The AZ21 differs from the others primarily by a slightly higher maximally rectified current, and it also has a middle heating terminal to which a + high voltage output can be connected. This, albeit minimally, reduces the uneven wear of the ends of the heating outlet and makes the middle outlet of the heating coil of the rectifier unnecessary.No. AZ1, AZ11, AZ21 and AZ31 are the same tubes, but with different bases. They're all successors of the venerable Telefunken RGN 1064 or Philips/Valvo G 1064. First production run for all of them featured mesh plates, later replaced by solid plates - without altering their characteristics or ratings, mind you!
I guess you're after mesh plate rectifiers due to their more transparent sound, aren't you?
Best regards!
Andy, btw way, why did you wonder that plate voltage went up when using a AZ12? That is comparing like rgn1064 with a 1A filament with the rgn 2004 with 2 A filament... The bigger the type, the lower the voltage deop/internal resistance.
If you want the mesh sound AND a bit stiffer psu, so less lush, more dynamic, simply use the bigger types like rgn2004 or rgn4004... the make your psu lower resistance and have dynamic reserves.
By the way... the rgn types have a serial filament internally, not a parallel filament like the 5u4g types...
And yes, i have tons of rgn064 and rgn2004 types, mesh and solid... and the sound difference is obvious... if someone with the same inventory aegues in the other direction, just let me buy your mesh types...
If you want the mesh sound AND a bit stiffer psu, so less lush, more dynamic, simply use the bigger types like rgn2004 or rgn4004... the make your psu lower resistance and have dynamic reserves.
By the way... the rgn types have a serial filament internally, not a parallel filament like the 5u4g types...
And yes, i have tons of rgn064 and rgn2004 types, mesh and solid... and the sound difference is obvious... if someone with the same inventory aegues in the other direction, just let me buy your mesh types...
I only have some RGN 1064's and AZ 1's with mesh plates, but I prefer to keep them for old tube radios instead of amplifiers where I don't see any benefit over simple SS diodes.
Regarding different rectifier sound: I'd bet my *** that noone can discern spot on which rectifier is plugged into the amp without having seen it before.
Best regards!
Regarding different rectifier sound: I'd bet my *** that noone can discern spot on which rectifier is plugged into the amp without having seen it before.
Best regards!
Last edited:
Interesting day's work today. I have a second PSU which is nearly identical to the first. It uses parallel AZ1 mesh rectifiers. Fortunately this PSU is performing as it should, and the results seem to indicate that there is something wrong in my other AZ11 PSU - rectifiers or caps or whatever. The voltage drop is more than it should be.
I tried my twin AZ1 PSU out with a 40uF first cap, then a 6.6uF first cap and this now allows me to control the output voltage with the first cap. Voltages now look right. The transformer resistance is 26 ohms. Less than ideal.
The mesh sound has been restored and it sounds nice again. What a relief. I need the mesh sound.
I tried my twin AZ1 PSU out with a 40uF first cap, then a 6.6uF first cap and this now allows me to control the output voltage with the first cap. Voltages now look right. The transformer resistance is 26 ohms. Less than ideal.
The mesh sound has been restored and it sounds nice again. What a relief. I need the mesh sound.
Attachments
Last edited:
The main transformer is EI. The top-up transformer which adds 56v is toroid. I could use an EI instead, I have some 24+24v.
It would be good to use a "proper" mains transformer, but would this make any real difference do you think?
How important is it to increase the 26 ohm resistance to something like 100 ohms?
It would be good to use a "proper" mains transformer, but would this make any real difference do you think?
How important is it to increase the 26 ohm resistance to something like 100 ohms?
... or mercury vapour rectifiers.
Pity these things have become so expensive. Lucky you who got a stock 15-10 years ago.
If you use cLC filter, the first capacitor charging pulses stressed mainly the rectifier.
In your case even at 4.7uF the pulses can reach 0.7A peek ... if you don't use limiter (mainly DCR of secondary).
You use PSU 7/24 ... because with this low DCR secondary sparking occurs at turn on.
If you want long life rectifier, never use it over its limits.
If you use choke input filter, the stress significantly lesser ... but no free lunch, the output DC also significantly lesser.
In your case even at 4.7uF the pulses can reach 0.7A peek ... if you don't use limiter (mainly DCR of secondary).
You use PSU 7/24 ... because with this low DCR secondary sparking occurs at turn on.
If you want long life rectifier, never use it over its limits.
If you use choke input filter, the stress significantly lesser ... but no free lunch, the output DC also significantly lesser.
... or mercury vapour rectifiers.
Yep...I use 872A for my 300B/842 amp as those old RCA have enormous fine resolution and tone, better than 866A which aound a bit treble heavy.
But indeed for driver stages I prefer rgn2004 and for DAC output stage rgn1064. All dual mono.
For the DAC stage i tried old globe versions of 866a with V filament, that has a different character than rgn1063, but as well very nice as they sound warmer than normal 866a types..amd the purple glows is the coolest thing I ever saw...check Dave's site hifiheroin I believe...
Last edited:
No, you are wrong. The transformer resistance is important and when you dig deep in the old datasheets of the rgn types, there are even formulars in ther which show the resistance allowed is depending on both, HV and current. As well, the trick is NOT to use resistors between the transformer winding and the rectifier. This sounds really bad. It must be a piece if copper, so ideally the secondary windings need to have the right resistance.
You can as well see immediately the effect in PSUD...as Euro21 is refering to...
I tPically like to use a transformer with 70-120ohm per leg from CT. Lower with the 872a...because...the resistance is depending on the HV as well (in the datasheets they suggest higher resistance, but that is for 3000V....)
My PSU are all choke input with a small Cap infront as this lowers noise and help to reduce the hammer effect on the first choke. athe size again depends on the HV and current, but is between 0,33uF and 2uF max currently. The quality is important here, I tried old paper in oil types but that sound slow. My favourite is Miflex Kpcu1. and for the rest if the psu Cornell cde 947d..hard to beat and I tried a lot....really a lot...they are worth every penny...just compare their priece to black gates and be sure they are better than BG.
Soundwise, you ll see a change from a very dry, very precise, but as well stiff tone characteric (choke input) to a very loose, less accurate 3D and warmer presentation (pi filter)...amd yes, I have multi-tapped transformers to generate in both cases the same HV to compare. Bias was the same for the tubes... The mixture I suggest gives you best of both worlds. This point is easy to find as u start with pure choke input, than add small caps until HV starts to raise very slightly like a few volts. Bingo.
Last edited:
Yes, you're right. It's the conductive angle that counts. With low impedances in the tranny and the 1st filter cap, the rectifier encounters small conductive angles, i.e. short pulses of high charging current and half the mains frequency that are literally hammering onto the cathode(s) - and vice versa, of course. With a properly designed L input, the inductor's impedance automatically enlargens the conductive angle which is a benefit to the rectifier. With a C input you need to obey the rectifier's datasheet in terms of maximum filter input capacitor and mininal DC resistance. It doesn't matter at all if this resistance is in the tranny wire (both primary and secondary windings need to be regarded accordingly) or in additional resistors.
Best regards!
Yup, that is how PSUD will look at it...and than comes the practise. You simply can hear the materialmix of an added resistor or the type of cap. Actually if you ever tried power transformer with amorphous cores, you ll be surprised that they even make a bigger difference than chokes with amorphous or nano cores...so as the inventor of the ongaku always said...its not about a new circuit...its all about the right ingredients on the materials level...and the PSU is the basis of all.
Kondo san taught me how to make his Audio Note cartridges, at their Tokyo workshop in the early 1990s and I can confirm material choices - e.g. age annealed silver wire, permendur pole pieces etc. - were central to his thinking. My contribution was to convert a hand made ad hoc variable production supply, into a repeatable process, which is still available today.
Btw, DHTRob recently tested copper wire wound cathode resistors and is pretty impressed. I believe the evil comes mostly from the materials. Also, chokes/inductors are high mass components. Resistors are low mass. Try mechanically coupling them to the chassis. Some food for thought.
P.S. You can play with the resistor sounds by removing the default wires and soldering your prefered on your own, or adding mass to it using prefered materials. I believe in the attempt of potting a resistor into tonewood.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.