bcarso
Well-known member
A bunch of stuff comes up about the 1626 doing a google search with just 1626 tube.
Take Five Minutes And Help the Tube Noobs -First Tube Stuff
- Thread starter northsiderap
- Start date
Help Support GroupDIY Audio Forum:
kiira
Well-known member
[quote author="bcarso"]A bunch of stuff comes up about the 1626 doing a google search with just 1626 tube.[/quote]
There was one single ended power amp with an output of about half a watt that used a 1626 tube but that is the only example of a circuit for the tube I've ever found. I've seen other low power triodes used in line stages and wanted to design someting for this tube because it has not been done yet that I know of - I'd have to really design it myself which would be a good learning thing for me. Plus I payed big money ($3) for these tubes and I want to use them. And they're real cute.
Kiira
There was one single ended power amp with an output of about half a watt that used a 1626 tube but that is the only example of a circuit for the tube I've ever found. I've seen other low power triodes used in line stages and wanted to design someting for this tube because it has not been done yet that I know of - I'd have to really design it myself which would be a good learning thing for me. Plus I payed big money ($3) for these tubes and I want to use them. And they're real cute.
Kiira
bcarso
Well-known member
"And they're real cute. "
Kiira, it doesn't sound like you get out enough either ;-)
Kiira, it doesn't sound like you get out enough either ;-)
> Take Five Minutes And Help
> 100ma at 200v = 20R plate resistance?
> and 50ma at 125v =6R?
> That seems off to me.. I almost know it's wrong...
{sigh...} Looks like more than 5 minutes of help needed.
100ma at 200v is 2,000 ohms. Where do you see 20?
When befuddled, don't work in "milli"anything. Convert to whole units. 100mA is 0.1A
Ohms is Volts divided by Amps.
200V/0.1A is 2,000. If you forgot your elementary education, any pocket calculator will figure this for you, or Windows Calc will give an almost-right answer.
That is the Large Signal or "static" plate resistance for 200V 0.1A. See red line.
Tubes pass current one way, audio swings both ways. We usually set a large DC current, then swing back and forth around that point. See blue line. We can swing 70V and about 70mA (0.07A), so the Dynamic plate resistance at this point is 1K.
But get back to the real world. The 1626 has a plate dissipation limit of 5 watts (orange curve). 200V at 0.1A is 20 Watts! (do you get the same answer?) In audio, we usually can't beat the plate dissipation limit. (In Class C radio, we can peak much higher, and that's what the 1626 was sold for, and why the curves cover territory way past the steady-state plate rating.)
A likely operating point for audio might be near the green line. Here the curves are steeper. The dynamic plate resistance is around 50V/25mA which is.... 2K. The Static plate resistance however 175V/30mA or about 6K.
Large-signal design can not be done correctly from simplified single-number data like plate resistance. The curves are too bent for any single-number to be true in all cases. However you can rough-sort tubes by the slope of the zero-bias grid line as it approaches the plate dissipation parabola, because that is usually the upper limit of how much current you can suck out of the bottle. (You can go higher if you drive the grid positive, but in audio it is too hard to do that and meet a distortion spec too.)
Small-signal design, where signal swings are MUCH less than supply voltage, can often be done well-enough with just a rough idea of the dynamic plate resistance at an operating point similar to what you plan to use.
> There was one single ended power amp with an output of about half a watt that used a 1626 tube ... I've seen other low power triodes used in line stages and wanted to design someting for this tube because it has not been done yet that I know of - I'd have to really design it myself
No you don't. What is the difference between a speaker output stage and a line output stage? Just power and impedance. Speakers need many-watts at a few ohms. Lines need a part-watt at hundreds of ohms. And in this case you HAVE a part-watt amplifier... you just need to find a transformer with 600Ω secondary instead of 8Ω secondary.
Typically, speaker-amps are more desperate for power than line amps. OTOH the complete audio path (artist to listener) passes through one speaker-amp but maybe many line-amps. You want low distortion in a line amp. So you might also increase the tube load impedance (primary impedance) which drops power a little but drops distortion even more. For old-school small power triodes, you may run into trouble winding a good high impedance with high enough inductance and low enough leakage. For 1626, a 10K:600Ω ratio may be a good first try.
> 100ma at 200v = 20R plate resistance?
> and 50ma at 125v =6R?
> That seems off to me.. I almost know it's wrong...
{sigh...} Looks like more than 5 minutes of help needed.
100ma at 200v is 2,000 ohms. Where do you see 20?
When befuddled, don't work in "milli"anything. Convert to whole units. 100mA is 0.1A
Ohms is Volts divided by Amps.
200V/0.1A is 2,000. If you forgot your elementary education, any pocket calculator will figure this for you, or Windows Calc will give an almost-right answer.
That is the Large Signal or "static" plate resistance for 200V 0.1A. See red line.
Tubes pass current one way, audio swings both ways. We usually set a large DC current, then swing back and forth around that point. See blue line. We can swing 70V and about 70mA (0.07A), so the Dynamic plate resistance at this point is 1K.
But get back to the real world. The 1626 has a plate dissipation limit of 5 watts (orange curve). 200V at 0.1A is 20 Watts! (do you get the same answer?) In audio, we usually can't beat the plate dissipation limit. (In Class C radio, we can peak much higher, and that's what the 1626 was sold for, and why the curves cover territory way past the steady-state plate rating.)
A likely operating point for audio might be near the green line. Here the curves are steeper. The dynamic plate resistance is around 50V/25mA which is.... 2K. The Static plate resistance however 175V/30mA or about 6K.
Large-signal design can not be done correctly from simplified single-number data like plate resistance. The curves are too bent for any single-number to be true in all cases. However you can rough-sort tubes by the slope of the zero-bias grid line as it approaches the plate dissipation parabola, because that is usually the upper limit of how much current you can suck out of the bottle. (You can go higher if you drive the grid positive, but in audio it is too hard to do that and meet a distortion spec too.)
Small-signal design, where signal swings are MUCH less than supply voltage, can often be done well-enough with just a rough idea of the dynamic plate resistance at an operating point similar to what you plan to use.
> There was one single ended power amp with an output of about half a watt that used a 1626 tube ... I've seen other low power triodes used in line stages and wanted to design someting for this tube because it has not been done yet that I know of - I'd have to really design it myself
No you don't. What is the difference between a speaker output stage and a line output stage? Just power and impedance. Speakers need many-watts at a few ohms. Lines need a part-watt at hundreds of ohms. And in this case you HAVE a part-watt amplifier... you just need to find a transformer with 600Ω secondary instead of 8Ω secondary.
Typically, speaker-amps are more desperate for power than line amps. OTOH the complete audio path (artist to listener) passes through one speaker-amp but maybe many line-amps. You want low distortion in a line amp. So you might also increase the tube load impedance (primary impedance) which drops power a little but drops distortion even more. For old-school small power triodes, you may run into trouble winding a good high impedance with high enough inductance and low enough leakage. For 1626, a 10K:600Ω ratio may be a good first try.
kiira
Well-known member
Here is a schematic of the power amp using B+ 265V.
http://www.2tough.com/~kiira/rec/darling.gif
I would like to use two of the tubes in something like srpp/cascode follower/cathode follower instead. I have a Haufe 1:4 (600:10000 I think) input xformer. [/code]
http://www.2tough.com/~kiira/rec/darling.gif
I would like to use two of the tubes in something like srpp/cascode follower/cathode follower instead. I have a Haufe 1:4 (600:10000 I think) input xformer. [/code]
kiira
Well-known member
[quote author="bcarso"]"And they're real cute. "
Kiira, it doesn't sound like you get out enough either ;-)[/quote]
I've forgotten what a date is. :?
Kiira, it doesn't sound like you get out enough either ;-)[/quote]
I've forgotten what a date is. :?
northsiderap
Well-known member
We had a decent session this weekend and I'm back to reading about tubes...
I guess I need a starting point. How about biasing the grid?
If my signal is +/- 1V would I want to bias the grid at -3, or anything above -2 or -3 that matched my desired plate resistance?
Basically, I'm askiing which parameter that one would start with, based on the assumption that one knows the input voltage swing/ impedance.
1) Desired plate resistance and work backwards to find V and I at desired plate impedance?
2) Desired plate voltage bias, working backwards to match I with gate swing
3) Start by biasing the gate so the input voltage swing is linear on the plate, then find I based on typical V???
I guess I need a starting point. How about biasing the grid?
If my signal is +/- 1V would I want to bias the grid at -3, or anything above -2 or -3 that matched my desired plate resistance?
Basically, I'm askiing which parameter that one would start with, based on the assumption that one knows the input voltage swing/ impedance.
1) Desired plate resistance and work backwards to find V and I at desired plate impedance?
2) Desired plate voltage bias, working backwards to match I with gate swing
3) Start by biasing the gate so the input voltage swing is linear on the plate, then find I based on typical V???
> I guess I need a starting point.
People have written whole books. And been paid. Why should I duplicate their work?
Yes, today's Whirled-Wide Web is 99% horse-pucky, and it takes time to find the Good Old Stuff (much of which was not so good).
Start with RDH 3rd Chapter 1 (3 Meg PDF file)
RDH 4th Chapter 12 (1.2 Meg PDF file) is more detailed.
People have written whole books. And been paid. Why should I duplicate their work?
Yes, today's Whirled-Wide Web is 99% horse-pucky, and it takes time to find the Good Old Stuff (much of which was not so good).
Start with RDH 3rd Chapter 1 (3 Meg PDF file)
RDH 4th Chapter 12 (1.2 Meg PDF file) is more detailed.
