Hi! This is the current scehmatic ... current values.
Please do not consider yesterday's data, slightly different, especially in the B+.
Currently (V2) the voltage value set by the 1n4148 diodes is slightly different from the value obtained with the 2K resistor. I will update it with right diodes.
Basically it's an "almost" copy of Audioresearch ...
🙂
Please do not consider yesterday's data, slightly different, especially in the B+.
Currently (V2) the voltage value set by the 1n4148 diodes is slightly different from the value obtained with the 2K resistor. I will update it with right diodes.
Basically it's an "almost" copy of Audioresearch ...
🙂
?
Do not consider yesterday's data?
I thought the whole point of this thread was why diode bias on V1 did not work?
Do not consider yesterday's data?
I thought the whole point of this thread was why diode bias on V1 did not work?
I think the forest is being lost 'cuz of all the trees. Post № 3 had the answer, and remains true.
Changing out R8 for a LED significantly changes the behavior of the ECC83 above it; moreover, the feedback loop is essentially 'ruined' as well. Do not do this at all. The value of 1.8 kΩ was chosen for a variety of reasons, and should be left alone.
Changing out whatever resistor was at the second triode's cathode for a pair (in your diagram above) of LED diodes … is going to increase its gain significantly. Also, it will increase its nonlinearity. The counter-side is that the levels involved are fairly low, so the distortion added won't be very annoying. Yet it remains. The diodes increase that stage's gain substantially.
In the end, and I understand you are not really asking for this advice, but I wouldn't have changed either cathode resistor for a LED diode. Phono preamp circuits, especially with their RIAA networks are electronic “gemstones”, which are mathematically designed to significant precision, and then further perfected by measurements, parts-value adjusting, and 'balancing' real-world parts variations with the economics of making hundreds or thousands of the things reliably.
But that's not the advice you are looking for.
Still … it is worthy.
RE-READ post № 3 again. It is true.
⋅-⋅-⋅ Just saying, ⋅-⋅-⋅
⋅-=≡ GoatGuy ✓ ≡=-⋅
Changing out R8 for a LED significantly changes the behavior of the ECC83 above it; moreover, the feedback loop is essentially 'ruined' as well. Do not do this at all. The value of 1.8 kΩ was chosen for a variety of reasons, and should be left alone.
Changing out whatever resistor was at the second triode's cathode for a pair (in your diagram above) of LED diodes … is going to increase its gain significantly. Also, it will increase its nonlinearity. The counter-side is that the levels involved are fairly low, so the distortion added won't be very annoying. Yet it remains. The diodes increase that stage's gain substantially.
In the end, and I understand you are not really asking for this advice, but I wouldn't have changed either cathode resistor for a LED diode. Phono preamp circuits, especially with their RIAA networks are electronic “gemstones”, which are mathematically designed to significant precision, and then further perfected by measurements, parts-value adjusting, and 'balancing' real-world parts variations with the economics of making hundreds or thousands of the things reliably.
But that's not the advice you are looking for.
Still … it is worthy.
RE-READ post № 3 again. It is true.
⋅-⋅-⋅ Just saying, ⋅-⋅-⋅
⋅-=≡ GoatGuy ✓ ≡=-⋅
Well , you put a schematic in which the first stage wasn't in the feedback loop , and the autor didn't say it was wrong ...
Of course with a LED or 2 diodes all the feedback is shorted to ground , no RIAA , no nothing , there must be a resistor . This is not just increased gain , it render the circuit useless
Of course with a LED or 2 diodes all the feedback is shorted to ground , no RIAA , no nothing , there must be a resistor . This is not just increased gain , it render the circuit useless
Last edited:
Hi, forgive me.
I didn't mean don't "totally consider"!
Only the tensions have varied a little.
The problem remains the same as yesterday.
Thanks 🙂
So , for the first stage you can't use a LED or 2 diodes for bias because the AC impedance is almost zero and the feedback network would be shorted to ground .
In post 1, I seem to see an input voltage of 400 mVpp, but 4 mV would
be enough or do you want to show a clipper/limiter ? Do you really want
to debate this or is the scope reading wrong ?
I thought this was about "improving" a phono amp, which was allegedly
a modified EAR834, but in reality it is not.
So you set people on the wrong track from the beginning. Even then,
post 3 had the correct hint by coincidence.
My post 8 already had the final answer to the original question, because
Rk of the first cathode is essential for the modified feedback loop (see last
part of your original post).
Using diode bias on the second valve is a matter of taste.
EAR834 schematic does not apply here, but if you insist on diode bias
of the first tube also it can be done using this very schematic of post 4.
On the oscilloscope measurement posted, I was certainly wrong.
I was using a higher signal to calculate the gain of the preamp as signals of a few mV input are struggling to measure them exactly (with my equipment).
With polarization (V1) with resistance and 400mv input I did not get any output distortion. I then changed the resistance for an IR led and the square wave came out, probable clipping .. due to higher gain.
Later I forgot to lower the input signal to a "real" value and check what I was getting at the output.
Probably it would not distort but it is a test that I do not do as I have understood that Rk of V1 should be left as it is, being part of the riaa ring.
I am listening to the sound with the 1n4148. I'll see if they are to my taste.
Can CCS be implemented on V1 and V2? To see if I get sound improvements?
Many thanks and I apologize again for the inaccuracies but unfortunately it will still be time before it becomes a "professional".
😱
I was using a higher signal to calculate the gain of the preamp as signals of a few mV input are struggling to measure them exactly (with my equipment).
With polarization (V1) with resistance and 400mv input I did not get any output distortion. I then changed the resistance for an IR led and the square wave came out, probable clipping .. due to higher gain.
Later I forgot to lower the input signal to a "real" value and check what I was getting at the output.
Probably it would not distort but it is a test that I do not do as I have understood that Rk of V1 should be left as it is, being part of the riaa ring.
I am listening to the sound with the 1n4148. I'll see if they are to my taste.
Can CCS be implemented on V1 and V2? To see if I get sound improvements?
Many thanks and I apologize again for the inaccuracies but unfortunately it will still be time before it becomes a "professional".
😱If you can't measure the RIAA curve to remain unaffected by those changes better don't do it ... I think Audioresearch is engineered good enough for most people 😀
In part you are right but if I use a reverse riaa circuit I could measure it.
I ask the last question, if you were kind enough to clear my mind.
Using something other than a resistance for the cathode of V2 (as I am doing now) and for the two possible CCS on V1 and V2 instead of the resistors ... they go (all three modifications) to alter - surely - the riaa?
Thanks again. I finished!
I ask the last question, if you were kind enough to clear my mind.
Using something other than a resistance for the cathode of V2 (as I am doing now) and for the two possible CCS on V1 and V2 instead of the resistors ... they go (all three modifications) to alter - surely - the riaa?
Thanks again. I finished!
Copy the AR circuit identically from the SP3 or SP6 and only mess with it when you become a better designer than the AR engineers. It will save you a lot of grief.
I get the impression that clipping is occurring due to excessive voltage again. There is no cathode degeneration when using a LED in lieu of a resistor to bias up the tube, so the tube running wide open.
In part you are right but if I use a reverse riaa circuit I could measure it.
I ask the last question, if you were kind enough to clear my mind.
Using something other than a resistance for the cathode of V2 (as I am doing now) and for the two possible CCS on V1 and V2 instead of the resistors ... they go (all three modifications) to alter - surely - the riaa?
Thanks again. I finished!
At first glance it shouldn't 😀 , but ... you can't be 100% certain that something small wouldn't change , only if you check it .
Last edited:
Thanks to you and your knowledge. 😉
I'll try to equip myself with one or two reverse riaa circuits and start learning how to make these measurements.
I also say two to have comparisons, I don't think they're expensive.
If you have one to recommend, trustworthy, thank you very much, otherwise I look for a bit on the net.
Good evening or good morning!
PS: I started with tubes with the most complicated thing: a prephono! the rest goes downhill ahahhaah
😀
I'll try to equip myself with one or two reverse riaa circuits and start learning how to make these measurements.
I also say two to have comparisons, I don't think they're expensive.
If you have one to recommend, trustworthy, thank you very much, otherwise I look for a bit on the net.
Good evening or good morning!
PS: I started with tubes with the most complicated thing: a prephono! the rest goes downhill ahahhaah
😀
It is obvious why it does not work. Please put a big capacitor over the LED's and test again (just for fun).
Hi, ok, just for fun, I try! 🙂 Capacitor as big as? You reverence the led / diodes of the V1 stage, right? Why the V2 goes ... Thanks!
That is simple, the LED acts as a Zener, but a Zener will only work if the voltage is higher then the conducting theshold of this Zener. This is true for DC. But your tube is getting an AC signal and the voltage will drop under the threshold sometimes. The LED will stop conducting and strange behavior is to be expected. By adding a 100uF capacitor the voltage will become a DC voltage and the LED will not stop conducting. You could also create a DC bias from the heater supply to make sure the LED will not get "starved"
Ok, thanks, yes I understand what you mean.
But I think this is just a test out of curiosity or could it even sound good? Because LEDs or diodes with a cap in parallel .. in these uses .. I have never seen them.
The cap should be of quality I guess.
The situation is the same as when you put a cap parallel to the Rk ... and therefore it must be excellent and of great quality? Or am I wrong?
As for letting more current flow in the led ...... it was known that I could draw current through a resistor from the B + and let it pass through the led.
Are you telling me that it can also be done with the 12.6v or the 6.3v heater of the same valve? Doing so would be better. Less waste of anode current. 🙂
thanks for your interest and advice. 😉
But I think this is just a test out of curiosity or could it even sound good? Because LEDs or diodes with a cap in parallel .. in these uses .. I have never seen them.
The cap should be of quality I guess.
The situation is the same as when you put a cap parallel to the Rk ... and therefore it must be excellent and of great quality? Or am I wrong?
As for letting more current flow in the led ...... it was known that I could draw current through a resistor from the B + and let it pass through the led.
Are you telling me that it can also be done with the 12.6v or the 6.3v heater of the same valve? Doing so would be better. Less waste of anode current. 🙂
thanks for your interest and advice. 😉
The point is that you need current flow, but this could add noise. As mentioned also in an earlier post, it could be that the gain is too high of the first stage now. If you offer a smaller signal to this stage will the clipping disappear? The 12AX7 is a high gain but low current tube. So replacing the resistor with diodes will be similar to adding a capacitor over the resistor. If this works fine then the starving issue could be your problem. Do not forget that all devices have a capacitance and even oscillation could occur especially in a high gain stage.
Last edited:
Thanks, I understand.
When I tried LEDs and diodes for the first time on v1 (with a normal Rk it has about 0.5mA), but, not working, I already tried to give about an additional 1.3mA through a 220k R from the B +.
However, I had mistakenly set a huge input level (400mV).
Since I'm curious ... I do the test again with a normal input signal and see. Then I try as you told me, with the capacitor.
You are absolutely right about the swing speech.
Now there are no wobbles and everything is stable as a rock, but it is also easy, by changing "things" ... to occur.
For your curiosity .. I'll let you know.
Thanks again. 🙄
When I tried LEDs and diodes for the first time on v1 (with a normal Rk it has about 0.5mA), but, not working, I already tried to give about an additional 1.3mA through a 220k R from the B +.
However, I had mistakenly set a huge input level (400mV).
Since I'm curious ... I do the test again with a normal input signal and see. Then I try as you told me, with the capacitor.
You are absolutely right about the swing speech.
Now there are no wobbles and everything is stable as a rock, but it is also easy, by changing "things" ... to occur.
For your curiosity .. I'll let you know.
Thanks again. 🙄