Member
Joined 2009
Paid Member
So there is a benefit
Are there any JFETs with nice specs that are still readily available? I've been trying to source some 2SK170 devices to complete a pair of Rudy's SYMASYM boards but these parts have reached "unobtainium" status with the high prices that ensue.
I might play with the idea of a singleton JFET fed input in my Quasi complimentary project thread.
Hi Ranchu
Just got some 2sk170BL from Reichelt at very reasonable prices..... I believe this is going in my preferred direction....
Thank you so much, I will start by sim the simple solution from post 266.
If it goes ok in my schematic, I will build it.
I read Ranchu had some trouble inittialy and got the solution with some judicious mod but I can not recall what it was....
@ Ranchu
Would you please point me any possible issues with the schematic from post 266 ?
PS: In my sim I needed to reduce gain by playing with the feedback resistor values because the output buffer is not too strong.
As I recall the only issue I had with the amplifier section itself was when testing with low 25v rails and I needed to adjust some R values under guidance of Bigun. The speaker protection section has/had a few issues that are resolved now for the most part.
I'll double check my notes tonight and let you know if I remember anything else.
I'll double check my notes tonight and let you know if I remember anything else.
Member
Joined 2009
Paid Member
Member
Joined 2009
Paid Member
In my mind, for TGM8, there's not requirement to use 2N5551 - you just need a modern small signal silicon transistor with a voltage rating adequate for use and if you want to use it in the same pcb then it should have the same pin-out.
The BC337 has a max Vce of only 45V, a lot less than the 2N5551 so I'd be nervous about that depending where in the circuit you will use it. The VAS experiences the larges range in voltages, the input device should normally only see one rail voltage. You want some margin of safety too so you'll be very limited in rail voltage. With the 2 x 32V transformer you should find a different transistor.
The BC337 has a max Vce of only 45V, a lot less than the 2N5551 so I'd be nervous about that depending where in the circuit you will use it. The VAS experiences the larges range in voltages, the input device should normally only see one rail voltage. You want some margin of safety too so you'll be very limited in rail voltage. With the 2 x 32V transformer you should find a different transistor.
Are you referring to Destroyer DX?
DX is a ClassA non believer, he has stated repeatedly that ClassA does not perform well enough for him and that it consumes too much power for Brazil.
I can't believe DX has posted a sch of his own making, for a ClassA amplifier.
Member
Joined 2009
Paid Member
Yes, Destroyer DX.
He's a man of many talents so I don't doubt he could build a nice Class A amplifier if he wants - not sure why his earlier forays into Class A were not satisfying but I suspect that the desire for power leads to Class AB more often than not. He'd be right that a hot amp is not ideal for a hot country, but then again, this hobby exerts a powerful addiction over us
I sent him some triodes a few years back - maybe we'll see a DX hybrid amp one day.
He's a man of many talents so I don't doubt he could build a nice Class A amplifier if he wants - not sure why his earlier forays into Class A were not satisfying but I suspect that the desire for power leads to Class AB more often than not. He'd be right that a hot amp is not ideal for a hot country, but then again, this hobby exerts a powerful addiction over us
I sent him some triodes a few years back - maybe we'll see a DX hybrid amp one day.
I have some 2SC2240GR that I got from a good friend... please let me know if it is a suitable contender.
Member
Joined 2009
Paid Member
Member
Joined 2009
Paid Member
Where are you probing the input offset ? the input transistor base current has to flow through R5 and R6 (in your schematic attached to your post). A few uA of base current will develop a voltage drop across those resistors of a few mV which will appear at the base, but C3 blocks it from appearing at the source. I don't see any issue with that.
I would like to omit the input cap... I know it is possible by choosing the correct input impedance.
In this case I am using 10k because it suits my preamp but if I choose 27k as per your schematic I alter output offset and get even higher input offset.
I am probing at the base of Q4 in my schematic
In this case I am using 10k because it suits my preamp but if I choose 27k as per your schematic I alter output offset and get even higher input offset.
I am probing at the base of Q4 in my schematic
Member
Joined 2009
Paid Member
It's a bit more complicated but not much.
What you care about is the dc offset at the output because it is going to cause a dc current to flow in your speaker - so you want it to remain relatively small - I think +/-100mV is safe but since most amps can be set up to keep the offset at the output to less than +/-10mV we often choose to do so. Also, if you accidentally short the output of the amplifier to ground without any music playing then having almost zero offset limits the current flow through the short - it might save the amplifier from damage.
The Singleton input creates a dc voltage difference between input and output unless steps are taken to null it out. As you now know, VR1 is used to null out this difference so that for a fixed dc voltage at the base, the amplifier output sits close to ground. The amp is designed so that this set-up will be maintained fairly accurately over temperature. However, if the dc at the input changes then this null condition is lost.
The shunt capacitor in the feedback loop ensures that the amplifier has a gain of '1' at zero frequency, i.e. dc. This means that a change in the dc-offset at the input will create a change in the dc-offset at the output of the same magnitude. For example, say there was -34mV at the input and you adjusted VR1 to set zero dc at the output, then for some reason you force the dc offset at the input to changes to the new value of + 10mV. This is an increase of 44mV in the dc level at the input and we would find the amplifier will produce a corresponding change in the dc offset at the output of +44mV.
Let me note that I don't consider this a problem, 44mV is nothing to a normal speaker.
Now let's think about what happens when we remove the input capacitor. The input is now dc-connected to the RCA output connector on your source. However, most, if not all, sources usually include a capacitor in their output. This cap is inside the box so you don't know it's there and so in most cases removing the input capacitor from the amplifier produces no real change in dc-conditions. However, if the source has no capacitor in the output then your amplifier is exposed to whatever dc level the source is sitting at.
Let me note that I am happy to check all my sources and would be perfectly happy not to have an input capacitor on any of my DIY amplifiers.
A source without an output capacitor likely offers a relatively low dc path to ground for the base current to your Singleton input amplifier which means the base current will not have much resistance to develop any voltage across and it will be close to zero. When you unplug the source the base current now has to flow through R5 and the dc offset will jump back to where it was before you plugged in the source. A similar situation could arise if the source has an output transformer (e.g. a tube based line amp) in which the transformer provides a different resistance to ground than R5. This will likely be inaudible since the change in voltage at the output will be tiny. If the source has significant dc on it's output due to a fault condition then it can put some nasty dc onto the front end of your amplifier - this is the risk that makes people want to see an input capacitor on an amplifier so I put space for one on the TGM8 pcb.
If you don't want to include one in your design and later you wish you had, it's easy enough to wire one in line between the amplifier and the RCA input socket - in fact you'd likely have more space and could use something fancy.
I've been rambling, did I even answer your question !
What you care about is the dc offset at the output because it is going to cause a dc current to flow in your speaker - so you want it to remain relatively small - I think +/-100mV is safe but since most amps can be set up to keep the offset at the output to less than +/-10mV we often choose to do so. Also, if you accidentally short the output of the amplifier to ground without any music playing then having almost zero offset limits the current flow through the short - it might save the amplifier from damage.
The Singleton input creates a dc voltage difference between input and output unless steps are taken to null it out. As you now know, VR1 is used to null out this difference so that for a fixed dc voltage at the base, the amplifier output sits close to ground. The amp is designed so that this set-up will be maintained fairly accurately over temperature. However, if the dc at the input changes then this null condition is lost.
The shunt capacitor in the feedback loop ensures that the amplifier has a gain of '1' at zero frequency, i.e. dc. This means that a change in the dc-offset at the input will create a change in the dc-offset at the output of the same magnitude. For example, say there was -34mV at the input and you adjusted VR1 to set zero dc at the output, then for some reason you force the dc offset at the input to changes to the new value of + 10mV. This is an increase of 44mV in the dc level at the input and we would find the amplifier will produce a corresponding change in the dc offset at the output of +44mV.
Let me note that I don't consider this a problem, 44mV is nothing to a normal speaker.
Now let's think about what happens when we remove the input capacitor. The input is now dc-connected to the RCA output connector on your source. However, most, if not all, sources usually include a capacitor in their output. This cap is inside the box so you don't know it's there and so in most cases removing the input capacitor from the amplifier produces no real change in dc-conditions. However, if the source has no capacitor in the output then your amplifier is exposed to whatever dc level the source is sitting at.
Let me note that I am happy to check all my sources and would be perfectly happy not to have an input capacitor on any of my DIY amplifiers.
A source without an output capacitor likely offers a relatively low dc path to ground for the base current to your Singleton input amplifier which means the base current will not have much resistance to develop any voltage across and it will be close to zero. When you unplug the source the base current now has to flow through R5 and the dc offset will jump back to where it was before you plugged in the source. A similar situation could arise if the source has an output transformer (e.g. a tube based line amp) in which the transformer provides a different resistance to ground than R5. This will likely be inaudible since the change in voltage at the output will be tiny. If the source has significant dc on it's output due to a fault condition then it can put some nasty dc onto the front end of your amplifier - this is the risk that makes people want to see an input capacitor on an amplifier so I put space for one on the TGM8 pcb.
If you don't want to include one in your design and later you wish you had, it's easy enough to wire one in line between the amplifier and the RCA input socket - in fact you'd likely have more space and could use something fancy.
I've been rambling, did I even answer your question !
Last edited:
Yes, you made it completely clear now. Thank you so much.
So the only real problem is if my dc coupled preamp develops a fault, it might inject dc into the amp and that might damage the speakers.
my preamp is a dcb1 without any coupling cap.... never had any issues but if one of the psu's shts down, there will be considerable output dc.
The input cap is mandatory in this amp.
So the only real problem is if my dc coupled preamp develops a fault, it might inject dc into the amp and that might damage the speakers.
my preamp is a dcb1 without any coupling cap.... never had any issues but if one of the psu's shts down, there will be considerable output dc.
The input cap is mandatory in this amp.
Yes.
The input bias currents flow through resistors that are DC coupled to the +IN and -IN pins.
If you add another DC route with a low DC resistance (by deleting a DC blocking capacitor), then the unbalanced input offset voltages result in an output offset.
It is the balancing of the input offset VOLTAGES, that results in near zero OUTPUT OFFSET.
If you can achieve this with near equal input bias currents feeding through near equal DC resistances then that ensures near equal Pq in the input devices and results in low levels of output offset drift with changing chip temperatures. Low drift indicates good input balancing.