scope readings

[east electronics]

this is the out put of one preamplifier when i feed it with square wave 10 khz 200mv

please notice the rounding in the corners and if possible explain me what does it mean .....

forum member from tube area said to me that this represents a loss of 0.1db @20khz ....

is that so or is it worst ?????

in a good preamp shouldnt it be perfect square to at least 30KHZ ???

thanks

 

[Elvee]

It looks like a nice, clean first order roll-off. The kind of response you expect from an RC placed at the input to limit the ingress of RF signals and similar trouble makers. Nothing abnormal IMHO.

 

[bear]

we need to see the input square wave image for comparison...

preferably on the same screen, although not essential.
_-_-

 

[east electronics]

bear .....

of course in the input there is a totally square image ....what you expect ..... ( normaly signals like that when coming from average signal generators are expected to be totally square ....)

in the input ....there is nothing ....since this is a tube preamp there is no r-c to restrict the input signal ..... just 100k resistor to signal connection with 100k resistor to ground .....

 

[wg_ski]

Re: bear .....

in the input ....there is nothing ....since this is a tube preamp there is no r-c to restrict the input signal ..... just 100k resistor to signal connection with 100k resistor to ground .....

Tubes, like semiconductors, have capacitors hidden inside them. And the circuit may or may not have NFB, and therefore makes its own RF/ultrasonic filter - probably some pesky 100k plate resistor and a few pF somewhere.

 

[east electronics]

very well .....

in this case can you ensure me about that rounding in the corners ????

does it really represent a los of 0.1db at 20khz ???or is it more ???

 

[Salas]

Looks like being fed by large source impedance, creating a rather large time constant due to Miller. I have made 12AY7 line circuits and they go much higher.

 

[unclejed613]

a 3db loss would have rounding that intersected the flat top at about 70% (actually 67%)of the way through the square wave, just as a way of comparison. the 0.1db estimate is probably very close.

 

[wg_ski]

Re: very well .....

in this case can you ensure me about that rounding in the corners ????

does it really represent a los of 0.1db at 20khz ???or is it more ???

You've got a scope and a signal generator - measure it. If you have trouble resolving 0.1dB, measure where it's down 1dB and back-calculate where the 0.1dB point would be assuming it's a singe-time-constant roll off.

 

[east electronics]

well salas ....

thanks for join in .... the circuit is fairly simple ...please take a look and give us your sugestion if you have any ....

thanks

 

[east electronics]

small mistake

the 470k resistor is located after the output capacitor

 

[Salas]

Well, if you really need to halve the input signal, try a 25k resistor before the pot, and also try a 25k pot instead of 100k. That will wishfully keep the input resistive grid high enough still but better for response. Run the 10kHz square again and see if it got noticeably sharper corners.

 

[Salas]

Also see to attach a 1k carbon directly with short leg to upper section's grid (7) because some fuzziness on the flat top of your scope trace indicates possible oscillation.

 

[east electronics]

thanks salas

meanwhile ....i tryied many variations and a few diferent circuits other type of topology and all of them had more or less similar performance

also meanwhile i had a responce from tube forum and a member said to me that 12AY7 will be the last thing to use if performance is the target ....

so salas if you have a circuit for a 12AY7 tube i will be more than happy to have it or at least to try it ....

to goal is a simple line stage preamp based on 12AY7 tubes

thanks

PS as bout the oscilation that will be the last thing to bother with since the all construction is kinda messy .... i will have to shape it up and then see for oscillation issues ( but thanks i will keep that in mind also )

 

[Salas]

He misunderstood 12AY7 for 12AU7. Your choice (6072) is much more linear than what he thought you use. Try the lower impedance input resistor grid that I suggested first.

 

[nikwal]

For fun you could try this circuit..
http://www.geocities.com/rjm003.geo/rjmaudio/images/twi_mu_stage.gif
irf710 maybe, used it in a hybrid amp I made..

 

[east electronics]

did that allready

has absolutelly no effect

as we speak i modify this circuit and seems to perform a lot better and now square curve is almost fine up to 20khz

please see att

 

[Salas]

So it was something with the upper section...

 

[east electronics]

not really

this is a totaly diferent circuit than the one i worked with originally

 

[east electronics]

this is

the final schematic used and seems that performs very fine ....and square wave is very fine for at least 20khz

comments please