Constant impedance relay-resistor logarithmic attenuator

[promitheus]

If you don´t understand the german Text you can just use my equations in the previous postings.

[OliverD]

Here you go:

[Brian Guralnick]

Originally posted by promitheus
If you don´t understand the german Text you can just use my equations in the previous postings.

Already read yours. I just enjoy gathering many different means of describing the exact same formula.

Thank you, both, AMT-freak & promitheus.

[Brian Guralnick]

Since it would cost less to trim our own resistors for this log pot, NE1 Know of the cheapest ohm meter which would allow me to trim a 5.xxxx Ohm resistor to better than 0.05%?

Perhaps, a milliohm meter, except then I stilll need a .05% meter for the above 100.00 KOhm range resistors.

[Steven]

Hi Brian,

There is no need to use such accurate resistors. Normal 1% resistors will do. Suppose an MSB of 32dB, then 1% resistors make an error of approx 0.2dB maximum, which is probably less than your LSB, that will be 1dB or maybe 0.5dB. So 1% resistors keep the volume control still nicely monotonic.

Steven

[Brian Guralnick]

You missed the 2 points, If I can trim resistors, why not do it right the first time.

OK, .5% is good enough for me, but, my current crummy meter reads 5.0 Ohm on a 5 ohm resistor, also, it's trim drifts. I need a meter which is at least 10x better for single digit ohms, without drifting due to local humidity, local temperature & it's lousy calibration pot. If I'm going to buy a new meter, make it a good 1.

[ErikdeBest]

Is it possible to use the same microntroller code as Till developed at http://home.arcor.de/dddddd-/relais/relais.html ?

Both are 8 bits and switch in a binary mode.

Remember that Till is selling the programmed microcontrollers.

[Brian Guralnick]

Originally posted by ErikdeBest
Is it possible to use the same microntroller code as Till developed at http://home.arcor.de/dddddd-/relais/relais.html ?

Both are 8 bits and switch in a binary mode.

Remember that Till is selling the programmed microcontrollers.

Who's design?

Mine uses the PIC16F72, it has a few more IOs.

[Magura]

Originally posted by Brian Guralnick
I need a meter which is at least 10x better for single digit ohms, without drifting due to local humidity, local temperature & it's lousy calibration pot. If I'm going to buy a new meter, make it a good 1.

My father surprised me once by dragging an old resistor bridge out of the desktop drawer. I believe the instrument is more than 30 years old....and its spot on with better presicion than ive been able to obtain from any other meter. Unfortunately a resistor bridge seems to be hard to find for reasonable money these days, but if you should stumble on such at a flea market.....dont hessitate to buy it

Cheers


Magura

[hifiZen]

Oops, I wish I'd seen AMT-freak's post earlier... I derived all the formulas by hand for this... Oh well. Just a word of caution: I believe the formulas on Steven's diagrams are incorrect. Please double-check before you use them.

Here is my version of the spreadsheet. It includes some diagrams from this thread, along with some useful notes. The worksheet also calculates some min and max values based on resistor tolerances used. Enjoy!