Thanks for the info, I think that resistor has to be the best quality it can be in Filament Bias, I also think it's important to try different values as well. If trying to build things using other People Schematics, it has to be identical, otherwise you'll probably not get the same results, the Power supply has to be right for starters & the Valves/Tubes that we are using are all different, especially 01a & 26 Types, some of them are also nearly a century old, Ale's Gyrator is Great for that & sounds amazing, especially if You take notice of the Mosfets used, also use a good quality resistor for RMU, I am now using Miflex Caps with my 2P29L & Russian Teflon with my 01a(accurate without coloration).
matejsirk
thank for info about anode chokes...
Is there any problem with 1m long wire from HV supply to gyrator
and same from LV to Coleman print
i tried this ,,but the gyrator is now defekt,,,found out that the j310 fet is finish
so therefor i will try choke....
have not used my pre for a while,because i have some serious noise issues,so i think maybe remove all the AC like trafo and caps away from the tube
best Bjarne
thank for info about anode chokes...
Is there any problem with 1m long wire from HV supply to gyrator
and same from LV to Coleman print
i tried this ,,but the gyrator is now defekt,,,found out that the j310 fet is finish
so therefor i will try choke....
have not used my pre for a while,because i have some serious noise issues,so i think maybe remove all the AC like trafo and caps away from the tube
best Bjarne
J310 if I'm not mistaken will only do 10mA(ok for 01a or 26), so You need a more robust N Channel FET, I use BSH111K(SMD) as it will do 60mA, but if You are not happy soldering SMD, try 2N7000, which will do 200mA. There is no problem running Coleman Regs at one meter or more, just read Rod's notes though, 2 or even 3 boxes is best, as You could separate Filament & B+ supplies, although that isn't necessary, I got 2P29L to work in one chassis, but I have tried 4P1L this way with little success, also use PSUD2 to design B+Power Supply to make sure that is as clean as can be, make sure You only ground Filaments after Cathode Resistor or You will get a lot of Hum. Rod Coleman explains this in His notes. A proper grounding scheme is most important, all grounding should go to one point before being taken to the Chassis/Safety ground.
Hmmm that j310 is easy to destroy,,happened before..i will look at the 2n7000
humm is not so bad ,,more hiss and other high noise
i will keep on the long wire,,,
i use kkaudio print for the 4P1L ,,
now i have 2 things i must try,,anode choke and a stronger trans on the gyrator
so can i hear what sound best
edit,,j310 i think is 60ma
thanks Bjarne
humm is not so bad ,,more hiss and other high noise
i will keep on the long wire,,,
i use kkaudio print for the 4P1L ,,
now i have 2 things i must try,,anode choke and a stronger trans on the gyrator
so can i hear what sound best
edit,,j310 i think is 60ma
thanks Bjarne
Last edited:
Unless you have strong reasons for using a 4P1L I would use a 01A or 26 with SIC diode bias. I couldn't get many of my old 01As to bias correctly with Ale's gyrator for some reason, so at that point I went over to anode chokes and stayed with them. The 26 sounds surprisingly good with SIC diode bias - better than it did with any other kind of biasing - so for now that's what I'm using and very happy with the sound. The 7mA current could be higher for driving the 4P1L PSE outputs but it's just about in the ballpark, which the 01A at 3mA was not.
The 10Y and 01A have a sparkle and freshness that is quite unique. The 4P1L for me is an output tube in PSE unless I can think of a very good reason to use it elsewhere e.g. it's a good driver tube. The 26 has a warm sound usually, where 4P1L is cleaner. But with SIC diode bias the warm 26 turns into quite another beast - very clean and detailed but still with that lovely 26 midrange that is so nice on vocals.
I have plenty of 01A - that's not a problem. It's just that most of the globes don't bias with the gyrator. I didn't go into why at the time. The ST ones were a bit better but I had less of them. The 01A isn't particularly microphonic, neither is the 26, but you can always come across noisy examples. These are very old tubes!
Last edited:
Yes, this is correct. Locating the transformer and rectifier/caps some 1m or more from the signal/tube chassis will usually make the system quieter and better-sounding, because the rectifier recharge pulses - and the magnetic field they generate - do not get coupled into the signal-handling parts.
For the same reason, please keep transformers & rectifiers away from DACs and MC-cartridge amplifiers.
OTOH please try to keep the regulator near to the tube socket.
24... 60mA.
My J310 batch (few dozens of NOS ones) dispersion is 28...42mA.
I regularly use these JFETs for 801a and 4P1L gyrators.
Hi Andy, I have tried 01a in ST & Globe & I think the Globe versions sounds much better, but that is only with the samples I have. I must try choke loading also, as it seems to be a good way to go. The type of coupling cap seems to make a big difference also.
Happy New Year!
The hybrid mu-follower stage (aka gyrator) is fixing anode voltage. When using filament bias, also the cathode bias is fixed mainly by the filament current and variance of the anode current provides little change. In this way, 2 parameters are fixed in the circuit - anode voltage and cathode-grid bias voltage.
The 201a's globes are very old valves from the 1920s. If the parameters are far away due to low emission, etc. you should expect the valve to have very low (or very high) anode current as you have already fixed the 2 controlling voltages (anode and cathode/grid).
I've taken the time and effort to curve trace my 01a stash. I discarded the ones that are not very good.
Ale
Hi Ale,
It could well be a matter of emissions, as you say. The gyrator did work with a couple of my 01As. It's not impossible that I had several low emission 01A globes. It would have been useful if the design was able to cover lower emissions, but I can understand that you don't design for out-of-range tubes. I don't have emission issues with plate chokes, so that's one way to make use of the surviving pool of less than perfect 01As and other such tubes. Since I have quite a few of them I needed some kind of solution - just being practical.
Happy New Year!
Andy
It could well be a matter of emissions, as you say. The gyrator did work with a couple of my 01As. It's not impossible that I had several low emission 01A globes. It would have been useful if the design was able to cover lower emissions, but I can understand that you don't design for out-of-range tubes. I don't have emission issues with plate chokes, so that's one way to make use of the surviving pool of less than perfect 01As and other such tubes. Since I have quite a few of them I needed some kind of solution - just being practical.
Happy New Year!
Andy
So use a CCS. With fixed plate current and fixed grid to filament return voltage, the only variation will be in plate voltage. With an AC coupled circuit this is without consequence even with large variations.
Similarly, if you use a plate choke feed it with a CCS which is AC decoupled from the plate load by a suitable capacitor for the same result. Setting a fixed plate current will ensure consistent Rp (which is otherwise not the case), and allows to run optimally with the air gap and uniformly between channels (which is otherwise not the case), and you can expect a PSRR of >100dB.
The major advantage of the 'gyrator' IMO is fixed voltage at the plate of the tube, and when fed from a regulated B supply, if the voltage across the load can be made equal to the bias voltage of the following stage, it allows for simple stacking of supplies and DC coupling where all and any variation is then moved to the plate of the second stage.
All else being equal, and with practical use of all three mentioned above, I prefer the CCS fed plate choke, stacked supplies and DC coupling to the second tube. This exchanges a coupling cap, two connectors and a length of interconnecting per channel for 1.5" of wire.
HK
Last edited:
Most low emmision thoriated tungsten valves respond well to rejuvenation attempts.
