DIY stepped pots w/rotary switches & resistors

[djmiggymigz]

Cool thank you! I looked thru the datasheet as well and think these may be the ones. I don't understand how the ones I linked are around $13/switch and then the ones you linked are up to around $20/switch! Super unfortunate seeing as I need 12 switches.

I'm also no longer sure whether I need shorting or non-shorting lol. As I said, Harpo made an awesome excel sheet way back to calculate what the resistance values of each step should be. On that sheet, harpo specifically says that the switches should be non-shorting (bbm) switches. Seeing as there was so much thought and work put into this I'm inclined to follow that guidance.

 

[chrispbass]

Cool thank you! I looked thru the datasheet as well and think these may be the ones. I don't understand how the ones I linked are around $13/switch and then the ones you linked are up to around $20/switch! Super unfortunate seeing as I need 12 switches.

I'm also no longer sure whether I need shorting or non-shorting lol. As I said, Harpo made an awesome excel sheet way back to calculate what the resistance values of each step should be. On that sheet, harpo specifically says that the switches should be non-shorting (bbm) switches. Seeing as there was so much thought and work put into this I'm inclined to follow that guidance.

You want shorting…make before brake …MBB. That way, there is no disconnection when changing values

 

[djmiggymigz]

hmm yeah that does make sense to me as well. Only thing about this design is harpo's calculations take into consideration a resistor wired in parellel outside of the switch. Could this be the reason he recommended BBM? Here's a pic of the excel sheet for reference.

 

[ruffrecords]

From a circuit design point of view, there are two main factors which affect whether you use shorting or non-shorting contacts.

1. Gain changes as you switch. Does the gain go up or down? Generally speaking gain going down is preferable. In most circuits this indicates using non-shorting contacts.
2. Can you maintain a dc path at all times to avoid clicks? This can be achieved either by shorting contacts or a parallel resistor.

Cheers

Ian

 

[user 138639]

Cool thank you! I looked thru the datasheet as well and think these may be the ones. I don't understand how the ones I linked are around $13/switch and then the ones you linked are up to around $20/switch! Super unfortunate seeing as I need 12 switches.

I'm also no longer sure whether I need shorting or non-shorting lol. As I said, Harpo made an awesome excel sheet way back to calculate what the resistance values of each step should be. On that sheet, harpo specifically says that the switches should be non-shorting (bbm) switches. Seeing as there was so much thought and work put into this I'm inclined to follow that guidance.

Not sure if you would get them for around $13.-, I guess it’s geographic and the supplier. And yes, I would follow the masters advice too, absolutely, as they’re supposed to know what they’re doing, while I…

 

[craigmorris74]

I’m thinking of replacing the pots with rotary switches on my Pultec style EQ. Would a 12 position switch be enough?

 

[user 138639]

12 steps might be a little coarse, but if the bandwidth you work on in narrow (the adjustments are fine/small) it may work. I unfortunately don’t know pultecs but from the internet, the case look a bit narrow… you’re sure the pultec has the space for stepped attenuators? (They are rather bulky compared to a pot…)

 

[craigmorris74]

12 steps might be a little coarse, but if the bandwidth you work on in narrow (the adjustments are fine/small) it may work. I unfortunately don’t know pultecs but from the internet, the case look a bit narrow… you’re sure the pultec has the space for stepped attenuators? (They are rather bulky compared to a pot…)

I was afraid of that. The commercial versions have 21 steps. I’ll take measurements before ordering anything.

 

[user 138639]

If space, budget, and specs allow for it, you could use some prefab attenuators, made with smd-resistors.

Something like these (I’m not affiliated)

 

[djmiggymigz]

From a circuit design point of view, there are two main factors which affect whether you use shorting or non-shorting contacts.

1. Gain changes as you switch. Does the gain go up or down? Generally speaking gain going down is preferable. In most circuits this indicates using non-shorting contacts.
2. Can you maintain a dc path at all times to avoid clicks? This can be achieved either by shorting contacts or a parallel resistor.

Cheers

Ian

Thanks for the reply Ian, really appreciate you breaking by that down. That makes sense. I’ll be following harpo’s guidance and go with non-shorting.

Now, a potential total noob problem/question. So I bought the Lorain CK1458 1 Pol x 12 Pos switches. Just got them in today and immediately following my excitement: confusion. Hahah

They’re supposed to be 12 position, they have 12 solder terminals and 1 pole in the back as they should. However, there’s only 11 clicks when rotating from one side to the other?? Am I somehow confused about the switches work?

Attaching a screenshot of the specs from Mouser

 

[ruffrecords]

Of course there are only eleven clicks. You are initially at position one. You rotate clockwise one click and get to position 2, then a second click gets you to position 3 and so on until the 11th click gets you to position 12.

Cheers

Ian

 

[djmiggymigz]

Of course there are only eleven clicks. You are initially at position one. You rotate clockwise one click and get to position 2, then a second click gets you to position 3 and so on until the 11th click gets you to position 12.

Cheers

Ian

Okay I knew something was weird! Just double checked them and I actually mis-spoke: they actually only have 10 clicks?? I tripl checked and sure enough they only have 10 clicks from the bottom to the top position.?

 

[TwentyTrees]

In that case, they may have the special washer set for only 11 stops. Try removing the nut, the lock washer, and the flat washer with a single perpendicular tooth that's underneath - if it's there. That last can be used to set the maximum rotation for many Lorlin CK switches. If it's in place, maximum rotation will be 11 stops; remove it, and you should get 12.

 

[djmiggymigz]

Ahhh okay thats probably it! I did notice that there’s a washer with a tooth. I’ll give that a shot! Thank you!

 

[djmiggymigz]

In that case, they may have the special washer set for only 11 stops. Try removing the nut, the lock washer, and the flat washer with a single perpendicular tooth that's underneath - if it's there. That last can be used to set the maximum rotation for many Lorlin CK switches. If it's in place, maximum rotation will be 11 stops; remove it, and you should get 12.

BINGO Thank you!! Lol that was exactly what was going on. Appreciate all the patience here! Gonna add resistors to one of the switches now as a test run!

 

[TwentyTrees]

Great stuff, glad to hear it.

 

[djmiggymigz]

I'm baaaaaack!!! Looking for some help working through a problem here! So I finally got around to finishing up all my switches for the MAAG EQ clone. Got the resistors on all the steps as well as the resistor is parallel with the switch. I'll include the calculations that I used for the R values using Harpo's excel sheet.

Now the problem I'm having is in circuit, the switches are boosting beautifully but there is no cut happening!! It seems to be boosting correctly at around 3 dB however the cut (at step one which should be max cut) only seems to attenuating like 0.3 dB if even that. I double checked that all the resistor values were correct and followed these calculations and still nothing.

Im thinking it may be how I wired the switches into the pcb. I attached a picture of how the pots are supposed to be wired up. Based on this I currently have the switches wired as follows: Wiper and Pin 1 are wired as the red line in the diagram and then I have the loose end of the resistor on pin 12 jumped with a wire to where the blue line goes. Does this sound correct based on how the pots should be wired?? I can post a more detailed drawing later on that might make it easier to understand what I have going on right now.

 

[djmiggymigz]

Actually here is what I think I've done. I think I need to upload an actual picture of the switches later though because I'm thinking I just have the input/outputs wired wrong.

 

[DaveDC]

From a circuit design point of view, there are two main factors which affect whether you use shorting or non-shorting contacts.

1. Gain changes as you switch. Does the gain go up or down? Generally speaking gain going down is preferable. In most circuits this indicates using non-shorting contacts.
2. Can you maintain a dc path at all times to avoid clicks? This can be achieved either by shorting contacts or a parallel resistor.

Cheers

Ian

This is all true, however a set of switched resistors like is being described here does not behave identically to a potentiometer, strictly speaking. In the case of a shorting switch (make before break), when the switch is between detents, one resistor of the chain is shorted. That makes the end-to-end resistance (the resistance between the two terminals that are not the "wiper") is slightly different than when the switch is in a detent. Usually this doesn't matter, but it is not what a real potentiometer does. In the case of a non-shorting switch (break before make), the wiper is an open circuit when the switch is not in a detent. This also not what a real potentiometer does. So you need to take these factors into consideration when replacing pots with stepped resistors. I'm not saying not to do it (I am a fan of stepped controls), just make sure your circuit is okay with it. Pay particular attention to your circuit's DC situation (bias current, offset, etc.) which is often overlooked in audio devices. A sudden step in the DC operating point can make for a nasty click in the audio.

Another thing to consider is that while pots are indeed evil and unreliable and prone to getting noisy, switches are no panacea. Like any mechanical control, switches can also become unreliable with dirt and corrosion. Sealed switches are what you want here, the kind you'd find in aeronautic applications (Janco, etc). But they are bloody expensive!