Hi, is it better to use 2 separate cathode resistors for valves in parallel ?
Or should I treat them as 1 valve and use 1 cathode bias resistor ?
Thank you.
Or should I treat them as 1 valve and use 1 cathode bias resistor ?
Thank you.
Use separate resistors unless you want to use well matched tubes otherwise one will hog the current and run hot, the other will run cold.
You might need to use plate stoppers, too. 27R per tube stopped an RF oscillation I was having on a PPP 6P3S build.
You might need to use plate stoppers, too. 27R per tube stopped an RF oscillation I was having on a PPP 6P3S build.
Thank you. So 2 resistors is the way to go.
The well matched thing got me thinking, if it is even worth it... because if the 2 triodes are not matched, how will they reduce noise ?
The well matched thing got me thinking, if it is even worth it... because if the 2 triodes are not matched, how will they reduce noise ?
I don't match tubes but I use an automatic grid bias system that will strive to match the current between the tubes so it's not needed anyway.
Depending on the tubes you buy, matching isn't much more money. I recently got two matched pairs (ended up being a matched quad because both pairs matched) of KT88-98 for under 200$CAD
What is the design you have in mind? Link a schematic 🙂
Depending on the tubes you buy, matching isn't much more money. I recently got two matched pairs (ended up being a matched quad because both pairs matched) of KT88-98 for under 200$CAD
What is the design you have in mind? Link a schematic 🙂
It is for the input circuit, the 2 triodes of an ecc83 in parallel. A mono signal.
Something like this but with 2 valve.
https://www.tubecad.com/articles_2002/RIAA_Preamps_Part_2/img4.gif
Something like this but with 2 valve.
https://www.tubecad.com/articles_2002/RIAA_Preamps_Part_2/img4.gif
Most of the time you can just short the plates, grid, and cathodes together on those tubes since they are so low in transconductance.
The added cost of the resistors is only a few cents though so why not?
The added cost of the resistors is only a few cents though so why not?
Two triodes in the same bottle are usually matched to within 10%; in separate bottles but same type and age maybe 20%. You won't get the theoretical 3.01dB reduction in hiss, but 2.8dB is still very good.
They should not "hog" unless one is very weak, so sad that it should be discarded.
There is broadband hiss.
There is 1/f noise.
There is shot noise.
They are not exactly the same.
Get real aggressive, and use a single 416A RF Lighthouse Triode per channel.
(probably a lot more trouble than it is worth).
Just my opinions.
There is 1/f noise.
There is shot noise.
They are not exactly the same.
Get real aggressive, and use a single 416A RF Lighthouse Triode per channel.
(probably a lot more trouble than it is worth).
Just my opinions.
What is the application? A phono preamp for moving magnet cartridges? If so...
The input capacitance of an ECC83 is quite high, due to the Miller effect (gain multiplies input capacitance). ECC83 has high gain and fairly high anode-grid capacitance. Typically, the in-circuit input capacitance for one (cathode fully bypassed) ECC83 will be at least 150pF, and usually closer to 200pF.
If you take two ECC83 and wire them in parallel, the input C of each of those triodes will act like two capacitors in parallel. That is, the capacitance will sum (add). That means the input capacitance of your two ECC83 in parallel is likely to be from 300pF to 400pF. Add roughly 150pF for your tonearm wiring and cable from turntable to preamp, and you're looking at 450pF to 550pF input shunt capacitance. That will introduce a whopping resonance to your MM cartridge's response.
This may not be a problem with a high-output moving coil cartridge like Denon DL110 or a moving-iron (MI) cartridge like Grado, etc.
So... What is the application?
If it's a preamp tube then 1 shared resistor is fine. If it's power tubes then it's kinder to use 2 resistors.