• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

#26 pre amp

Hi Rod,

Can you explain your circuit for the electronically faint hearted. I'm only used to the LM1084 with adjust to the positive voltage, and I don't understand why R2 isn't 1.2R. I don't understand R1 at all!

I'm not Rod but R1 is the (simulated) tube filament, it even says so in the comment, ADJ is connected properly to the ground but R2 doesn't make any sense to me either (should be 1.2R IMO).
 
hi Andy - Yes, the circuit is an LTspice schematic - modelling the currents & voltages through R1, which represents the filament. And yes, R2 should be 1.2 ohm - I was trying diffeent values to get low voltage operation & forgot to switch it back!

The filament does not behave exactly like a resistor, of course... we have to keep this in mind when using simulation tools. But I thoroughly recommend LTspice for all of these kind of circuits - so quick to try out things you'd never get round to doing with the soldering iron. And it's free, and is designed for Linux use (via WINE)... DIYer's dream.
 
Rod,

Thank you for sharing that modified simpler circuit, I just have a few of queries.
Can one use a LM 350 or LT1086 instead?
Also can your circuit be placed in an outboard filament power supply or does it need to be placed near the tube? Finally would a 1.5A 7806 pre-regulator be sufficient.

Cheers,

Richard
 
Where would you connect the cathode resistor - LM1084 side or the other side or does it matter?

andy

Western Electric and recent designers on this forum have all used the negative side for the cathode resistor. The sound may suffer if you get it wrong.

I use two resistors for my 300B design (see 'new DHT heater' link above). They are calculated to give a voltage drop across each that is equal to the filament voltage.

Probably easiest to stick to the single R mounted on the -ve.
 
Rod,

Thank you for sharing that modified simpler circuit, I just have a few of queries.
Can one use a LM 350 or LT1086 instead?
Also can your circuit be placed in an outboard filament power supply or does it need to be placed near the tube? Finally would a 1.5A 7806 pre-regulator be sufficient.

Cheers,

Richard

hello Richard

The first thing to look at on the data sheet of these 3-terminal regs is the Noise - usually expressed as % of Vout.

LM350 offers 0.001 % vs LT1086 0.003% (typ) 10Hz..10kHz (bandwidth must be same for comparisons).

So it may well be a better part for the job... but only maybe. The dynamic behaviour of the feedback loop (internal to the chip) is very important too.

So LT1084/6 or LM350 meet the functional requirements, provided your heatsink is big enough, but you have to listen to them to be sure.
 
O, and I mount the power transformer and filter caps >2m from the amp (this is a low impedance circuit). Mount the CCS/gyrator right near the valve since it's a high impedance circuit, and may suffer pickup.

I think Andy has done some work on location of the modules for this kind of design
 
So LT1084/6 or LM350 meet the functional requirements, provided your heatsink is big enough, but you have to listen to them to be sure.

@Richard and others with regards to above comment: I used a couple of LM350 (supposedly Fairchild part, purchased via Farnell) in a project recently and it was as if the tab on the back side was stamped out of a tuna can :confused: Any LM317 (in same TO-220 package) I came across has tab more than twice the thickness of that LM350 and there's no way it would get bent by merely looking at it the wrong way (unlike that of LM350). I have never seen such a flimsy tab on any TO-220 device.

The result was that regulator kept shutting down due to overheating because that thin tab couldn't take the heat away and thermal contact between the heatsink and the tab wasn't good enough. Higher datasheet rating aside, 1A filament doesn't require 3A-specced regulator so I wholeheartedly recommend you guys NOT to use LM350, at least not one coming from same stock. What good is that rating anyway if device keeps shutting down at lower dissipation than LM317 ?
 
Hi Rod,

Thank you once again for the excellent advice, I should be able to squeeze it into the preamp chassis.

Do you think it would be beneficial to mount all the parts for this circuit (except the big cap) in a small sealed aluminium diecast box (also using it as a heatsink) and have leads out for the tube filament and voltage in?

As for cathode resistor position, I’m using the centre tap of 2, 20ohm resistors. Looking at a lot of the data sheets I’m at a loss as to which filament pin is the +and the -. Can anyone enlighten me?

Arnulf, It’s a pity you may have a bad batch of LM350’s. I just measured the mounting tabs thickness of my Fairchild LM 350T at 1.34mm and the LT1086 at 1.26mm. Perhaps the T version is a bit more robust.

Cheers,

Rich
 
Hi Rod,

Could you post a schematic just of the gyrator circuit? I'm not a star at reading solid state schematics...

I would be easier for me to see what the gyrator looks like, build that into my positive supply and build a ccs for the negative supply. Would that work?

Kind regards,
Bas

PS. for the ccs I am thinking of a cascoded IXTH20N50D.

Bas,

The gyrator is M1 plus R3/R4/C2 in the circuit I posted in this thread. That one is just an idea for those that don't like to build with bipolar transistors - I haven't tested it, although the simulation models just fine.

The schematic on this old thread:

http://www.diyaudio.com/forums/tubes-valves/38248-new-dht-heater.html#post446973

Has been tested and known to sound wonderful with 300Bs. The gyrator portion is Q1/T1 and the Rs/Cs around them.

I have updated this circuit a little, and will repost when I get a minute!

For sure, use the gyrator for the +ve, and CCS for negative.