Do you have a reason for saying you cannot use PWM? That is what I would use. It is by far the best way to go.
If your fix is to add a processor of some kind to control the switching, it can generate the PWM signal.
Otherwise, a center off DPST switch can work. This will give you 100R, 50R and 10R, but no off position.

View attachment 202209

Thank you for the diagram. I used PWM (ATMEl + MOSFETs) in my previous device and I had problems with the oscillation at higher frequencies because I run pretty long wires from the control unit to LED matrix. Anyway the diode matrix is for illuminating plants and I'm not sure what effect pulsing if any may have on bio-modulation so I don’t want to use it'

It seems that the mechanical switch is my best bet. I’m considering using RS26 4P3T 4 Pole 3 Position Rotary Channel Selector .

 

Anything with chips is likely to require more than one with that 24 volts.

Even if you could get a micro to sink the current, you would still need a regulator.

Certainly. I meant one chip to control

Anything with chips is likely to require more than one with that 24 volts.

Even if you could get a micro to sink the current, you would still need a regulator.

I have no problems with using step down voltage regulator for the chip.

 

Well then post #11 becomes very confusing, because you were offered a single chip solution.

Same solution was also offered in post #2.

 

control

Well then post #11 becomes very confusing, because you were offered a single chip solution.

Same solution was also offered in post #2.

Step

option #2 would only need 1 chip, if you'd use a hex inverter IC like a cd4069ub and make 3 latches out of those 6 inverters.

For option #1 in post #11, the CD4017 is the only chip. Granted you could get the switching transistors in a quad package, but that's not a "chip" in the traditional sense.

ak

I did not realise MOSFETs come in quads. CD4017 + QUAD can work.

 

option #2 would only need 1 chip, if you'd use a hex inverter IC like a cd4069ub and make 3 latches out of those 6 inverters.

It looks interesting but I'm having hard time understanding how it works. It looks like it operates almost like diode type of device and does not have any input resistance.

 

I use 3 LED drivers to drive 3 strings of LEDs. LED driver current can be adjusted via an external resistor. I want to be able to set three levels of current which means that I need to switch among 3 resistors per LED driver.
I will use following resistors 10 ohm, 50 ohm and 100 ohm. Considering the value of resistors switching device should not add any additional resistance above 10 ohm.
Since I’ve got 3 drivers I need to switch among 9 resistors. I don’t want to use rotary switches. Are there multriplexers or SS switches that don’t add additional resistance ?

Hi Alex,

It would be possible with only 2 toggle switches per string.



The resistance value would be 100 Ω with both switches open, 50 Ω with the 50 Ω switch closed and 10 Ω with the 10 Ω switch closed.

Nandu

 

It looks interesting but I'm having hard time understanding how it works. It looks like it operates almost like diode type of device and does not have any input resistance.

Heres a 4 bit counter I designed, built and tested awhile ago, using 1 cd4069ub IC .Modified to suit your application, it switches from 10, 47, 92 and 129 ohm

 

Here is a first pass at post #11 option #1. Note that the LED and circuit power systems must have a common ground, and that the circuit needs a regulated voltage source at 15 V or below. You can eliminate one transistor by using the switching scheme in post #26.

ak

 

Here is a first pass at post #11 option #1.

Probably should add a small resistor in series with SW1 so the capacitor high peak current doesn't weld the contacts.

 

Probably should add a small resistor in series with SW1 so the capacitor high peak current doesn't weld the contacts.

Sorry Carl, I used the circuit below to reset PIC micros, with no problems.

 

The schematic is an edit of another circuit from about a year ago. Back then I added the 100 ohm resistor at Wally's urging. I took it out today to see if he still cares.

ak