Electronics > Projects, Designs, and Technical Stuff
Find highest value of 4 inputs without microcontroller?
<< < (6/12) > >>
T3sl4co1l:

--- Quote from: spec on November 18, 2018, 04:53:22 pm ---Ah I see your clever scheme- use transistors as diodes and difference amplifiers.

The only thing is that  the A#15 circuit uses:  2 x LM234 , 4 x diodes, 3 x resistor

While the A#23 circuit uses: 1 xLM324, 9 transistors, 11 resistors

--- End quote ---

Yeah, like I said, there's the other way, which is detecting which amps are saturated.  (Did you even read my reply or just glaze over at it? :P )

And like I said, the PNPs can be "prebiased" types, for 16 components total.


--- Quote ---Don't you use a circuit capture package?

If not, you do know that there are some pretty good free versions available on the web.

I'm an EAGLE fanboy ::)

--- End quote ---

I use Altium.  No point in cranking it open just for a doodle. ;)

Tim
unitedatoms:
@Teslacoil:

That schematic looks like the best. Will work for negative voltages too. It is possibly fine to have opamps feedbacks with total n2 pathes = 16 feedbacks. I was not sure about AC stability. But may be it is fine and  stable
spec:

--- Quote from: T3sl4co1l on November 18, 2018, 05:11:49 pm ---(Did you even read my reply or just glaze over at it? :P )
--- End quote ---

The latter- far too long   >:D


--- Quote from: T3sl4co1l on November 18, 2018, 05:11:49 pm ---I use Altium.  No point in cranking it open just for a doodle. ;)
--- End quote ---

You should have. :)
spec:

--- Quote from: T3sl4co1l on November 18, 2018, 12:42:17 am ---Should be able to do it in slightly fewer, due to this being a sorting problem (of sorts) and sorting problems being guaranteed N log N worst case complexity.
--- End quote ---

But you didn't :)

That's why you should post a circuit rather than just say what you should be able to do.

Your circuit does nothing more than my circuit and uses more components. It is also much more complex.

It also  does not operate down to the 0V rail, to what extent is not clear because you haven't put in values and it does not have an adjustable initial offset (50mV in my circuit).

SiliconWizard:
I think spec's approach is the simplest so far if using discrete components, and if using RR opamps and schottky diodes, the operating voltage range is pretty decent. The way the feedback loop is designed allows to get usable input voltages down to almost 0V and also very small detectable voltage differences, which wouldn't be the case with just diodes?
As is, it won't work properly for negative input voltages though, even if the opamps are powered by a dual supply. At least one of the input voltages would have to be positive in order for it to work I think, otherwise all outputs would be down to the negative rail.

Of course if it were to be implemented in silicon, using opamps in this way would be wasteful and full-fledged opamps would probably not be required. Using a few transistors instead would yield a much smaller area, although I'm still wondering about the effective usable voltage range for the 4 inputs.

I've also thought about a clocked version of this, using switches, only one comparator and a peak detection block. Would be effective area-wise if dealing with a lot more inputs too, as it would scale up well. Drawback would be the latency to get the end comparison, but that may be negligible depending on the application. Implementing this with discrete parts would take significantly more parts too.
Navigation
Message Index
Next page
Previous page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod