Audio Asylum Thread Printer
Get a view of an entire thread on one page
|For Sale Ads|
I ask because my pair 2a3 suddenly tested from medium good to borderline in a month.
My friend suggest I should give it up because the complexity of the power supply but I don't think I am ready to throw the towel yet.
My question is if I can tune the 2A3 to suggested operating condition, let the driver tube and 2a3 filament light up first with separate tranny, delay the B+ for above 1 min, at the same, use tube voltage regulation to ram up the B+ from zero to 425v. Will the nature of such kind of circuitry , I mean direct couple, the higher plate voltage shorten the tube life? Is there anything I need to address?
Long story short, I have been feeding my 2a3 with 5 v. It is all good now but one pair 2a3 bit the dust. OUCH!
For me, this is the most challenge project. In the beginning, to build and set the power supply to handle the need of direct couple design like slow turn on, adding a relay, etc, there were quite a few things to considerate. during the bead boarding phase, to get the tube bias correctly, it is like fitting a tight fitted bed sheet, pulling on one end then ended up losing other side.
I managed to have everything under control and dialed the operating plate voltage and current to the right point except one night after switching back and forth different amps, I had the jumper of the filament accidentally attached to the wrong slot. Lesson well learnt.
Dude, we all screw up from time to time...if you don't, you're not doing anything. I once connected a VT52 UP BACKWARDS and then only realized it when the $40.00 cathode bypass cap started popping...
The valve survived luckily...
We have all made such mistakes I am guessing at one time or another.
When you get new 2A3s, try running them at 10 to 10.5 Watts plate dissipation, by reducing their current to 41 to 43 mA. They usually will sound great and last 10s of thousands of hours. Drummerwill and Rage use these op points, and enjoy their 2A3 DC amps.
5v on a 2A3? Your option would have been to make a parallel 2a3 and wire the fil in series. We have all done crazy stuff in our old age when we hit 35.LOL
When problems become increasingly difficult,we can always rely on our collective knowledge and seek relief.
FWIW, I had an amp with switchable filament voltage for 2A3 or 300B, some years ago (late 90s?). I ran the Sovtek 300B (yes, accidentally! but they were fairly fresh at the time) at 2.5v for quite a while; it sounded pretty good actually. Measured 2.78v RMS. I gave it up after a while, there was not enough space charge margin to protect the cathode, but an interesting experiment. Running 18 watts plate dissipation and 55mA current - ol' Jeff would approve!
just a suggestion : install separate switch for heaters and start your PS with a choke.
this will : ensure sufficient heat inthe tubes before high voltage is applied, and help prevent cathode stripping; and the choke input will ensure "slow" turn-on.
I have a separate filament transformer for the driver and 2a3, basically I am doing what you have suggested, but I use a Omron delay instead of a choke. I feel more comfortable using a delay because in case of momentary power outage, that Omron will re-set automatically.
Is 425 VDC the highest voltage in the circuit?? What is the VDC (B+) that feeds the output transformer?
What does the 2A3 plate dissipate as your amp is configured??
What is the VDC after the output transformer, on the plate of the 2A3, at the tube socket?
What is the VDC on the grid of the 2A3, at the tube socket?
What is the VDC on the filament of the 2A3, at the tube socket, and what is the Ohmic value of your self bias cathode resistor for the 2A3?
Do you use solid state or tube rectification?? What type?
Solid state will require some design work to cope with "instant on".
Do you use any thermistor to reduce turn-on surges?
NOTE : I am familiar with DC 2A3 amps that give tens of thousands of hours of good use to 2A3s, and it is safe to run up to 480 VDC on the plate (maximum) of a VINTAGE 2A3, and, up to 510 VDC on the plate of a JJ 2A3-40, without arc-over on turn on, using conservative current draw (41 to 43 mA.) and about 245 VDC P-K. Define yours !!
What is your driver tube, and its op points?? Do you have a schematic with operating voltages you can share??
How sure are you that you were measuring the output tubes properly and uniformly both times??
There isn't enough info to make any suggestion.
It is 6sl7 // dc to 2a3. The setting is pretty what the book suggestded, 250vdc plate voltage, bias -45 +/- 2v, the plate current is around 55-60ma, depends on the condition of the tube. B+ is 425 +/- 3, plate ref to ground is 420v, filament ref to ground is around 170vdc.
I guess I am within the parameter except making one careless mistake. Wrong filament voltage!!
Lets take this opportunity for me to make some design comments on what I see is not optimized with your amplifier, in terms of op points, and the DC couple. This is all just my humble opinion.
You already know, IMHO, the best op point for the 2A3 is 245-250 VDC at 41 to 43 mA, which is usually -49 to -50 VDC of self bias. This is THE best trade off I know for long life and unstressed, relaxed sound. 'Takes good internal wiring to optimally pull off.
What you listen to the MOST in a DC two stage amp, is the DRIVER tube, and its op point has to be NAILED and optimized up the whazoo. The 2A3 is acting more like a unity coupler. You MOSTLY hear the driver tube.
In your case, with 420 VDC on the 2A3 plate, and 170 VDC on the cathode, this means, for -45 VDC of self bias, you have ONLY 125 VDC on the 6SL7 plate. This 125 VDC Ea is simply too low, too unoptimized, to have a brilliant sonic presentation. The too-low of an op point will forever hinder the ultimate level of your new amp's sonic performance. Ohh, it will pass a signal, last forever, but it will NOT sound as good as it can.
I find the absolute lowest B+ on the 2A3 plate in a DC 2A3 amp is 450 VDC, and I really prefer 480 VDC on the plate of the 2A3. This allows you to have a B+ on the driver tube's plate of about 175 VDC minimum, and sometimes as high as 195 VDC (maximum) which gives a far more brilliant presentation of the music than your present value of 125 VDC Ea on the 6SL7.
It is paramount to get the operating point of the 6SL7 driver NAILED and optimized. The tube data sheets suggest 250 VDC Ea at 2.3 mA. and I usually will try to keep it close to the Ea, but lower the Ia to develop the best sonics overall. Obviously, I think HALF of 250 VDC (125 VDC Ea) is too low on the 6SL7 plate for optimal sonics.
Additionally, your self bias on the 6SL7 needs to be OVER 1.5 VDC, and likely over 1.75 VDC for optimal sonics. If you have efficient speakers, one 6SL7 section, at low current, will usually sound better than a parallel driving a 2A3, as it will avoid a time smear from two not identical sections.
One 6SL7 section will easily drive the 2A3 in a direct couple. One can assume that a 6SL7 in parallel will outperform a single section, and I say, YOU better listen to both and decide.
The wire between the DC needs to be really good, try three in parallel, one run of Kimber AGSS, one of Kimber TCSS, and one run of teflon coated Percy 21 AWG solid silver. That works very well.
The final 6SL7 bias point you alone need to determine BY EAR, by inserting a 10 turn pot as the Rk, and adjust the bias till it sounds the best to YOUR ear, often it will end up at 1 mA. or less, as per all the Loftin White professional designs I respect, Isamu Asano, Nobu Shishido, and Dennis Fraker's. Replace the pot's Ohmic value with two fixed value Rks in parallel, matched to .1% if you can measure that.
Have fun, a Loftin White (two stage DC 2A3, etc.) is a rewarding design, which can perform nicely. I've heard it do so, many times, and I prefer a two stage DC amp above the other tube amps I have made.
OOOOHHH, I just noticed you have an adjustable high Voltage B+ supply, maybe you can get another 25 to 70 VDC out of it !!!
I recall using a neon lamp on the driver tube as a voltage clamp in a DC circuit before.
I have direct coupled D3A to 2A3 SE amplifiers. I use 6AX4GT damper tubes for the rectifier. The 6AX4GT provide low voltage drop and a B+ delay.
In direct coupling, it is very easy to put a high voltage on the output tube grid during the warmup transient. Some tubes are fairly tolerant of this abuse, and others are very touchy about it. I have seen tubes that withstand it for years of daily use, and tubes that die in a day. It's still abuse.
Worrying about these transients is one of the prices you pay for direct coupling. Done carefully enough, there is no reason such circuits should not have a very long tube life. I am not claiming that I have always done it carefully enough - how else would I have learned what's in the paragarph above?!
Incidentally, once the transient is eliminated, sometimes the tube's emission will come back after a few hours of running.
Hallo Paul... can the Bottlehead 2A3 kit be ordered with a 240v primary?
Seems like I remember seeing a schematic with a tube damper diode following SS rectifier for surge limiting or am I wrong?
I suspect that the damper diode is used as a time delay element (to delay B+ tirn on) and not for surge protection. Damper diodes such as the 6DE4 take a long long time to warm up - something greater than 20 seconds, if I recall correctly. Which brings up a point. If you need a damper diode for delay, why not get rid of the SS diodes and replace them all with damper diodes?
Post a Followup:
Post a Message!
This post is made possible by the generous support of people like you and our sponsors: