For some reason the 'quote' button is missing on the screen that I am looking at, so please excuse, and all to say: these tubes last a long time, particularly when run at such low dissipation. Its a bit like 1541A, you can find them now and again - for example there is a guy selling D3a NIB at USD 40 at the sales forum here at diyaudio. For me, I'm a bit spoiled for choice, but E180F is a clear winner at <10$.
If you don't need mu 60, and half that will suffice, say 2vRMS without step down, 6922 types set up at zero bias are great (they sound great that way too) but they aren't $10, either (and dont sound quite as good as what we're otherwise talking about, IME).
If you don't need mu 60, and half that will suffice, say 2vRMS without step down, 6922 types set up at zero bias are great (they sound great that way too) but they aren't $10, either (and dont sound quite as good as what we're otherwise talking about, IME).
Can we current manufacture to original Siemens standards for say 30 Bux each?
Nope. There is an OEM invoice floating around the web from the early 90' that had D3a invoiced out, direct from Siemens, at the equivalent of (then) Eu 140 per. 35 years ago.,
Nope. There is an OEM invoice floating around the web from the early 90' that had D3a invoiced out, direct from Siemens, at the equivalent of (then) Eu 140 per. 35 years ago.,
The 6AK5 I think is for now better value and cuter.
These days use 5670, electrically similar enough, good quality NOS under 10 bux.
Thor
Buggeration and botheration. But now everything is made in china for 1/100's the cost.
Shugang E810F and D3a at 2 euro each please!
Anyway, I pay less than 100 Bux for a CDP with TDA1541 to harvest and expect make a profit when I part out the players for spares on e-bay!
No way I pay 300 Bux for a pair of tubes. I'm too cheap for that!
Thor
i just purchased 8 pieces of d3a, 30e a piece.
Thing is, d3a is difficult to make, so i don't belive chinese will replicate it (soon). Frame grid tube.. But lets hope they do. 8 pieces will take a long time to run trough in this dac, so i'm set for a while. Now C3G is probably even better, BUT that thing is really expensive, and will be even more in the future (that tssssss feeling when you forget to turn off the dac 😁).
Thing is, d3a is difficult to make, so i don't belive chinese will replicate it (soon). Frame grid tube.. But lets hope they do. 8 pieces will take a long time to run trough in this dac, so i'm set for a while. Now C3G is probably even better, BUT that thing is really expensive, and will be even more in the future (that tssssss feeling when you forget to turn off the dac 😁).
Buy 6AK5 and 5670 while they are still affordable.
If importation into Thailand wasn't such a huge PITA I'd take some D3a off yer hands.
Thor
I'll stock up, i love miniature tubes 🙂
Anywho, since we are making this for standard size chassis, and we are talking tubes. Not a bad time to speak of tube rectification. Pros cons, never done it, heard it, heard good things about it. Also hybrid solid and tube rectification. Logic says the sound signature comes from variation of voltage for tube to tube whole it does it's job, any light shedding on the matter?
Now that we have all the schematics of the circuits discussed in recent weeks, I would like to try to implement them.
I have made several DACs based on TDA1541A, most of them on perfboard. Now I would like to venture into a 4-layer assembly using stripboard or perfboard.
A final summary of the routing scheme in the 4 layers for signals, power supplies, grounds etc would help me a lot.
I know this topic has already been covered in this thread, so I apologise for the question.
I have made several DACs based on TDA1541A, most of them on perfboard. Now I would like to venture into a 4-layer assembly using stripboard or perfboard.
A final summary of the routing scheme in the 4 layers for signals, power supplies, grounds etc would help me a lot.
I know this topic has already been covered in this thread, so I apologise for the question.
I was hoping we could use EasyEDA as an editor for pcb design. I'm not their promoter, and I won't insist on this obviously, I just selfishly thought that their .json files were now widespread.😊
Antonio
Antonio
It's a bit more complicated.
Vacuum diodes have no reverse recovery, a very soft turn on curve and if heated directly from an AC winding on the power transformer offer a mostly resistive partial load, preventing the power transformer going "open circuit" for much of the conduction angle.
While each aspect is easily simulated in solid state circuitry, replicating all and sundry full spectrum however is non-trivial and wasteful.
Any dual triode can be made a tube rectifier
So a cheap Chinese 6N6 proudly made in the Laogai by slave labour of Falun Gong and anti communist prisoners, prior to suffering vivisection without anastasia to harvest organs, will do fine.
This just a reminder that everything has a price. At times, a hidden one.
But yes, just using tube Rectifiers is different from all efforts using solid state.
I'm open to discuss means to reduce the gap to simulations of the results, for now it seems a lot of effort for no return, unless you know what you do and can implement with confidence.
Thor
This was an interesting schematic by Alex, that employs a hybrid rectification followed by a clc. Not really interested in beyond that point from the schematic, as shunt hv regulation is better.
What if the DAC is entirely current sourced/shunt regulated? Can a tube rectifier modify the sound in that case?
(And what about double shunt, as in ccs->shunt->ccs->shunt?)
I prefer tube rectifier slow start up (indirect heated type) and no need to snub the B+ secondary - prefer to use TV damper diodes because they are dirt cheap and high quality, sockets can be tricky to source. 6CJ3 for example. I think there are advantages in using a separate rectifier heater transformer.
*edit - but 3A heater current, the sockets and the space could be a bit 'overkill' ..
EZ80, EZ81 or cheap and cheerful 6X4?
Last edited:
My only thought is that you can't plug and play D3a, E280F or E810F as a substitute for 6AK5, whereas with a 9 pin socket wired for E180F, you can ..
This is a difefent and new concept.
Main difference is diferent usage if Tube stage with moderate amplification, and active iV as sort of biffer after the DAC.
Very different from Riv to DAC output then to tube stage.
.
this is not good chematic from the conceptual angle. Mixing a range of diferent technology levels even IC and tubes in one circuit?
For my opinion better way to use classic tube PS with also tube regulator at the end Or diskrete solid state.
...
For someone to have "a must" to use IC better otopn is classic LM723 in HV regultaotor mode?
Hi Zoran, It just gives more options = they all plug in the same. If you have say, CCS load.. then okay adjust the current to suit, all else the same. If you have choke (transformer same) or resistor load, use an adjustable HV reg - tube or solid state, as you suggest, no probs and just dial it in.
https://www.diyaudio.com/community/threads/6ak5-in-triode-plate-resistance.343410/
I can see what you are saying.
However, with limited voltage swing, horizontal load line, (close to) zero bias.. perhaps 12-16mA is acceptable.
Rp 3k3.. hmm.
*edit - sorry I missed your edit 🙂
Last edited: