Yaesu Incorrect Op-Amp Termination

[w3amd]

Hi all,

Is this an incorrect implementation of unused op-amp termination in Yaesu radios? I'm only a novice. I think this is probably incorrect but want to ask the experts.



Cheers,
John

[ciccio]

I've seen this done in many devices. The first time I've seen it (I was repairing it) it I've checked the circuit for instability, and could not find none.
The book you are referring to says NO to the the solution I always use, that is output shorted to inverting pin and non-inverting pin grounded.
I've done this from 1970's, without problems, NEVER any problem.
I generally use dual supplies, and when I have a single supply the non-inverting input is wired to virtual ground (1/2 supply voltage) and not to ground.
My experience is with audio amplifiers and DC amplifier used in industrial controls. Other application may have different requirements, but I doubt.
Maybe the book refer to single supply amplifiers.... or the book assertions came from other books, and not from direct experience.

Best regards

[T3sl4co1l]

Depends on type.  Grounding inputs is very common.  Or VEE, or VCC or something inbetween.  Or a skewed state (one input above the other, so the output is fixed high or low), or a follower connection (+in = VREF, -in = out) so it can do its job as usual (i.e., amplifying with feedback, you're just not using the signal).

Incorrect setting can cause problems, because multi-part amps often share bias circuitry.  Causing one section's circuitry to saturate (inputs exceeding common mode range, output saturating?) can cause problems in the others.  This is usually not a problem on modern parts, but RTFDS to be sure.

Tim

[w3amd]

Depends on type.  Grounding inputs is very common.  Or VEE, or VCC or something inbetween.  Or a skewed state (one input above the other, so the output is fixed high or low), or a follower connection (+in = VREF, -in = out) so it can do its job as usual (i.e., amplifying with feedback, you're just not using the signal).

Incorrect setting can cause problems, because multi-part amps often share bias circuitry.  Causing one section's circuitry to saturate (inputs exceeding common mode range, output saturating?) can cause problems in the others.  This is usually not a problem on modern parts, but RTFDS to be sure.

Tim

Thanks Tim. I'm going to try to take a look at the chip they used in both radios and see what the common mode specs are.

Here is another video from Texas Instruments itself:


The datasheet for the op-amp in the FT-450D radio is an LM2904:
http://www.ti.com/lit/ds/symlink/lm258a.pdf

Cheers,
John

[SeanB]

Page12 shows a schematic, showing the input stage is a PNP transistor,. so it is fine being grounded. If the input common mode does not include ground you need a simple voltage somewhere inside the common mode range to connect to the one input and simply use the opamp as a buffer if it is stable as a unity gain buffer, otherwise you will have to use a minimum gain configuration to keep it stable.

[chris_leyson]

Well spotted John, grounding both inputs of an unused op-amp was a popular method back in the day with bipolar LM358 type op-amps. I've seen this in a lot schematics. The correct method all depends on the op-amp. The safest method is to wire it as a unity gain buffer and park the bias voltage within the amplifiers common mode range, of course that assumes that the op-amp is unity gain stable.

A lot of early generation dual or quad op-amps shared a common bias circuit and you could pull the bias by driving one of the op-amps outside of its common mode range and that would pull all the other op-amps. I've fallen into this trap trying to use a dual op-amp as an amplifier and comparator, the comparator modulated the op-amp bias so the op-amp no longer worked as intended.

Anyway, it's nice that a software engineer ask a hardware question.