I think as Ian mentioned the double LTP is generally used for driving a 'push-pull' VAS. The symmetry, if well executed, seems to offer what some might say a palatable character to the sound. A certain amount of harmonic cancellation can be achieved this way. Cons as I see them really amount to added complexity. One could achieve really nice results without basically doing everything in the front end x2.
Thanks Ian .
This amplifier is from a budget range. yes the layout is quite symetrical too. if the componets where soldered in with some care it wouldlook pretty good. Asafter a few mods it is sounding much better . i have disconected the Preamp / control PCB as it was noisey and strangled the sound . I have also removed the dc blocking cap from the input of the power amp as my dac has no or ver yvery little dc on its output. That made a big diffrence most noticable was a much welcome reduction in sibulance type distortion .
This amplifier is from a budget range. yes the layout is quite symetrical too. if the componets where soldered in with some care it wouldlook pretty good. Asafter a few mods it is sounding much better . i have disconected the Preamp / control PCB as it was noisey and strangled the sound . I have also removed the dc blocking cap from the input of the power amp as my dac has no or ver yvery little dc on its output. That made a big diffrence most noticable was a much welcome reduction in sibulance type distortion .
Hi Jason
Added complexity normally means added cost or so i thought. Im guessing it is an attempt to steal away business from the more prominent manufacturers. inside does look quite tidy compared to others i have seen in this price range ... A couple of questions . This amp runs cool very cool and the heatsink is very small too . I suspect it is running a quite low bias . maybe it could be increased . Also i dont see any compensation capacitors well that is if i ignore the two 100nF film caps . if the 100nF caps are for compensation are they not quite high in value as normslly i see the caps with only say 30 pF in these. C4 and C5 are the capacitors in question.
Added complexity normally means added cost or so i thought. Im guessing it is an attempt to steal away business from the more prominent manufacturers. inside does look quite tidy compared to others i have seen in this price range ... A couple of questions . This amp runs cool very cool and the heatsink is very small too . I suspect it is running a quite low bias . maybe it could be increased . Also i dont see any compensation capacitors well that is if i ignore the two 100nF film caps . if the 100nF caps are for compensation are they not quite high in value as normslly i see the caps with only say 30 pF in these. C4 and C5 are the capacitors in question.
Benefits of complementary LTP.
1. Potential for reduction in input noise by up to 3dB.
2. Can be easier to control DC offset with cancellation of input currents - dependant on effort to match complementary Hfes.
3. Potential to reduce distortion.
4. Easy way to drive push-pull VAS.
5. It looks nice - to many people at least
There are drawbacks too. See Bob Cordell's book for a more detailed analysis. He is not a real fan of this layout.
Best wishes
David
1. Potential for reduction in input noise by up to 3dB.
2. Can be easier to control DC offset with cancellation of input currents - dependant on effort to match complementary Hfes.
3. Potential to reduce distortion.
4. Easy way to drive push-pull VAS.
5. It looks nice - to many people at least
There are drawbacks too. See Bob Cordell's book for a more detailed analysis. He is not a real fan of this layout.
Best wishes
David
Small signal stage costs are generally small compared to power supply, heat sink and chassis. It adds to the aesthetic and make some people feel all warm and fuzzy inside.
Q8 looks like the emitter and collector are reversed. C4 and C5 would be plain old miller compensation caps and 100pF isn't necessarily out of line, depending on LTP and VAS currents. The emitter resistor values in the VAS degeneration also don't appear correct, and are not equal.
Also absent is any emitter degeneration of the LTP and an input filter. The feedback network values look suspect (40dB gain) and there is no input resistor to allow for bias currents.
As for bias, it may well be somewhat low. If the heat sink is particularly small I wouldn't try to bias it too heavily. Typical values for 'optimal bias' puts about 26mV across each resistor R26 and R23. Trying to make the class A region too big will result in a hot amplifier and gm doubling which some say is every bit as bad as under-biasing.
Is it safe to assume you have drawn this by tracing the physical circuit? That might explain some of the discrepancies...
Discrepencies
As Jason has already pointed out for Q8, I think Q8 and Q7 should be npn's with e-c reversed.
Similarly C3 appears to have the wrong polarity shown.
I do wonder if the big difference between the degeneration resistors R7, R12 (1K, 100 ohms respectively) is a typo or an intentional way to try and set the vas idle current?
Thanks
-Antonio
As Jason has already pointed out for Q8, I think Q8 and Q7 should be npn's with e-c reversed.
Similarly C3 appears to have the wrong polarity shown.
I do wonder if the big difference between the degeneration resistors R7, R12 (1K, 100 ohms respectively) is a typo or an intentional way to try and set the vas idle current?
Thanks
-Antonio
Hi Antonio
All the errors look like the schematic has been derived from a sample so I would expect someone misread a resistor multiplier band. Makes no sense as an intentional difference.
Best wishes
David
BTW Nico, it is "complementary" and I rather like them, hence my list of benefits. But best to inform the OP of the other opinion too.
Last edited:
Hi all..
My oh my . i could see you all getting in a tiz about the scem being wrong . nico i agree that the amplifier will not work . I have also noticed the componet location numbers are totally diffrent to the ones used in the actual amplifier . Apart from that it is the right scematic . will see if i can redraw the scem in eagle using the amp i have to check the componet valuse .. Ps this amp sounds quite nice !
My oh my . i could see you all getting in a tiz about the scem being wrong . nico i agree that the amplifier will not work . I have also noticed the componet location numbers are totally diffrent to the ones used in the actual amplifier . Apart from that it is the right scematic . will see if i can redraw the scem in eagle using the amp i have to check the componet valuse .. Ps this amp sounds quite nice !
Bob is not a real fan but does actually write that "many fine amps have been made" with complementary symmetry, so the comment doesn't have much bite as a joke. You haven't read the book I think? But no problem, I've had a little discussion with him about symmetry in circuits and he's not doctrinaire about it. So I will take this as a chance to recommend his book to the OP and anyone else who hasn't read it, to learn some of the subtleties in the subject.
Best wishes
David
Last edited:
You appear to have forgotten to use the humour / sarcasm font.
For some reason when I quick reply the smilies do not attach, only when I do advanced. Sorry
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.