This a new linestage preamp using EF86 & 12AX7.
What do you think?
What do you think?
An externally hosted image should be here but it was not working when we last tested it.
Hi Edwind, Welcome to the forum!
In terms of your schematic I am wondering what the intended purpose would be? The gain for a line stage application will be very high. I've not done an analysis of this design, but generally gains of well under 20dB are the norm for modern designs, and in most cases less than 10dB is a reasonable choice.. (I would guess the gain here could be 60dB - in phono stage territory.)
The output impedance will be extremely high as well which will make it a poor line driver..
The input stage has limited signal handling capability, I suspect a couple of hundred mV would cause this design to clip as there is no way of dealing with the very high signal levels of digital sources (typically 2Vrms @ FS) and even most analog sources..
Filament circuit should not allow current to flow through chassis/circuit ground as this is an excellent recipe for hum.
In terms of your schematic I am wondering what the intended purpose would be? The gain for a line stage application will be very high. I've not done an analysis of this design, but generally gains of well under 20dB are the norm for modern designs, and in most cases less than 10dB is a reasonable choice.. (I would guess the gain here could be 60dB - in phono stage territory.)
The output impedance will be extremely high as well which will make it a poor line driver..
The input stage has limited signal handling capability, I suspect a couple of hundred mV would cause this design to clip as there is no way of dealing with the very high signal levels of digital sources (typically 2Vrms @ FS) and even most analog sources..
Filament circuit should not allow current to flow through chassis/circuit ground as this is an excellent recipe for hum.
Hi, kevinkr annd DF96 (is it tube type?).
This are the gain for every tube that use: +22dB (12AT7), +16dB (12AU7), +21dB (12AX7).
This is low quality circuit, fun for beginners (low cost) and a little homeworks for experts
I already build this. Input directly from CD player.
There is a very little of hum, noise, and distortion (for my ear, may be no for others).
This circuit not like Quad II phono stage.
The output impedance still to high since it's use common cathode and may add a voltage follower to boost the output signal.
Many-many thanks for advices, it will give a positive feedback for next circuit.
By the way, will somebody upload the new scheme for upgraded version of this circuit?
This are the gain for every tube that use: +22dB (12AT7), +16dB (12AU7), +21dB (12AX7).
This is low quality circuit, fun for beginners (low cost) and a little homeworks for experts
I already build this. Input directly from CD player.
There is a very little of hum, noise, and distortion (for my ear, may be no for others).
This circuit not like Quad II phono stage.
The output impedance still to high since it's use common cathode and may add a voltage follower to boost the output signal.
Many-many thanks for advices, it will give a positive feedback for next circuit.
By the way, will somebody upload the new scheme for upgraded version of this circuit?
Only in a guitar amp would someone think that ECC81,82 and 83 are directly interchangeable. Beginners can have much more fun with sensible circuits; cheaper too as you would not have to throw it away once you know better.
As a line stage this circuit is simply wrong. I will leave it to you to do the homework. It can't be upgraded, only replaced. Sorry to be blunt, but you asked what we think.
As a line stage this circuit is simply wrong. I will leave it to you to do the homework. It can't be upgraded, only replaced. Sorry to be blunt, but you asked what we think.
Hi Edwind,
As DF96 says the circuit is useless. But there is actually a solution for making it work:
1 Move the volumecontrol to the input
2 Triode-strap the EF86 Ra=47k Rk=820
3 Make the second stage a CF, DC connected to the first, and use ECC82.
The triodestrapped EF86 sounds great as lineamp.
As DF96 says the circuit is useless. But there is actually a solution for making it work:
1 Move the volumecontrol to the input
2 Triode-strap the EF86 Ra=47k Rk=820
3 Make the second stage a CF, DC connected to the first, and use ECC82.
The triodestrapped EF86 sounds great as lineamp.
It's been awhile since I've seen a more ridiculous design. Cascading a small signal pentode into a 12AX7 (high gain triode) is going to give one helluvalot of open loop gain. That would be OK if you needed that for NFB, but this isn't an NFB design.
Secondly, this has a -20bdv attenuator up front. Why include all that open loop gain, only to throw it away? That makes no sense at all.
Finally, driving the output directly from the 12AX7 plates is just asking for trouble. Long runs of cable, an amp with a smallish Zi, are going to load the 12AX7 right into distortion and nonlinearity. Triodes like light plate loading, and the output isn't buffered.
Take a pass on this one.
Secondly, this has a -20bdv attenuator up front. Why include all that open loop gain, only to throw it away? That makes no sense at all.
Finally, driving the output directly from the 12AX7 plates is just asking for trouble. Long runs of cable, an amp with a smallish Zi, are going to load the 12AX7 right into distortion and nonlinearity. Triodes like light plate loading, and the output isn't buffered.
Take a pass on this one.
I agree with the most said before from the other guys , and something more you include all that gain without any negative feedback , and you attenuated it in the input , but you still have high level of noise due to large amplification ( EF86 plus 12AX7 ) and due to partition noise of the EF86 in pentode mode amplifier .
20K, alone, is a near vertical loadline for a 12AX7, and is way to small. Add in cable capacitance, and it'll roll over and die. Find a better design.
For a short cable the low input impedance of the power amp will rescue the frequency response, but this is not the way to do audio!
Instead of fumbling around trying to patch up this design, why not try to understand all the things which are wrong with it and then start again, with your new knowledge?
Instead of fumbling around trying to patch up this design, why not try to understand all the things which are wrong with it and then start again, with your new knowledge?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.