Hi Disco!
Your cathode bypass cap on the output stage is way yoo small. You have to calculate it from the actual resistance at the cathode. That would be something like 750 ohm in parallel with ri/mu. I would use 470 or 1000 uF. The value you have used spoils the dampingfactor at low frequences. That could explain some of the loudness and boominess. Also try to put a resistor on top of the driver instead of SRPP. That might give you a little more swing.
Thorsten
Your cathode bypass cap on the output stage is way yoo small. You have to calculate it from the actual resistance at the cathode. That would be something like 750 ohm in parallel with ri/mu. I would use 470 or 1000 uF. The value you have used spoils the dampingfactor at low frequences. That could explain some of the loudness and boominess. Also try to put a resistor on top of the driver instead of SRPP. That might give you a little more swing.
Thorsten
Thorsten, thanks for the input. I followed your suggestion but the peculiar reproduction is not caused by the bypass cap. You are right indeed:
To obtain full gain, good amplitude response and to minimize the effects on phase and transient response, the feedback voltage over the cathode must be shunted by a bypass capacitor of low reactance up to 1 Hz. This cap forms a LPF with the output resistance at the cathode, which consists of Rk (the bias resistor) parallel to rk (the cathode resistance).
I haven't actually measured the relationship between Ug and Ua but from the many graphics it would come down to a mu of 2.5 or something close for my operating point. So, filling in the formulae rk = 1200+600/3.5 = 514 ohm, parallel to 750 ohm results in 305 ohm. Ck = 1/6.28x305 = 522 uF.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
To obtain full gain, good amplitude response and to minimize the effects on phase and transient response, the feedback voltage over the cathode must be shunted by a bypass capacitor of low reactance up to 1 Hz. This cap forms a LPF with the output resistance at the cathode, which consists of Rk (the bias resistor) parallel to rk (the cathode resistance).
I haven't actually measured the relationship between Ug and Ua but from the many graphics it would come down to a mu of 2.5 or something close for my operating point. So, filling in the formulae rk = 1200+600/3.5 = 514 ohm, parallel to 750 ohm results in 305 ohm. Ck = 1/6.28x305 = 522 uF.
That Bartola blog doesn't mention how the tubes were tested. It could be that the 6N6P was measured with a low plate current, since it really does like to see at least 15mA per triode. With 6N6P, I think distortion would be higher with 10mA or less.
There seems to be all sorts of contradictory data on 6N6P distortion characteristics. Some say it has high THD, many others say low.
?
--
Hey Rongon,
Here are my tests:
Valve Ia [ma] Va[V] Vg [V] THD
6N6P 27.5 271 -10 0.11%
6N6P-I 13 305 -15 0.13%
6N6P test point was well over Pa so should retake tests....
Enjoy
Ale
Do you remember how you loaded the tube for testing? Plate load (resistor, transformer, active?), cathode load, cathode resistor with bypass cap, not bypassed, etc? Just curious...
I usually use 6N6P at more like Va = 150V, Ia = 15mA, Vg = -5V. I wonder if that would change the distortion measurements by much...
Thanks!
When using proper 'audio' tubes I agree. Some 'crap' tubes on the other hand need a lot of correction.
Here's my second attempt with these house warmers, with an alternate steering.
The plan is to use the fine EC86 (E86C) for amplification and to boost it with the shunt cascode.
I'm aiming for this setting:
An externally hosted image should be here but it was not working when we last tested it.
but I'm still in the tweaking process, fiddling with resistor values.
Last edited:
Hi Rod, thank you for the support.
The OPT is a 150VA EI-model from China, seen in the above pictures.
I've no data on it but from what I gathered it's difficult to kill bandwidth
on low impedance transformers
It has decent bass control.For study I used this temporary setup, if results are encouraging I'll rebuild
it with all respect to HF properties.
This is the actual setup (FB has not been established yet):
An externally hosted image should be here but it was not working when we last tested it.
Base voltage is for now derived from a bypassed voltage devider on B+
Without cathode bypass cap and 10K collector resistor amplification is 47 times.
The amp isn't sounding half as bad as I recalled, can't wait to raise amplification and apply FB.
Last edited:
Bypassing only the lower cathode resistor gives 62 times amplification.
As a result output impedance sinks. This has impact on LF response as
the bass roll off frequency drops. Reproduction cleared up immediately,
giving powerfull yet detailed bass notes.
This was the one positive aspect of the previous build I recalled but now
it seems to be lacking the previously mentioned distortion. I like it so far
An externally hosted image should be here but it was not working when we last tested it.
As a result output impedance sinks. This has impact on LF response as
the bass roll off frequency drops. Reproduction cleared up immediately,
giving powerfull yet detailed bass notes.
This was the one positive aspect of the previous build I recalled but now
it seems to be lacking the previously mentioned distortion. I like it so far
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.