Unity gain solid state buffer

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi!

In these relatively quiet days around xmas. I've conceived the unity gain buffer depicted here below.

OltreSP-Buffer-DEF.png


Questions is: have you yet seen some similar circuit in the past? (also not in the audio electronic field). If the answer is "yes" can you tell me vhere you seen it and from what author?

Thanks in advance
Piercarlo
 
For who want to try out the schematic proposed above here below will find a track layout for build it on a small PCB, or better again, on a breadboard. BF245A (or BF245B with an Idss below 3 mA) may be replaced with any similar N channel device with an Idss comprised between 1 to 3 mA, The JFET is used in layout with swapped source and drain (operation possible with the most of general purpose JFETs) just because more straightforward to accomodate.

OltreSP-Buffer-PCBTRACK.png


Happy New Year! :).

Piercarlo
 
Last edited:
Disabled Account
Joined 2006
Whats is good or special or different with regard to the performance to warrant so much complexity for just a buffer ?? It must have THD 120db below on heavy loads , extreme speed or PSRR ??.... when we can come close to that performance by using just 4 transitors and less than 1/3 the passives. Coupling caps are not well regarded in audio world either.
 
Whats is good or special or different with regard to the performance to warrant so much complexity for just a buffer ?? It must have THD 120db below on heavy loads , extreme speed or PSRR ??.... when we can come close to that performance by using just 4 transitors and less than 1/3 the passives. Coupling caps are not well regarded in audio world either.

Who's care of these "questions"? If I wish i can replace all just with a single op-amp connected as voltage follower... If I don't try so is because i wan't try something of different. I CAN do it and I DO it! :)

Piercarlo
 
Before getting too involved with PCB design etc., have you made a lash-up prototype of this to make sure it all works as expected? I only mention it because spice simulations normally assume that all transistors are perfectly matched clones of one another and are closely thermally linked. In fact that all components are 'perfect'. Unless you deliberately put in variations that are more like real world scenarios. (?)

Just a thought....
 
Before getting too involved with PCB design etc., have you made a lash-up prototype of this to make sure it all works as expected? I only mention it because spice simulations normally assume that all transistors are perfectly matched clones of one another and are closely thermally linked. In fact that all components are 'perfect'. Unless you deliberately put in variations that are more like real world scenarios. (?)

Just a thought....

I have to mount it on a breadboard of course. Obviously I DON'T assume transistor as be perfect! ;-). Some variations during simulations however has been done just for wiping out some bad surprises (that I've presented is, for now, the fifth arrangement of a more sketchy circuit).

Thanks for the thought! I appreciate it! :)

Piercarlo
 
Disabled Account
Joined 2006
Who's care of these "questions"? If I wish i can replace all just with a single op-amp connected as voltage follower... If I don't try so is because i wan't try something of different. I CAN do it and I DO it! :)

Piercarlo

Many, much more than you think, just judge by the amount of members that are actually getting involved in the thread. ;) What does the circuit bring thats new or does better or worse than others ??
 
Your circuit looks similar to Maxim's BB3553 or LH0063 (used to be Burr Brown before).
www.datasheetcatalog.org/datasheets/270/493512_DS.pdf
These are not monolithic IC but thickfilm hybrid inside.

Ok, this is what I want to know. Too beautiful believe my circuit as a really new one! :). The inner schematic of BB3553-LH0063 is conceptually similar to those from I started a couple of week ago. There are however a couple of improvements...

Homemodder, see you where are these improvements and what are the improvements obtained? ;-) Not the distortion, not the bandwidth, non only the SVR... (why?) ;-)

Piercarlo
 
What are the improvements obtained? ;-) Not the distortion, not the bandwidth, non only the SVR... (why?) ;-)

The major real improvement is input acceptance, which is limited only by voltage supply. Once provided that Q2 and the bipolars used in current generators - QGN2, 3, 4 - have enough VCE, the Vs may any you need for your required signal acceptance, even 300 volt if you will... Of course you have to choose even the right working voltage of capacitors but this is obvious. The JFET may be any with IDSS in the range of 2-3 mA. Intrinsic output impedance (before output capacitor) should be below 10 ohm.

The improvement of SVR (and linearity) is simply a biproduct of that featured above.

Piercarlo
 
Last edited:
Hi!

Circuit revision due to a serious design fault - a spurius SCR between a couple of transistor non revealed during simulation but (of course) readily "revealed" during breadbord testing. Here below the new schematic:

OltreSP-Buffe-Revis3.png


The new file for simulation on LTSPICE is downloadable at the link below.

https://www.dropbox.com/s/14iyee1jzmybfmz/OltreSP-Buffe-Revis3.asc

The new circuit is physically again under construction.

Piercarlo
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.