Sorry, confusion for me.
Tell me if i'm wrong,please:
3,5k push-pull =3,5kΩ plate to plate, so 1,75kΩ for one plate?
Tell me if i'm wrong,please:
3,5k push-pull =3,5kΩ plate to plate, so 1,75kΩ for one plate?
I run them at 520V (triode) and 100mA. Yes, 52W disipation. They last more then 3 years...
Yes in this case since GU50s operate in class A.
As soon as one output tube goes to cutoff, the load impedance is 3k5/4.
Last edited:
Thanks.
In this case my project is viable:
300v .
3,5k 50w push-pull transformer
Gu50 triode strapped
Thanks but i don't need so much voltage.
A such voltage transformer is expensive.
I have an amish budget.
Don't forget the gu50 is the 300b of the poor man.
GU-50 has Ra of about 900-1000 Ohm at this OP, 1.75k load per tube is kinda low
I would go for 5k (2.5k per side) or above
Hello TG.
What do you mean with "OP",please ?
When i look the gu50 triode strapped curves,1,75k looks good. Even 1,5k seems be better to me.
But can you me explain please ?
What do you mean with "OP",please ?
When i look the gu50 triode strapped curves,1,75k looks good. Even 1,5k seems be better to me.
But can you me explain please ?
OP is operating point (Ua, Ia).
Plate curves might look good in terms of power, but low load values will cause poor speaker damping - like 6-7 Ohm (depending on the OPT's primaty and secondary DC resistance) of output impedance for 8 Ohm speaker tap.
That might work OK for some types of speakers, but for conventional modern ones it most likely won't.
Plate curves might look good in terms of power, but low load values will cause poor speaker damping - like 6-7 Ohm (depending on the OPT's primaty and secondary DC resistance) of output impedance for 8 Ohm speaker tap.
That might work OK for some types of speakers, but for conventional modern ones it most likely won't.
I laid out some loadlines on FL152 triode curves at Va = 375V, Ia = 65mA (attached). They show single-ended lines. As I understand it, for push-pull operation in class A, double the impedance (a 3.5k ohm SE loadline means a 7k plate-plate Push-Pull OPT).
Assuming FL152 characteristics are close to the same as for GU50...
As I see it, GU50 in triode needs a pretty high primary impedance compared to the more commonly used power tubes like EL34-triode, 6L6-triode, KT88-triode, 300B, 2A3. To me, it looks like you would want to use a 10k p-p OPT for GU50 push-pull class A operation. You'd also want that OPT to be capable of delivering at least 30W before saturating. That's a tall order. (Higher impedance OPTs are more challenging to make perform well than lower impedance ones.)
Take a look at the 1750 ohm loadline (3.5k p-p for push-pull). That does not look good to me at all. Nearly vertical, which suggests high levels of odd harmonics in the distortion spectra.
Do you agree, or did I get that all wrong?
--
Assuming FL152 characteristics are close to the same as for GU50...
As I see it, GU50 in triode needs a pretty high primary impedance compared to the more commonly used power tubes like EL34-triode, 6L6-triode, KT88-triode, 300B, 2A3. To me, it looks like you would want to use a 10k p-p OPT for GU50 push-pull class A operation. You'd also want that OPT to be capable of delivering at least 30W before saturating. That's a tall order. (Higher impedance OPTs are more challenging to make perform well than lower impedance ones.)
Take a look at the 1750 ohm loadline (3.5k p-p for push-pull). That does not look good to me at all. Nearly vertical, which suggests high levels of odd harmonics in the distortion spectra.
Do you agree, or did I get that all wrong?
--
Attachments
But how close is the sound? Anatoliy suggests 4P1L is better for PSE use, and he knows these tubes well.
First thanks for your pains.
But perhaps i missed something or i don't understand how push-pull amps are working but the voltage at input B+ is 300v .
Working point will be near 90ma 150v vg=-10v
With 100Ω at screen.
R=1,75k
Last edited:
1) The OPT stores energy, so you plot the loadline differently than you do using a load resistor. Here's a good basic explanation of single-ended tube output stages:
The Valve Wizard -Single Ended
And here's the push-pull explanation:
The Valve Wizard -Push-Pull
2) If your B+ is 300V, then the voltage at the GU50 plates will be 300V minus whatever drop there is across the primary resistance of the OPT.
If you're using cathode bias (self-bias) then you'll need to subtract the cathode-to-ground voltage from the GU50 plate voltage (actually the plate-to-cathode voltage).
So, with a B+ of 300V, minus (let's say) 5V dropped across the OPT primary resistance, that will now leave 295V at the GU50 plates.
3) With -10V bias, and assuming you're using cathode bias, the GU50 plate-to-cathode voltage (Va) will be 285V.
Where did you get operating point of Va = 150V and Ia = 90mA?
--
Thanks for explanation.
Wavebourn had show the Gu50 triode curves with 100Ω resistor at screen in this website.
I can't post the picture to show you.
Wavebourn had show the Gu50 triode curves with 100Ω resistor at screen in this website.
I can't post the picture to show you.
Not really a problem, it's just far from optimal.
GU-50 is a cheap high power tube, and it really makes sense to use it at high power only - like 400V/100mA or something like that.
There are a plenty of 12W tubes (say, EL84 and 6V6 families) that are much better sonically and require half of the heater power.
GU-50 is a cheap high power tube, and it really makes sense to use it at high power only - like 400V/100mA or something like that.
There are a plenty of 12W tubes (say, EL84 and 6V6 families) that are much better sonically and require half of the heater power.