Variable Headphone Attenuator Pot - Source?

[cedd]

Hi All

I'm trying to do what sounds to be something pretty simple - build a volume control for my desk headphones. I'm currently tweaking my flightcase (see other thread) and one of the negatives from the change is that my volume control (under the wrist rest at the front of the desk) ends up being hidden. I can get a connector in to the socket but won't be able to adjust the level. 

This isn't the end of the world as I'm building an interface panel that lets me switch between a coms circuit and my desk, so it seems straightforward to stick an attenuator pot on there, turn the desk volume up full before I stick it in the case, and just passively attenuate. As far as I can see I just need to use a dual gang logarithmic pot as a potential divider. 

Here comes the problem, I'm struggling to find such a beast that looks like it'll handle the power. There's one on CPC that'll handle 200mW, but that just feels a little too low. My normal headphones are either Sennheiser HMD-46's which are 300R, but could equally be a pair of DT-250's which are 80R. Am I worrying over nothing, or do I need to find something more capable? If so then what??? I've tried a few different search terms but can't seem to find anything suitable. 

I know Behringer have their PM1, which seems to be exactly what I want in a box, but what's inside one of them? £30 seems a lot if all I need is a few quid's worth of potentiometer. 

[timsabre]

The thing with Behringer is they manufacture enough quantity to make custom pots viable, so they could easily have had a high power low resistance pot made. Or indeed a weird resistance taper or something.

I would have thought 200mW would be plenty, it's only going to be 200mW very intermittently unless you're blasting white noise through them all the time. Watch out for how you connect headphone ground as sometimes headphone outputs play tricks with the ground to get a larger signal, so headphone ground isn't audio ground (probably not on a desk output though).

[David Duffy]

Agreed on the switching part of the equation. I would switch all 3 legs (L/R/G) of the headphone feed, not just the L/R.

As for the dual gang pot, what resistance values can you get?

[cedd]

Yeah I'm switching all 3 legs - my comms beltpack is a Bluecom one, so it's a H-Bridge output. 

Fair point about it being intermittent, given it's audio. That 200mW is likely to be DC isn't it. 

Looking at a 10k log pot at the minute. 

[alistermorton]

That's going to give you an interesting curve, if you're connecting to 80 ohm phones, whether you just put it in series or wire it as a divider like in an amp, the relative resistances are going to make for a gain control that will be very sensitive. 1k in series would give you approx 10:1 level change end to end, unless I've done my mental arithmetic wrongly. 

[kgallen]

Yea I was going to suggest more like 1k too.

For this sort of thing I often look in the service manual for a Sony minidisc deck. They always have the schematics and are useful for circuit ideas! Many minidisc decks (unlike CD players) have headphone outs with volume control.

Look at MDS-E10 or E12 or 9xx machines. Don’t bother with late entry-level 4xx machines (eg MDS-JE440) as they don’t have headphone outs. 

eg https://www.minidisc.wiki/_media/equipment/sony/pro/sony_mds-e10_sm.pdf

(pdf page 69, doc page 78, bottom right)

Here: https://www.minidisc.org/part_Decks_Sony.html

Or here: https://www.minidisc.wiki/equipment/sony/deck/start

Edited March 25, 2022 by kgallen

[cedd]

10k was the lowest resistance dual gang pot I could get through CPC, who I've realised through this process are definitely no longer an electronics components company! RS have a fair few more, however the only dual gang 1k log pot they've got is 24 quid! Can get a linear one for 7, but I'd really prefer log. 
At this rate I may as well buy a PM1.  

[DrV]

You could use a wafer switch and fixed resistors. Very audiophile to have a switched attenuator!

Actually forget wafer - I just looked at the prices, but a standard 12way single pole Lorlin switch is only £2.63 plus 11 resistors won't break the bank. And you get to design your own curve.

Just remembered you want stereo so you're back to wafer switches, unless you can manage with 6 positions

Edited March 25, 2022 by DrV
Didn't read the question

[alistermorton]

That's a good point - a 2 pole 6 way switch might work rather well. 

[boswell]

Will this do?

https://tinyurl.com/25hsjp5y

 

 

[Sound Man]

I have used a dual gang 1K pot in this application before. I've only tried it with 8ohms headphones and it worked over the full resolution of the pot.

David

[cedd]

11 hours ago, boswell said:

Will this do?

https://tinyurl.com/25hsjp5y

 

 

I'd seen those, but 100mW seemed a little on the low side. If folks seem to think it'll be ok then it's a decent option.

[kgallen]

12 minutes ago, cedd said:

I'd seen those, but 100mW seemed a little on the low side. If folks seem to think it'll be ok then it's a decent option.

The Sony MD machines I referred to above are using an Alps RK09L series and just looking at the data sheet now they are 50mW pots. 

[David Duffy]

I'd be trying for a 500 Ohm pot, but that will be harder to find in dual gang.

Don't worry too much about the log taper as you may find a linear tape works perfectly fine in this application.

[cedd]

Oh he has done it! He has found the world's classiest potentiometer! (Fans of the musical Our House will get the reference!).

Turns out if you want to find a dual gang potentiometer on CPC, you need to make sure you don't filter for dual gang! 

https://cpc.farnell.com/vishay/14920f0gjsx13102ka/potentiometer-1w-dual-1k/dp/RE04604

Aside from being linear I reckon it's just what I need. Big more than I wanted to pay, but not heaps. 

Have found a 5k as well, but the consensus seems to be that 1k is best. 

Thanks all for your input!