OK...My 813 SE has been dropped in favor of a much more 'practical' design...the 2A3.
Note if you will the 500 watt HV B+ supply, the six amp supply for the two 2A3s' , the 1.2 amp supply for the lonely 5691, and the sixty watt 2A3 cathode resistor...I do not want any heat generated ....thermal runaway and all that jazz.
I am aware of the B+ noise issue (Diode switching) but I really do not want tube rectification......Voltage drops thru tube rectification really don't do it for me & LCLC strings seem too................dare I say complicated as I'm trying to keep it "simple".
I have the green light to start...as I explained my strategy to my wife...the chassis comes first with the cheapest components next and on up until completed.
Most all parts are coming from one place with a handful coming from other sources.
_____________________________________Rick.........
Note if you will the 500 watt HV B+ supply, the six amp supply for the two 2A3s' , the 1.2 amp supply for the lonely 5691, and the sixty watt 2A3 cathode resistor...I do not want any heat generated ....thermal runaway and all that jazz.
I am aware of the B+ noise issue (Diode switching) but I really do not want tube rectification......Voltage drops thru tube rectification really don't do it for me & LCLC strings seem too................dare I say complicated as I'm trying to keep it "simple".
I have the green light to start...as I explained my strategy to my wife...the chassis comes first with the cheapest components next and on up until completed.
Most all parts are coming from one place with a handful coming from other sources.
_____________________________________Rick.........
Attachments
Ok...great feedback, why are tantalums "bad" , are they 'noisy', do they have an inherent poor quality for the signal path?
The two 2A3s' are JJ Teslas. Do you all have better choice for the 2A3s?
The 5691 is the red base version.......
It is a simple execution, with the high prices going where they should go...the Plitron OPTs' being a good chunk of change.
___________________________________Rick..........
The two 2A3s' are JJ Teslas. Do you all have better choice for the 2A3s?
The 5691 is the red base version.......
It is a simple execution, with the high prices going where they should go...the Plitron OPTs' being a good chunk of change.
___________________________________Rick..........
Nebraska Surplus Sales is seriously overpriced for a lot of the parts you need.
Try handmade electronics, Angela Instruments, diycable, madisound, digikey, mouser and AES for a lot of these parts - you'll save some money.
Tantalum caps are bad for a couple of reasons, the two most important ones for you are that they are pretty nonlinear with audio across them - generating measurable thd in many instances, and they are potentially unreliable as well - they catch on fire when they short, which is their most common failure mode.
My employer has banned tantalum caps in all new designs because of the risk.
I also don't understand the all of the extraneous connections in your power supply, the extra switches, and etc..
Tube rectification has the benefit of being kind to capacitors and tubes during the warm up phase, and is less troublesome and costly to implement than soft turn on circuitry. I recommend one or the other with DHT.
Given JJ current QC issues I would be hesitant to recommend their 2A3. Sovtek makes a pretty decent sounding 2A3 for a lot less money and it is reliable last time I checked. If you really want the hot tube you should check out the TJ 2A3 - very nice if a bit expensive.
Try handmade electronics, Angela Instruments, diycable, madisound, digikey, mouser and AES for a lot of these parts - you'll save some money.
Tantalum caps are bad for a couple of reasons, the two most important ones for you are that they are pretty nonlinear with audio across them - generating measurable thd in many instances, and they are potentially unreliable as well - they catch on fire when they short, which is their most common failure mode.
My employer has banned tantalum caps in all new designs because of the risk.
I also don't understand the all of the extraneous connections in your power supply, the extra switches, and etc..
Tube rectification has the benefit of being kind to capacitors and tubes during the warm up phase, and is less troublesome and costly to implement than soft turn on circuitry. I recommend one or the other with DHT.
Given JJ current QC issues I would be hesitant to recommend their 2A3. Sovtek makes a pretty decent sounding 2A3 for a lot less money and it is reliable last time I checked. If you really want the hot tube you should check out the TJ 2A3 - very nice if a bit expensive.
kevinkr said:
If you are going to cap couple, then the cap you use makes a huge difference. Indeed, I'd say you would be better off with a $10 cap and a $10 driver tube than with a Tantalum cap and an expensive driver tube. Indeed, there are a lot of caps on the market that sound pretty good and that are not overly expensive. I have a page with some that I have tested that might give you a starting point: http://www.ecp.cc/cap-notes.html
Finally, I don't think that your expensive driver tube is worth the money. You will do much better finding something that is not in every other commercial amp, and which thus has not had the price driven up. Try, for instance, a 12V version of the 6SL7, or the loctal version, or try the single triode version, or the 12V single triode version, or branch out even more and try a 5842 or 6688 or EC8010. If you like new tubes, how about 6c45 of 6h30, etc. The point is that there are lots of cheaper choices many of which have more drive capability and sound at least as good, if not better.
You should check out Valve Art 2A3s. They are great sounding and are very affordable. I don't understand why you want to use a 60w cathode resister. Any thing that large is going to be wire wound and not the best. If you want to use something in that wattage range get the NI power resistors from Michael Percy. These are aluminum body with tabs so you can bolt them to the chassis for heat sinking and they 1% 50ppm.
Richard Ellis said:
Hi Rick, for less than $200, I got a pair of RCA bi-plate 2A3. But that's here in the Philippines, and it is the going rate. The Sovtek 2A3 is pretty hard to beat for the price you'd pay. My friend also use Valve Art, and again, they're so good at that price range. No personal experience with the Tesla, but I was talking to a friend yesterday who was pissed with the brand as almost every year, he has to change tubes when they go bad.
I own the 5691 red base too, which I originally use for my 2A3. Got them for less than $80, for a pair, NOS. Now, they're in my bin, and lonely, as I've replaced my 2A3 drivers with 76 which I got for $10 each. Better over-all tone than the 5691, for me.
The money you'll save from the 2A3 will buy you a nice Auricap, or even a Jensen for C2. I'm just not comfortable with an expensive tube *then* Tantalum coupling cap. To further, you will still have enough change to even afford a Black Gate standard for C3.
I'd say you try 330K for R5, and a Kiwame brand for R2.
You seem to be operating the 2A3 in a conservative mode... You can go with a wirewound 50W 820R R7. For this part, I'm not really that anal, as I've got NOS high power resistors and aluminum clad to compare. For the sake of reliability, I went for aluminum clad.
Considerations
OK...All opinions in....It seems unanimous, No Tantalums coupling the two stages, I checked the coupling caps page, more to research there is in order.
It seems that the JJ Tesla is not the cats' meow I guess ?, I recall reading somewhere that 'JJ' had moved its manufacturing to Russia, perhaps that is so and as a result thier QC has suffered.
I will definately rethink my 2A3 choices, maybe Valve Art or Sovteks.
I should stay with the circuit (Tube compliment) as designed so as not to have to redesign the whole thing all over.
The PS circuit as written has a switched warm-up procedure.
The 'standby' has all three TXs' up & running, The B+ is disconnected on the secondary, The two filiment TXs' are running thru dividers set for appx 63%."start" has the B+ running thru a PWR resistor to start-up the caps, the Filiments now have full power, "On" now just bypasses the start-up resistor & of course the filiments are still at full power.
_____________________________________Rick.................
OK...All opinions in....It seems unanimous, No Tantalums coupling the two stages, I checked the coupling caps page, more to research there is in order.
It seems that the JJ Tesla is not the cats' meow I guess ?, I recall reading somewhere that 'JJ' had moved its manufacturing to Russia, perhaps that is so and as a result thier QC has suffered.
I will definately rethink my 2A3 choices, maybe Valve Art or Sovteks.
I should stay with the circuit (Tube compliment) as designed so as not to have to redesign the whole thing all over.
The PS circuit as written has a switched warm-up procedure.
The 'standby' has all three TXs' up & running, The B+ is disconnected on the secondary, The two filiment TXs' are running thru dividers set for appx 63%."start" has the B+ running thru a PWR resistor to start-up the caps, the Filiments now have full power, "On" now just bypasses the start-up resistor & of course the filiments are still at full power.
_____________________________________Rick.................
Attachments
Total overkill on the power supply warm up.. 
It's not required with the 2A3 which warms up very quickly. Tube rectification would remove this as an issue, but you can safely ignore it with SS rectification as well.
AC is fine on the 2A3 filaments, anything above 2.5V I generally recommend DC heating, but I have both 45 and 2A3 amplifiers with no audible hum or buzz on their outputs when connected to my current ~100dBspl/w/m speaker system or their predecessors which were about 6dB more efficient.. (Both have usable response down to 35Hz or better.)
The Sovtek 2A3 offers a lot of bang for the buck, and with cathode bias and the right transformer primary Z sounds pretty good. (I use 3K transformer at ~60mA and 250V plate to cathode - pretty standard actually.)
JJ are made in the Slovak Republic not Russia last time I checked.
QC has really bombed on certain tube lines. Jim McShane can tell you a few stories about bad JJ tubes. A shame as the older 300B was quite good (and what I use)
The JJ 2A3 is actually a 300B with a 2.5V filament.
I would highly recommend the Sylvania 6SN7 as a driver, I had lots of 5691 and find older Sylvanias to be far superior sonically.
It's not required with the 2A3 which warms up very quickly. Tube rectification would remove this as an issue, but you can safely ignore it with SS rectification as well.AC is fine on the 2A3 filaments, anything above 2.5V I generally recommend DC heating, but I have both 45 and 2A3 amplifiers with no audible hum or buzz on their outputs when connected to my current ~100dBspl/w/m speaker system or their predecessors which were about 6dB more efficient.. (Both have usable response down to 35Hz or better.)
The Sovtek 2A3 offers a lot of bang for the buck, and with cathode bias and the right transformer primary Z sounds pretty good. (I use 3K transformer at ~60mA and 250V plate to cathode - pretty standard actually.)
JJ are made in the Slovak Republic not Russia last time I checked.
QC has really bombed on certain tube lines. Jim McShane can tell you a few stories about bad JJ tubes. A shame as the older 300B was quite good (and what I use)
The JJ 2A3 is actually a 300B with a 2.5V filament.
I would highly recommend the Sylvania 6SN7 as a driver, I had lots of 5691 and find older Sylvanias to be far superior sonically.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.