key thing here is that I'm not regulating the voltage here in this design. I'm using the MOSFET to clean up the power supply as I'm using an SMPS for the 12VDC which is then feeding the ZVS booster, so there is plenty scope for chirps and weirdness. A big fucking reservoir cap and an equivalent 47H choke will squash a lot of that. It may seem overkill but with it being a high gain preamp there isn't any common mode rejection in the poweramp which would normally tame things as the poweramp is using a separate power supply.

To answer you question: I've used a MOSFET regulator in a VVR circuit to control the poweramp voltage in a 2x6V6 amp I built for a friend and that worked real well. I have also used MOSFET regulated screen supplies to lock the screens at 400VDC in a 2xKT88 amp with 480VDC on the plates (4k Zp). This improves tube life and removes poweramp squish, which is great for authoritative high volume palm mutes but would blow for rock and more classic tones as it might sound a bit sterile. It's not hard to tryout if you have the space for a tag board and somewhere to mount the MOSFET on a heatsink. What is it you're looking to achieve?

[CrunchBerries]

lol..

Yeah man, i was wonderin.. can you do a center tap mosfet dropper and a mosfet voltage regualtor simultaneously? I want to take the voltage down 60 V at tye center tap, then another 60 at the regulator for 330V B+... right now PT supplies 444V, need it down bout 120V down ..

[CrunchBerries]

I really need to to do an update on this thread I've built it up and pulled it apart since I last posted.

The choice of MOSFET isn't super important for what I'm doing here. IRF820 or 840 is more than adequate to act as a gyrator choke without a heatsink for the couple of mA a preamp will pull. Key thing here is that I'm not regulating the voltage here in this design. I'm using the MOSFET to clean up the power supply as I'm using an SMPS for the 12VDC which is then feeding the ZVS booster, so there is plenty scope for chirps and weirdness. A big fucking reservoir cap and an equivalent 47H choke will squash a lot of that. It may seem overkill but with it being a high gain preamp there isn't any common mode rejection in the poweramp which would normally tame things as the poweramp is using a separate power supply.

To answer you question: I've used a MOSFET regulator in a VVR circuit to control the poweramp voltage in a 2x6V6 amp I built for a friend and that worked real well. I have also used MOSFET regulated screen supplies to lock the screens at 400VDC in a 2xKT88 amp with 480VDC on the plates (4k Zp). This improves tube life and removes poweramp squish, which is great for authoritative high volume palm mutes but would blow for rock and more classic tones as it might sound a bit sterile. It's not hard to tryout if you have the space for a tag board and somewhere to mount the MOSFET on a heatsink. What is it you're looking to achieve?

When speaking about 'poweramp squish'.. if I were to add series resistors to simulate sag (switched on/off), would that create that squish? Or what if I get the regulator's reference to be a simple resistor divider network (no zeners or caps)?

[CrunchBerries]

Do you think this will work?

Like this:

Solid rails with no budge:

Power.png

Sag rails with series resistors + floating reference:

Sag rails.png

*** Load is 2 x EL84 and 2 x 12ax7. About 70-120 mA total.

[Zozobra]

Since I last bothered to update this I've replaced the front panel and messed around with an op amp output buffer. Photos and clips soon. Maybe