Tube Preamp Project: advise and discussion

[thermionix]

12V x 0.15A (amount of current 12ax7 draws)

Considerably more at startup.

[anotherjim]

I was referring to the 1.6VA used backwards as the "step up" transformer and it definitely isn't supplying the filament. To put the incoming AC supply step-down transformer inside the pedal means having to deal with potentially dangerous power - we don't encourage that for amateur/DIY work, although there are and have been plenty of commercial guitar pedal products with the AC wall supply on board.
If you use a commercial 12vAC power adapter, you entirely bypass the problems and dangers of producing that part yourself.

[johngovan1234]

My first plan is to use the backward transformer setup having the main power supply outside like how tube pedals work. (eg. Kingsley) but what is more advisable to use in terms of space saving and noise or hum or interference? Will backward transformer will produce a better result or smps?

[amptramp]

I would go with the backwards transformer as the lower noise solution.  But if the plate voltages you are running are not that high, you can use a Cockroft-Walton voltage multiplier directly off the filament windings.  This would allow you to not bother with the second transformer in the unit.

[printer2]

> Dangerous to do work with AC? And the dc going to the plate is of no concern?

Connection to a wall-plug can burn-down the house, or at least scorch the workbench-- current available is very large.

Plate supply is still lethal but less dangerous.

Not too sure about this one. At least if we are talking about North American voltages. A 120V service is going to give 170V peak potential. Whereas a tube amp (I was talking more tube amps than what is needed for this circuit) can supply 50-100mA easy. That is in the kill range given that voltages are 250-450V. And with the AC line voltage the current gets interrupted 120 times a second. DC is continuous and there is no current interruption that gives you a chance to break the circuit. I have been zapped by 120V many times in life with line voltage. You work with it enough stuff happens. But other than feeling crappy afterward I have walked away.

Thankfully I have not been zapped much with higher voltages (only twice, big photocopier transformer putting out couple of thousand volts and resetting breakers in an electrical storm), 208 V and above are the real dangerous ones in terms of people and not being able to get up and walk away from. Not that I advise people being cavalier with any voltage above 50V but if I am going to be poked I would prefer it be 120 V than any higher voltages.

[printer2]

I have used switching 12 Vdc power supplies, it is rare for transformer filled adapters to be made today, for the filament and switched up the high voltage with the module shown earlier. No audible noise given a reasonable about of gain. I have been won over by the little switching power supplies and no longer use transformers in my amps and pedals. It is easier finding a switched power supply now days than an adapter that puts out 12V AC. You can go in some 12V dc adapters and remove the rectification but most adapters are sealed in a way where you destroy the seal going in.

[Jeema]

I was referring to the 1.6VA used backwards as the "step up" transformer and it definitely isn't supplying the filament. To put the incoming AC supply step-down transformer inside the pedal means having to deal with potentially dangerous power - we don't encourage that for amateur/DIY work, although there are and have been plenty of commercial guitar pedal products with the AC wall supply on board.
If you use a commercial 12vAC power adapter, you entirely bypass the problems and dangers of producing that part yourself.

Agreed - dealing with wall voltage directly requires very careful attention to safety due to the high current potential.  The safest way is to use a 12VDC or 12VAC adapter and step up internally.

Let's also not forget that even with isolated HV DC you can still potentially electrocute yourself hand-to-hand if you don't obey the "one hand" rule when messing with energized circuits and capacitors.  In fact anyone messing with HV anything should make sure they're aware of ALL of the safety considerations first.

[johngovan1234]

I am aware how we need to be careful enough when we are working on HV devices. I think i will use a backward transformer since i do not need big transformer for 150ma of current to make the tube pedal work. But the only downside of it is i need to use an external AC 12v powersupply rather than a typical dc 12v power supply.

[rankot]

I'm finishing testing of 5670 based bass preamp, it sound nice, but I didn't make it in pedal format. Sound's good, but maybe too sharp compared to 12AX7 based Alembic F2b.

[MJ_Sound_Cubed]

My design would be based in the following reasoning after seeing all the posts in this thread:

- Being aware of the extreme care held when dealing with mains AC voltage and high DC voltages, I would choose to only deal with one of them DC.

- I would run the SMPS and the tube heaters from a standard 9V DC PSU because makes the pedal board simpler and if you forget the AC transformer, there is no problem. Generally we can assume that the 12AX7 operates between 5,5V to 13V AC or DC, 150mA to 300mA, this means that they work at 1,9W. At 9V they consume 200mA.

- With the HV SMPS I posted the output, besides being very very quiet, is adjustable.

The block diagram would be like this:

[rankot]

Just quick note - the best way to provide correct voltage for heaters is to use a resistor. If you have 9V DC supply and 6.3V heaters requiring 300mA, then you must put 9.1 ohm resistor in series with heaters.

[MJ_Sound_Cubed]

Indeed sir, a 10R /5Watt resistor, thanks, forgot about the series parallel wiring of the 12AX7 filament . Just a note, the 9V in heater in series works well the tube just glows slightly less and is a bit grungier. 

[johngovan1234]

i think i will try a fender preamp first. here is my current ugly layout without power supply.
how to smps supply with this layout?

[rankot]

Indeed sir, a 10R /5Watt resistor, thanks, forgot about the series parallel wiring of the 12AX7 filament . Just a note, the 9V in heater in series works well the tube just glows slightly less and is a bit grungier. 

2W rated resistor is quite sufficient in this case, or you can use two 18 ohm / 1W resistors in parallel.

[johngovan1234]

will this smps work?

[rankot]

Breadboard and try

[pravudh]

I do use back to back power transformer 12VAC 300mA. it work great without problem. I did try SMPS  it have noise issue. 

[johngovan1234]

I do use back to back power transformer 12VAC 300mA. it work great without problem. I did try SMPS  it have noise issue.

Amazing work! Do you have a demo of this sir?

[amptramp]

Exactly as expected, the SMPS does generate more noise than a reversed transformer when implemented as a boost converter.  There are other topologies for an SMPS which include resonant converters and current-mode converters that are much quieter.  Also, with the current drain of a couple of 12AX7's, you can make a boost converter work if you push the frequency up into nearly the megahertz region, but an NE555 will definitely not do that.  I like the idea of a voltage multiplier rectifier array (Cockroft - Walton) because it saves the space, weight and expense of a transformer but the transformer is simpler, uses fewer parts and gets better regulation.

One idea that you might look into: biasing the heaters to a positive voltage.  Small parts of the heater may protrude above and below the mica insulator in the tubes that hold the electrodes in position.  The heater is capable of emitting a small amount of current but if it is biased positive, the protruding part of the grid will completely block the current flow.  This has been shown to reduce hum by a slight amount.  Some tubes are more of a problem than others from this point of view.

Although it is not a wall wart as such, there are doorbell transformers that plug into a standard power socket and provide a nominal 10 VAC or 14 VAC output.  The advantage of a doorbell transformer is that there is so much leakage inductance designed into it that shorting the output will not cause damaging or dangerous currents to flow.

[vigilante397]

I have a whole "line" of SMPS-based tube pedals and have never had issues with noise. I also use a 6V regulator with a schottkey between middle pin and ground for 6.3V as I don't want to cram a 5W resistor into my pedals, I can get regulators much smaller. I use Russian submini tubes for everything because I like to keep everything in 1590B's, so space is a bit of a premium.

Also cheers to cab42 on mentioning the Omnibox It's a little more than a preamp, but all it would take to turn it into a straight F-2B preamp is a jumper. That being said it's an SMD build, so not everyone's favorite thing