Another passive monitor controller

Help Support GroupDIY:

bjosephs

Active member
Joined
May 10, 2021
Messages
42
Location
Massachusetts
Hi all,


I know there are plenty of threads on this already but we all take a slightly different path, so here’s another.

Attached is a scribble I did (it’s drawn in two parts but ignore the gap) and I’m looking for some double checking.

I defaulted to 2k2 series resistors and a 10k audio taper “pot”. I will probably buy a nice stepped attenuator. Is there a reason I couldn’t use a larger value (say, 25k)? Mono simply ties the L/R hots and colds together using the 2k2s to mix. I wanted a DIM function that didn’t require a 4P switch so I’m going to just slap a small value across the attenuator pot to drag the level down… the taper will be weird but this seems unimportant if you just want to bring the level down to talk or give your ears a rest. The rats nest thereafter is a L/R flip followed by a cut L or R function achieved by tying hot and cold together at the output.

I may do it all point to point on the backs of toggle switches mounted to a rack blank with some XLR pig tails hanging off the back.
 

Attachments

  • 9FD340A5-D224-47DD-95CD-490213C4EC61.jpeg
    9FD340A5-D224-47DD-95CD-490213C4EC61.jpeg
    152 KB · Views: 65

bjosephs

Active member
Joined
May 10, 2021
Messages
42
Location
Massachusetts
Using relays would afford me some more options on switch selection. This might be the shortest path to a prototype.
 

Attachments

  • D1EE88F8-951E-4BE1-BC06-59462B8A262A.png
    D1EE88F8-951E-4BE1-BC06-59462B8A262A.png
    408.9 KB · Views: 47

ruffrecords

Well-known member
Joined
Nov 10, 2006
Messages
13,403
Location
Norfolk - UK
I have seen a lot of passive monitor designs and I have often wondered why they work so hard to minimise the insertion loss. Back in the 70s when I was at Neve, the monitor section of a mixer was nearly always preceded by a 31267 transformer which stepped down the nominal +4dBu signal level down to about -8dBu. One reason for this is that signal typically peaks 16dB above the nominal +4dBu and if you fed +20dBu into the typical loudspeaker amp of the day it would clip. Things are not much different today with powered monitors so there is no real reason not drop the level significantly in the monitor chain. The second benefit of the transformer was it dropped the source impedance down to below 600 ohms which mean there was a lot less chance of HF loss in long cables.

So rather than increasing your pot to 25K, I would recommend you reduce it to about 2K which will give you a nominal insertion loss of about 10dB. It would also drop your source impedance to a minimum of about 1K3 rather than the current 4K or so. The problem with using pots for stereo controls is the tracking of log pots is poor. Linear pots are much better. At Neve we sometimes used a 2K LIN pot slugged with 620 ohms from wiper to the lower end. This gives you a pot which is 10dB down at the centre and tracks a lot better than a LOG pot.

Cheers

Ian
 
Last edited:

bjosephs

Active member
Joined
May 10, 2021
Messages
42
Location
Massachusetts
Hi Ian,


Salient points as usual. Ok, no fretting about insertion losses, lower value pots are better. Higher value doesn’t get me anything but higher impedance. With this implementation though (a rheostat strapped across phase and anti-phase) is there really any difference between a 2k pot at max vs. a 10k pot turned down to 2k? Seems like the result is the same and the lesson is to go turn my monitor input gain up to unity and be to dump some amplitude off at the attenuator.
 

abbey road d enfer

Well-known member
Moderator
Joined
Jan 22, 2008
Messages
12,360
Location
Marcelland
Increasing the impedance generally results in more sensitivity to interference and increased HF loss due to cable capacitance. IMO the latter is very often neglected.
 

bjosephs

Active member
Joined
May 10, 2021
Messages
42
Location
Massachusetts
Fortunately I am using a passive attenuator now - a 4 ganged 10k gold point attenuator - and ran a white noise test recently for unrelated reasons. There was no roll off at 20k. My cables are pretty short, maybe 10’ each so I think the ballpark of 5k is acceptable.
 

abbey road d enfer

Well-known member
Moderator
Joined
Jan 22, 2008
Messages
12,360
Location
Marcelland
Fortunately I am using a passive attenuator now - a 4 ganged 10k gold point attenuator - and ran a white noise test recently for unrelated reasons. There was no roll off at 20k. My cables are pretty short, maybe 10’ each so I think the ballpark of 5k is acceptable.
White (or pink) noise tests are to be taken with precaution. They often tend to be optimistic. this is related to the measuring filters.
 

ruffrecords

Well-known member
Joined
Nov 10, 2006
Messages
13,403
Location
Norfolk - UK
Hi Ian,


Salient points as usual. Ok, no fretting about insertion losses, lower value pots are better. Higher value doesn’t get me anything but higher impedance. With this implementation though (a rheostat strapped across phase and anti-phase) is there really any difference between a 2k pot at max vs. a 10k pot turned down to 2k? Seems like the result is the same and the lesson is to go turn my monitor input gain up to unity and be to dump some amplitude off at the attenuator.
Electrically there is no practical difference between a 2K pot at 100% and a 10K pot turned down to 2K. Operationally, the 10K pot is likely to have much less available movement to allow the operator to set the desired level - only you can decide if that works for you.

Cheers

Ian
 

bjosephs

Active member
Joined
May 10, 2021
Messages
42
Location
Massachusetts
Ok well I’ll definitely experiment a bit with values of cheap pots before investing in anything fancy. I decided to buy that relay board to shortcut some of the development work here - fingers crossed it isn’t junk. Now I have a much more important problem: switches

Torn between retro looking paddle rockers with a BBCish layout (replacing “mono to a” with a L/R flip) or pretty square illuminated push buttons…
 

Attachments

  • F8A54627-AB65-475C-A6C3-F9C1FD0E96CE.jpeg
    F8A54627-AB65-475C-A6C3-F9C1FD0E96CE.jpeg
    30.3 KB · Views: 36

abbey road d enfer

Well-known member
Moderator
Joined
Jan 22, 2008
Messages
12,360
Location
Marcelland
Ok well I’ll definitely experiment a bit with values of cheap pots before investing in anything fancy.
You'll find that cheap (and not so cheap) pots often don't track well, which results in lateral shift of the stereo image.
Torn between retro looking paddle rockers with a BBCish layout (replacing “mono to a” with a L/R flip) or pretty square illuminated push buttons…
 

bjosephs

Active member
Joined
May 10, 2021
Messages
42
Location
Massachusetts
Yup, I’m aware of tracking issues. I switched to the stepped attenuator I have now because I couldn’t find a potentiometer based setup that didn’t drive me crazy with balance errors. By experiment I’m referring to figuring out how low of a value I can go and still get enough level to my monitors. I’m sure Ian is right on the money with 2k though, just don’t want to drop $150 on a nice attenuator without some due diligence.

Thanks for the link on the switches
 

bjosephs

Active member
Joined
May 10, 2021
Messages
42
Location
Massachusetts
Very wise. Try with a couple of different value pots to get the level where you are comfortable.

Cheers

Ian
Ian,


Would you mind answering a couple of follow-ups on the dual linear + slug resistor approach?
  • You suggest a 2k total resistance to get about 10db down at max but wouldn't a 2k pot slugged with a 620 ohm from bottom to wiper be 620 ohms when turned all the way up? I'm thinking if I take this approach I should start higher than 2k so the slug resistor can do some work (5k with 1k5 slug, maybe?).
  • In THIS thread you describe matching the channels by scaling the slug resistor to match the total end to end resistance. That works with a potential divider arrangement but for the topology I'm doing I think I need to scale the slug resistors *and* the series droppers (2k2) for the channels to match. I'm not missing anything, right?

Thanks,
Brian
 

ruffrecords

Well-known member
Joined
Nov 10, 2006
Messages
13,403
Location
Norfolk - UK
Ian,


Would you mind answering a couple of follow-ups on the dual linear + slug resistor approach?
  • You suggest a 2k total resistance to get about 10db down at max but wouldn't a 2k pot slugged with a 620 ohm from bottom to wiper be 620 ohms when turned all the way up? I'm thinking if I take this approach I should start higher than 2k so the slug resistor can do some work (5k with 1k5 slug, maybe?).
Correct. I meant a regular 2K (LOG) pot would be 10dB down. If you want to use a slugged linear pot then you do need to account for the slugging resistor. For 10dB down at the centre of the pot the slug resistor needs to be about half the pot values (which means I was wrong when I said we used 620 ohms at Neve. In fact we used 1K. When the pot it fully up you have 2K//1K which is about 666ohms which is not too far away from the minimum load the Neve circuits were designed to drive). So if you want the parallel combination of the pot and slug to be 2K with the pot full up you need to scale everything up by three times. But nobody makes a 6K pot so your idea of using a 5K pot is fine. I would suggest using 2K4 slug resistor for the pot to be about 10dB down at the centre.
  • In THIS thread you describe matching the channels by scaling the slug resistor to match the total end to end resistance. That works with a potential divider arrangement but for the topology I'm doing I think I need to scale the slug resistors *and* the series droppers (2k2) for the channels to match. I'm not missing anything, right?

Thanks,
Brian
Good question. You can scale the slug resistor to match the pot but then the total value of pot plus slug may well not be the same for each channel so the static insertion loss for each channel could be different.. This means you would need to scale the resistors of one channel to match the static insertion loss of the other. This is not a problem if the pot is driven directly from a low impedance source but with 2K2 in both the hot and cold leads it is a different story. Starting to get a bit too complicated to be honest. We only used the slugged pot method on mono outputs at Neve; maybe this is why.

Starting to look like a stepped switch would be better.

Cheers

Ian
 

bjosephs

Active member
Joined
May 10, 2021
Messages
42
Location
Massachusetts
Starting to look like a stepped switch would be better.
Ya but I was probably going to end up there eventually after trying a few values. Looking around there is a low limit on stepped attenuators anyway - most are 10k or higher, goldpoint’s lowest is 5k. I have some Alps Blue pots in 10k so I’ll slap one in and see how high I end up turning it. That’s the nice thing about using a rheostat in this way, any pot can be a lower value pot by turning it down. If I only turn a 10k up half way I know a 2k would be fine, etc

Anyway I’m going to do some front panel layout stuff tonight, tomorrow I’ll wire my relays and try out the switching. Hopefully pick switches and stuff by the weekend.
 

totoxraymond

Well-known member
Joined
Oct 25, 2013
Messages
128
Location
Paris - France
Hi,

Thinking about it, does anyones really need 60dB range on a monitor controller?

Explanation: i find that most of the time i set my monitoring to the same exact level.

I have to say that i mostly mix TV shows (documentarys, cartoon shows, fictions...) and therefore aim at -23 LUFS. Having a monitoring level i'm comfortable with allows to not look at the meter too much and just get the thing as I want to hear it.

The few times i mix music, that's when i adjust my monitor level a bit more, depending on the situation, but even then, i tend to not go more than +5/-10dB.

So, regarding your thoughts about expensive attenuators, linear pots and all... this is what comes to my mind:

If i were to make such a device, i'd like to get:

1- Normal operation: adjustable level from -10dB to +10dB with center detent. (Allow the use of a lin, center detented pot if i'm not mistaken) This should be set around a reference level adjustable through a trim pot.

2- dim level: switch a fixed resistive bridge to -20 or -40. Again, switched with a trim pot

3- mute. An easy one...

4- trimpots to adjust each monitors output.

And of course, all the mono/st reverse that you wish.

Just my two cents.

Cheers,

Thomas
 

ruffrecords

Well-known member
Joined
Nov 10, 2006
Messages
13,403
Location
Norfolk - UK
Ya but I was probably going to end up there eventually after trying a few values. Looking around there is a low limit on stepped attenuators anyway - most are 10k or higher, goldpoint’s lowest is 5k. I have some Alps Blue pots in 10k so I’ll slap one in and see how high I end up turning it. That’s the nice thing about using a rheostat in this way, any pot can be a lower value pot by turning it down. If I only turn a 10k up half way I know a 2k would be fine, etc
If you build it yourself it can be any value you like. This is DIY after all. ;)

Cheers

Ian
 

Brian Roth

Well-known member
Joined
Aug 20, 2005
Messages
1,708
Location
Salina Kansas
My comments/recollections below are semi on topic....

Years ago I had to replace the CR volume pot in a small Neve 51xx (forget the correct series). The decades old OEM pot was a Bourns 10K dual linear in the square 5/8" blue package.

I had no schemos so I had to do some digging. I discovered two trim pots per channel. One was a variable "slug" from wiper to 0V. I guess Neve used that one to match the L/R audio levels at some position...mid rotation?

The second trimmer was more interesting. It was in series with the "low" end of the CR volume pot and 0V, thus establishing the minimum CR audio level when the pot was fully CCW.

Since that series of Neve desks was intended for the broadcast market, limiting the range of the CR volume pot makes some sense. More than a few times, I've been in TV station control rooms watching all the folks do "News9 at 10:00 PM" live broadcasts. Everyone....producers, directors, Chyron and Teleprompter ops....and the audio board op are all in one big room and hearing audio going on-air from the Neve (or whatever) desk through speakers in the front "monitor wall".

Anyway, I replaced the Bourns with an Alps "Blue Velvet" 10K dual audio taper pot. It was too large to fit into the Neve's CR module, so I "flew" leads to an adjoining blank panel. I disconnected the slug trimmers, but left the CCW trimmers in circuit.

BTW, the desk's owner LOVED the feel of the Alps vs. the old Bourns <g>.

Pic of that desk below.

Bri
 

Attachments

  • IMG_1383.JPG
    IMG_1383.JPG
    220.4 KB · Views: 25

Winston OBoogie

Well-known member
Joined
Jun 3, 2004
Messages
2,467
Location
UK.
+1 on the TKD, these track very well from side toi side. Worth the money I think.
The only complaint I've heard about them was that the feel, although very smooth, could do with being a little more stiff. Never bothered me, but I suppose a rubber "O ring" on the shaft with the knob providing some pressure would fix that.

The two Neve desks I owned, and others I worked on, used a Type 72 switch attenuator for control room and studio level. Series string of resistors with a nominal 1K total resistance. Angle of throw is 12 degrees per step so there are a few more positions available over an Elma 04 type which, I guess, was why Neve went with them.
You can still get the switches from Blore Edwards in the UK, but I've not yet been brave enough to ask the price.
 

Attachments

  • Type 72 switch.pdf
    868.2 KB · Views: 11

Latest posts

Top