Hi Guys
I’m after suggestions for the best tube to drive an SE 300B- via an interstage transformer. The latter is a 1:1 bifilar wound design, with a primary inductance of 50 H. It was intended for currents in the 15 – 20 mA range, although its specified up to 35 mA. Over this range the inductance is pretty fixed.
Other restrictions-
Not a DHT (for now)
Mu ~ 20 (preamp line stage gain is high, so noise a problem with higher gain drivers)
Operating current no higher than 35 mA (~ 20 mA preferable)
Operating bias at least – 6 V, preferably – 8V, to give an output swing considerably greater than the 300B’s bias voltage (currently about – 95 V. Keeps driver await from its cut off region and noticeably increases the out put level at which audible clipping occurs on, e.g. loud piano)
Plate resistance 2k or less (for good bass response)
Plate voltage ~ 180 V (?)
9 pin base if at all possible
It’s quite tricky to come up with suitable types within these criteria. The best sounding compromise I’ve found so far is a triode connected 6P15P- 215 V on the plate, 20 mA, - 6.0 bias. I’m trying a 6N6P at the moment- 210 V plate, 20 mA, 8.0 V bias- but I preferred the 6P15P. For a long time, I used a 6P14P, 230 V plate, 26 mA, -8 V bias. This is a lot below its ideal operating current though, and the 6P15P brought extra clarity (though I’m using that a somewhat low current, to get the bias V up, the plate V against current characteristics are still pretty linear.
For what it’s worth, I’ve found that I prefer a zener diode as cathode bias over a resistor + capacitor. I can’t directly hear any zener noise, and dynamic impedance at 20 mA of the type I use is only 7 ohms!)
Any thoughts?
Paul N
I’m after suggestions for the best tube to drive an SE 300B- via an interstage transformer. The latter is a 1:1 bifilar wound design, with a primary inductance of 50 H. It was intended for currents in the 15 – 20 mA range, although its specified up to 35 mA. Over this range the inductance is pretty fixed.
Other restrictions-
Not a DHT (for now)
Mu ~ 20 (preamp line stage gain is high, so noise a problem with higher gain drivers)
Operating current no higher than 35 mA (~ 20 mA preferable)
Operating bias at least – 6 V, preferably – 8V, to give an output swing considerably greater than the 300B’s bias voltage (currently about – 95 V. Keeps driver await from its cut off region and noticeably increases the out put level at which audible clipping occurs on, e.g. loud piano)
Plate resistance 2k or less (for good bass response)
Plate voltage ~ 180 V (?)
9 pin base if at all possible
It’s quite tricky to come up with suitable types within these criteria. The best sounding compromise I’ve found so far is a triode connected 6P15P- 215 V on the plate, 20 mA, - 6.0 bias. I’m trying a 6N6P at the moment- 210 V plate, 20 mA, 8.0 V bias- but I preferred the 6P15P. For a long time, I used a 6P14P, 230 V plate, 26 mA, -8 V bias. This is a lot below its ideal operating current though, and the 6P15P brought extra clarity (though I’m using that a somewhat low current, to get the bias V up, the plate V against current characteristics are still pretty linear.
For what it’s worth, I’ve found that I prefer a zener diode as cathode bias over a resistor + capacitor. I can’t directly hear any zener noise, and dynamic impedance at 20 mA of the type I use is only 7 ohms!)
Any thoughts?
Paul N
Hi Shoog
5687 is a classic choice for sure, but it’s a bit thin on the ground these days. And the guy who designed the interstage transformer compared it to E182CC years back- and preferred the latter. Not saying I’d find the same, though I did prefer the E182CC as a common cathode connected driver- back in the darks days when I thought this was a good idea! Either way, the E182CC is a rarity too now.
ECC99. I’d forgotten about this. I think I’ve read some underwhelming reports on it in this role, but will definitely give it a go, as it looks an excellent fit on paper.
Any other thoughts?
Paul N
5687 is a classic choice for sure, but it’s a bit thin on the ground these days. And the guy who designed the interstage transformer compared it to E182CC years back- and preferred the latter. Not saying I’d find the same, though I did prefer the E182CC as a common cathode connected driver- back in the darks days when I thought this was a good idea! Either way, the E182CC is a rarity too now.
ECC99. I’d forgotten about this. I think I’ve read some underwhelming reports on it in this role, but will definitely give it a go, as it looks an excellent fit on paper.
Any other thoughts?
Paul N
A (triode connected) 7788/E810F will satisfy most of your requirements and is hard to beat sonically. Also check Sound Practices Iss.15 pages 11-12-13 for more details about this tube and its use in a transformer coupled driver for a VV52B triode in S.E mode. I don't really like zener diodes used for cathode biasing, I'm allways hearing some soiling of the sound. Zeners are know to be temperature-dependant and noisy components. The proposed mehod using NiCad cells is much better, IMO.
Attachments
Member
Joined 2009
Paid Member
I wouldn't expect much difference between the 6N6P you are trying and an EC99.
The gain is possibly too high without some degeneration, but the 6C45Pi is a nice tube, low resistance and good driver. I've heard an SV811 being driven via an IT by one of these.
If I remember correctly, there is a sweet spot for zener diodes for minimal noise, at around 5.6 to 6.2 V.
The gain is possibly too high without some degeneration, but the 6C45Pi is a nice tube, low resistance and good driver. I've heard an SV811 being driven via an IT by one of these.
If I remember correctly, there is a sweet spot for zener diodes for minimal noise, at around 5.6 to 6.2 V.
Hi!
Have you ever seen the plate curves of a triode connected E810F? Not datasheet curves, real ones. I have examined it with a curve tracer. While good for small signals it gets quite non-linear at the outer regions. I would not use it as a driver in a power amp.
I have used it in LCR phonostages and there it works nicely.
Best regards
Thomas
Have you ever seen the plate curves of a triode connected E810F? Not datasheet curves, real ones. I have examined it with a curve tracer. While good for small signals it gets quite non-linear at the outer regions. I would not use it as a driver in a power amp.
I have used it in LCR phonostages and there it works nicely.
Best regards
Thomas
Hi Tubologic, Bigun
Possibilities abound if we allow higher mu tubes- with 6C45 and triode connected 7788 both strong contenders. But I didn't, and you cheated!
I was very wary of zener bias until I tried it. I certainly wouldn't use it in a low level stage. I've never tried batteries- many like them, but many loathe them. I use LEDs to bias CCS loaded stages, but would be wary of them here, as several in series would be needed, raising their dynamic impedance
One problem with this topology is we can't lower the stage gain by using an unbypassed cathode resistor- this just raises the anode impedance and thus curtails bass response
Paul N
My long term plan is to try Rod Coleman's shunt cascode driver topology, discussed elsewhere on the site, in my amps. I installed this to brilliant effect in a friend's amps, even though the latter were somewhat compromised in comparison- weedier power supplies, output transformers etc
Possibilities abound if we allow higher mu tubes- with 6C45 and triode connected 7788 both strong contenders. But I didn't, and you cheated!
I was very wary of zener bias until I tried it. I certainly wouldn't use it in a low level stage. I've never tried batteries- many like them, but many loathe them. I use LEDs to bias CCS loaded stages, but would be wary of them here, as several in series would be needed, raising their dynamic impedance
One problem with this topology is we can't lower the stage gain by using an unbypassed cathode resistor- this just raises the anode impedance and thus curtails bass response
Paul N
My long term plan is to try Rod Coleman's shunt cascode driver topology, discussed elsewhere on the site, in my amps. I installed this to brilliant effect in a friend's amps, even though the latter were somewhat compromised in comparison- weedier power supplies, output transformers etc
You beat me to it, I was going to suggest the EL84 or EL86. Both drivers with real gain, lots of muscle and a universally liked tone.
Do not discount the option of using a higher mu valve and then stepping down in the interstage - that allows you to beat the residual losses in the IT into submission with a very large stick.
Shoog
Do not discount the option of using a higher mu valve and then stepping down in the interstage - that allows you to beat the residual losses in the IT into submission with a very large stick.
Shoog
Look for the "supertube" E282F Used by many in this application.
mu 27 and output impedance of less than 1 KOhm in triode mode
Triode connection data aviable here: http://frank.pocnet.net/sheets/009/e/E282F.pdf
Johan
Hi Paul,
6П15П is very hard to beat in this position! Many higher-gm triodes and triode-connected pentodes are excellent performers when the anode voltage is fixed (and therefore great in cascode) - but at high voltage the curves slump to the right (to use a crude approximation).
Purpose-designed Output valves, including line & frame video outputs are the best field.
Have you checked out the 6Ф6С?
Nominal point is at 250/250v -16,5v for 35mA. Only 3mA/V though, and the base is not what you want.
6П15П has different builds though - and some Reflektor (Saratov) made types are characterised like 6П14П. Other Reflektors may be different, and there are also Orel-factory parts. Also, the 6П15П-EB tough version. These are the examples I have here, but there may be others.
6П15П is very hard to beat in this position! Many higher-gm triodes and triode-connected pentodes are excellent performers when the anode voltage is fixed (and therefore great in cascode) - but at high voltage the curves slump to the right (to use a crude approximation).
Purpose-designed Output valves, including line & frame video outputs are the best field.
Have you checked out the 6Ф6С?
Nominal point is at 250/250v -16,5v for 35mA. Only 3mA/V though, and the base is not what you want.
6П15П has different builds though - and some Reflektor (Saratov) made types are characterised like 6П14П. Other Reflektors may be different, and there are also Orel-factory parts. Also, the 6П15П-EB tough version. These are the examples I have here, but there may be others.
Hi Babbe
What are your operating conditions for the 6N6P and EL84M? I'm trying 6N6P at 210 V anode, - 8 V bias, 20mA at the moment, to maximise the available voltage swing. But might sound better with -6 or -7 V of bias and a lower anode V?
I used a triode connected 6P14P- aka a "EL84M"- for a long time at 230V/26 mA/26 mA. Compromise to optimise output swing available, while running within ITX's current limits- far away from its spec sheet optimum conditions (48 mA)!
Hi Shoog
Step down ITX sounds interesting. Do have any links to relevant documentation? Though it would require considerable investment in new ITX's
Hi JohanB
Had a quick look at the triode connected specs of the E282F. Ia (not including Ig2!) 40 mA at 110V on the anode and - 2 V bias! Nowhere near enough output swing V to drive a 300B- or am I missing something?
Many thanks
Paul N
What are your operating conditions for the 6N6P and EL84M? I'm trying 6N6P at 210 V anode, - 8 V bias, 20mA at the moment, to maximise the available voltage swing. But might sound better with -6 or -7 V of bias and a lower anode V?
I used a triode connected 6P14P- aka a "EL84M"- for a long time at 230V/26 mA/26 mA. Compromise to optimise output swing available, while running within ITX's current limits- far away from its spec sheet optimum conditions (48 mA)!
Hi Shoog
Step down ITX sounds interesting. Do have any links to relevant documentation? Though it would require considerable investment in new ITX's
Hi JohanB
Had a quick look at the triode connected specs of the E282F. Ia (not including Ig2!) 40 mA at 110V on the anode and - 2 V bias! Nowhere near enough output swing V to drive a 300B- or am I missing something?
Many thanks
Paul N
Hi Rod
Glad you weighed in!
I used the 6П15П-EB (6P15P-EV) version in triode connection, after your earlier (emailed) suggestion. My operating conditions- 215 V plate, -6.0 V bias, cathode current 19 mA, are a compromise- the plate V is a bit higher, and the current a little lower than I'd ideally like, to get the bias up to -6 V. Even then I'm getting only a possible +/- 125 V swing. The TJ Mesh plates I'm using are fixed bias at -95 V for 68 mA with 420 V on the plate (28 W dissipation), so I'm not avoiding, at full output, cut off of the driver by as much asI'd like, yet the plate will instantaneously go over 300V- not sure if curves still good here. Whatever, I think I prefer the 6P15P to all comers so far. Though the 6P14P- despite its low current operating condition- had its charms
6Ф6С/6F6S looks a bit of a brute
Paul
Glad you weighed in!
I used the 6П15П-EB (6P15P-EV) version in triode connection, after your earlier (emailed) suggestion. My operating conditions- 215 V plate, -6.0 V bias, cathode current 19 mA, are a compromise- the plate V is a bit higher, and the current a little lower than I'd ideally like, to get the bias up to -6 V. Even then I'm getting only a possible +/- 125 V swing. The TJ Mesh plates I'm using are fixed bias at -95 V for 68 mA with 420 V on the plate (28 W dissipation), so I'm not avoiding, at full output, cut off of the driver by as much asI'd like, yet the plate will instantaneously go over 300V- not sure if curves still good here. Whatever, I think I prefer the 6P15P to all comers so far. Though the 6P14P- despite its low current operating condition- had its charms
6Ф6С/6F6S looks a bit of a brute
Paul
If you look at the triode connected caracteristics on page 7 in the data sheets,
http://frank.pocnet.net/sheets/009/e/E282F.pdf
you can see that a E282F:
@Ua=125V, Ug1=-3V, Ia (+Ig2)=25mA, the E282F can swing +-80V at no load & +-70V at 8KOhms shunted load
This will give 140 to 160Volts Peak to Peak in the inductive transformer load and will be more than enough to drive the 300B, even into the grid current area.
I also think the distorsion caracteristics of the triode connected E282F vill cancel much of the distorsion in the 300B.
Can be fine tuned by resistive loading of the transformers primary/secondary to perform better transient response.
Johan
Hi Vinylsavor
Thanks for that. Information about real world behaviour in this application is just what we need!
Paul N
Thanks for that. Information about real world behaviour in this application is just what we need!
Paul N
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.