After some related telephone talk I want to add the source for the differential amp drawing of post
11210, it is Redirect Notice ,
seemingly Seoul National University, lecture chapter about diff amps, 90 pages.
In the context given I just had to find a simple diagram with picture search.
There has been some discussion here about a noise penalty in differential amps due to gain difference
("noise differential" has 39 results in this thread).
However if you use current mirror load on a long tailed pair full gain is on the single ended output.
11210, it is Redirect Notice ,
seemingly Seoul National University, lecture chapter about diff amps, 90 pages.
In the context given I just had to find a simple diagram with picture search.
There has been some discussion here about a noise penalty in differential amps due to gain difference
("noise differential" has 39 results in this thread).
However if you use current mirror load on a long tailed pair full gain is on the single ended output.
Perhaps the sources in 11227 are referenced to the supplies and not to ground.
IF3602
Sometime ago I bought some IF3602. They are very exotic parts. J-Fets wit a whopping 750mS of trans conductance.
For a semiconductor they are quite expensive but nearly not so much then people pay for cartridges and cables.
Last time I checked at Mouser they where 68,88€ a pop and not in stock with a lead time of 27 weeks.
Sometime ago I bought some IF3602. They are very exotic parts. J-Fets wit a whopping 750mS of trans conductance.
For a semiconductor they are quite expensive but nearly not so much then people pay for cartridges and cables.
Last time I checked at Mouser they where 68,88€ a pop and not in stock with a lead time of 27 weeks.
Attachments
As a consequence of the high trans conductance they have high Ciss too with 300pF.
I assume that they are made internally from a lot of J-Fets in parallel.
Other things to watch are the high Idss between 30mA to 300mA.
When we run this Fet at Idss and 10V V DS and when we assume that our sample has an Idss of 200mA we burn 2W of power. The IF3602 can only stand 300mW. My first experiments went in that direction. The cans got really hot but survived.
I assume that they are made internally from a lot of J-Fets in parallel.
Other things to watch are the high Idss between 30mA to 300mA.
When we run this Fet at Idss and 10V V DS and when we assume that our sample has an Idss of 200mA we burn 2W of power. The IF3602 can only stand 300mW. My first experiments went in that direction. The cans got really hot but survived.
So one possibility is inserting a negative voltage into the gate. An input cap is needed with this arrangement so one of the charms of J-Fets , the direct coupling , is lost.
Interesting idea to derive a negative bias from single supply is here :
My version of the G = 1000 low noise measurement amp (for Ikoflexer).
My version of the G = 1000 low noise measurement amp (for Ikoflexer).
Then it occurred to me that the classic differential amplifier is a good way to go. The IF3602 comes as a matched pair in the can anyway. When I was searching for a circuit I found an application by Bob Cordell. Me getting lazy I used this circuit with some modifications to turn it into an MC phono instead of MM.
Attachments
