I have made a transistor based push pull preamplifier and have written a document.
In the document, two approaches have been discussed : the made one, with standard common emitter capacitors and the better one, the one without standard common emitter capacitors.
The circuit without emitter capacitors provides gain and buffer like linearity. The price to pay is a gain dependent maximum output ( and input ) amplitude.
Yet, because of the advantages, I prefer the circuit without emitter capacitors, which is a standalone transistor circuit with a local feedback and without a general feedback from the very output to the very input. The local feedback circuit, however, provides performance similar to the common collector buffer.
Of course, because of a lack of differential amplifier, at high gains, the power supply same phase noise is also amplified, but, by a lower gain than the signal gain. At low gains, this noise is attenuated, but, not by much.
Of course, with a provision for high inputs, such as 4.35V, the circuit does not reach the rails and, at high gains, the maximum output amplitude can reach 10V with +- 15V power supply.
Again, please, note, this output can be achieved at high gains. At lower gains, the output is lower, but, the dependence is not linear, thus, only gains - 1 through -3 provide rather low outputs ( 5V through 6.67V ).
I do NOT have linearity analysers and oscilloscopes. I only have a simple, inexpensive multimeters and ears. By ear, the sound is fantastic and the noise is OK. With a battery power supply, the noise is not bad even at the maximum gain of around 400!
I have not installed, but, provided room on the PCB to put large electrolytic audio power supply capacitors. These would reduce the power supply cable noise significantly as with other projects.
The document became huge. There is no way not to have made any mistakes. An attempt has been made to turn all stones around, but, of course, I may have forgotten some. An attempt has been made to provide only logical explanations and not useless terms. Of course, some things may have not been explained well. ( Well means by logic and simple language only. Logic is science. Logic is not sense. )
Thus, please, report whatever you have found in this sense.
Here is the link to the document ( the schematics and pictures are in the addendum ) :
Transistor Based Push Pull Preamplifier - Google Drive
Also report in case of any problems with the link and the document download and ability to view.
In the document, two approaches have been discussed : the made one, with standard common emitter capacitors and the better one, the one without standard common emitter capacitors.
The circuit without emitter capacitors provides gain and buffer like linearity. The price to pay is a gain dependent maximum output ( and input ) amplitude.
Yet, because of the advantages, I prefer the circuit without emitter capacitors, which is a standalone transistor circuit with a local feedback and without a general feedback from the very output to the very input. The local feedback circuit, however, provides performance similar to the common collector buffer.
Of course, because of a lack of differential amplifier, at high gains, the power supply same phase noise is also amplified, but, by a lower gain than the signal gain. At low gains, this noise is attenuated, but, not by much.
Of course, with a provision for high inputs, such as 4.35V, the circuit does not reach the rails and, at high gains, the maximum output amplitude can reach 10V with +- 15V power supply.
Again, please, note, this output can be achieved at high gains. At lower gains, the output is lower, but, the dependence is not linear, thus, only gains - 1 through -3 provide rather low outputs ( 5V through 6.67V ).
I do NOT have linearity analysers and oscilloscopes. I only have a simple, inexpensive multimeters and ears. By ear, the sound is fantastic and the noise is OK. With a battery power supply, the noise is not bad even at the maximum gain of around 400!
I have not installed, but, provided room on the PCB to put large electrolytic audio power supply capacitors. These would reduce the power supply cable noise significantly as with other projects.
The document became huge. There is no way not to have made any mistakes. An attempt has been made to turn all stones around, but, of course, I may have forgotten some. An attempt has been made to provide only logical explanations and not useless terms. Of course, some things may have not been explained well. ( Well means by logic and simple language only. Logic is science. Logic is not sense. )
Thus, please, report whatever you have found in this sense.
Here is the link to the document ( the schematics and pictures are in the addendum ) :
Transistor Based Push Pull Preamplifier - Google Drive
Also report in case of any problems with the link and the document download and ability to view.
Last edited:
Sorry Steven I fell at the first hurdle --can,t access Google Drive as I have zero Google on my minimal LInux PC and block its tracking.
But has not this been discussed before many decades ago in relation to tube cathode bias capacitors ?
I have read about it many times over the decades in relation to tubes and all the discussions on various technical magazines but if you are applying new engineering principles could you elucidate at least a glimpse of your findings.
But has not this been discussed before many decades ago in relation to tube cathode bias capacitors ?
I have read about it many times over the decades in relation to tubes and all the discussions on various technical magazines but if you are applying new engineering principles could you elucidate at least a glimpse of your findings.
Last edited:
An input source follower is followed by two emitter followers and the identical output signals applied to two more emitter followers, this time complementary. There is quite a bit of tautology so far. Why so many followers? Some even providing identical signals. What is the purpose of T21/22?
Update : The file has been updated with a filter between the two output buffers of Schematics PPA 0.
I have received an email the document could be downloaded. I, also, tried to do this while signed out of Google. Worked OK.
There may be misunderstanding here. The summary at the beginning of the thread meant to say the standard emitter capacitor was possible to be omitted on one of the circuits.
A standard common emitter amplifier has an emitter capacitor which conducts AC. This brings a phase shift, no AC feedback and non linearity. The linearisation resistor ( emitter degeneration ) introduces some feedback, yet, far away from 100%, and slightly improves the linearity, yet, not as much. Also, transistors are very linear, so, the overall linearity is not bad.
However, on one of the circuits, shown on Schematics PPA 0. in the addendum ( which has not been physically made, but, has been analysed as well as the circuit is similar to the one made ), the circuit does not need emitter capacitor and works with AC and DC. The circuit has similar linearity to a standard, common collector buffer and can amplify.
The push pull amplifier is a dual common emitter amplifier. However, the idea is to avoid the emitter capacitor.
I unblocked Google Drive using UBO but it only gave me a blank box.
Poster on UBO website states same shows "credential_picker_container" and he was signed on in Google . -
Its a secure context container interface of the CM API and exposes methods to request credentials and NOTIFY the user agent when a sign in or out happens .
In plain language if it doesn't recognize you as a "normal " person(like being labelled a "Bot " ) you don't get in , I took out "Google Safety Checks " in --"about :config as it uses it to track you round the web .
Not your fault but mine --why should I make it easy for all the giants to track me ?
Amazon could even tell the location of my public server till I did some re-coding now it only know I live in the UK.
Poster on UBO website states same shows "credential_picker_container" and he was signed on in Google . -
Its a secure context container interface of the CM API and exposes methods to request credentials and NOTIFY the user agent when a sign in or out happens .
In plain language if it doesn't recognize you as a "normal " person(like being labelled a "Bot " ) you don't get in , I took out "Google Safety Checks " in --"about :config as it uses it to track you round the web .
Not your fault but mine --why should I make it easy for all the giants to track me ?
Amazon could even tell the location of my public server till I did some re-coding now it only know I live in the UK.
I like to disconnect chains with buffers, but, in this case, I think the buffers are good to have.
T21 and T22 :
1. The input JFET J201 buffer better not to provide current for higher linearity.
2. The two input offset bias circuits better be disconnected, so, they do not influence each other. T21 buffers the capacitor and the resistors of the upper input offset bias chain made by C31 and the voltage divider R311, R310 + RT310. T22 buffers the capacitor and the resistors of the upper input offset bias chain made by C32 and the voltage divider R321, R320 + RT320.
T41 and T42 buffers are necessary at high gains of T1 and T2 only where the overall emitter impedance is low. The input impedance of T1 and T2 is :
Rin = Ze * ( B + 1 )
When Bmin = 220 and Ze = 10 Ohms, the input impedance is 2.2K, which is lower than R310 + Rt310 and R320 + RT320.
Thus, T41 and T42 provide higher input impedance for the main circuit. Also, Darlingtons can be used for T1 and T2.
T41 and T42 must provide an input impedance 10 fold bigger than the input impedance of T1 and T2 or higher. Better be 100 fold or higher.
The circuit on Schematics PPA 0. in the addendum has a fixed emitter resistors of 1K and, therefore, T1 and T2 provide high input impedance. Therefore, T41 and T42 have not been used.
The PNP NPN combination is for symmetry and, probably, for higher linearity, although, some of the parameters ( mainly B, but, the buffers are, somewhat, B independent ) are not the same even in PNP NPN complimentary transistors.
In case you are interested, please, provide a way, such as email, messages, etcetera, where I can send you the file, which is around 8MB.
I prefer to have the file uploaded to Drive, so, I can change and update the file whenever necessary and to avoid any size limitations.
Thats kind of you Steven can you send it via PM on Contact me when you click on my name or will that not work ?
That wouldn't work... PM's won't allow attachments of any kind which is by design to help keep everyone safe.
The file would attach to the forum (I think) if the extension were changed to mp4 as these have a 9.4Mb limit.
The file would attach to the forum (I think) if the extension were changed to mp4 as these have a 9.4Mb limit.
I think, the best solution for you may be to download the file on your cell phone. Android would allow this and, I hope, Apple too.
To read the file on your phone is, theoretically, possible, but, not practically. However, you can upload the file from your cell phone to your Linux PC. I hope, Linux allow this either directly or when your cell phone is considered by Linux to be a Flash Drive.
The easiest way may be to go to this forum on your cell phone and, then, to the first post. Then, just click on the link. Should work. You may be asked a question whether you want to download the file, open with Google Drive doc viewer or with your phone doc viewer or something alike. I think, all these three or more options download the file to the folder where your downloads are.
Please, inform whether this has worked.
Update : more big capacitors have just been install on the power supply. Schematics PPA 2. has been updated in the document.
As the "document" gets itself updated by the author, whenever you download it, it will be out of date. You will have to download it everyday to stay updated. Rightly, the posts on DIYAudio are not allowed to be edited after a small delay.
That is generally true on the 'edits', however as I recall the thread starter does have the ability to edit the first post indefinitely.
Information on the updates are posted here. Downloads are good to do after each update as posted.
Updates provide a way for a better document.
I was able to do an edit to the first post. However, even when the edits are difficult, one can always make a new post, which is probably the better way. Whoever wants the old text, this can be copied, pasted and edited. Or, even better, one can reply to the original post and provide explanations.
In regards to the original document, a new document, called " Push Pull Clamp.doc " has been uploaded to the same folder : Transistor Based Push Pull Preamplifier - Google Drive
This document provides a circuit which, probably, does not have any practical application, but, is interesting.
This document provides a circuit which, probably, does not have any practical application, but, is interesting.