You are here

Project: monitor controller

Customising, building or repairing your own gear? Need help with acoustic treatment or soundproofing? Ask away…

Project: monitor controller

Postby Frank Eleveld » Thu Apr 21, 2005 9:51 pm

Due to the apparent current interest in monitor controllers, I thought it would be nice to start outlining my plans for building a passive monitor controller here.

Mind that I'm occupied with a lot of other things at the moment and it could easily be a few months before I can actually start building a prototype, but this has the advantage readers can give their input as well - although I've read a lot on the subject lately, it doesn't mean I'm an expert on this.

Prelude
Most of the audio work I get to do here at home involves using my DAW system (currently Pro Tools LE/Logic Express on a G4 using a Digidesign Audio Media III card, though there are plans to upgrade the hardware and get a better, more versatile interface). I use a Spirit Folio mixer for monitoring, but it isn't very flexible in the routing department, and its sound quality is nothing to write home about.

A while ago, I started looking at units like the Mackie Big Knob and the Presonus Central Station, but found that especially the first unit received many negative comments from users who thought it cluttered the sound too much. The Presonus did better in this respect, but many American studio owners pointed at Coleman Audio monitor controllers and praised them for their sonic qualities. However, these units are not exactly cheap: prices start from USD 700-800 upwards, and I'm talking street prices here. RRPs are actually starting beyond USD 1,100.

Why are the Coleman units so good? We'll come back to that later. First of all, the Colemans are not the only monitor controllers getting a lot of praise, there are some others. In the 'Drawing Board' subforum of Prodigy Pro (pro audio DIY), the subject of making an affordable routing switch pops up quite frequently and as such I started to dive a little deeper into the subject.

If I were to make a simple switch, what features would I implement? Often, projects like these start off simple but all too soon they'll continue to gain weight and features and this project is no different - so far at least.

Designing
My original requirements were: having a simple, passive routing switch, capable of routing at least three balanced stereo inputs (CD player, mixing console main outs and 2-trk output of the DAW) into at least three stereo outputs (for routing the signal to my monitor amp, and, for example, the first two inputs on the interface of my DAW, and to another recorder).

Volume of the monitor amp output needed to be controlled; this way my Quad 34 control centre - a nice unit but mine needs serviceing and it colours the sound because it's active - could be omitted too. Building a passive routing switch with the above features implemented isn't overly difficult, and basically, one could also buy one of the GoldPoint monitor controllers. However - these don't have the required number of inputs and outputs, and are a tad pricey (though very good quality).

So I started thinking... 'If I want to do this properly, what features would I need more?' I don't do a great deal of recording in my home studio (which is called 'Leisure Studio' BTW, hereafter referred to as 'LS'), but nevertheless having a talkback feature in the unit would be a nice option.

But I don't want to spend a lot on high end mic pres, a simple affair employing a back-electret module will do for the purpose. Basically, it means that a simple op-amp circuit in 40dB gain topology would do. As I want to keep things simple, I'm not intending to design such a circuit myself; I'd rather prefer a readily-available design. Suppliers like Velleman and Conrad have kits available that seem to fit the requirements of a talkback mic and pre amp quite nicely, for not too much money (around a tenner, talking in Sterling here).

Another option would be to buy a few master or routing modules of a scrapped desk, removing the bits unnecessary and using the other parts in the controller unit. Virtually any modular desk will do, the pre amp doesn't need to be highest quality because we'll be using it for brief speech communications with guest musicians only. Choosing the 'scrapped desk' option would also fit the 'headphone amp' bill quite nicely...

I've built a few headphone amps (all PIMETAs) in different configurations and generally speaking, these are much better (especially in terms of sound quality) than the average desk's headphone amp output. Obviously, the engineer needs to have a crisp and clean headphone output, not a compromised one - (s)he'll need to hear everthing as close to the original sound as is rightly possible. A PIMETA headphone amp employs parts that may be difficult to obtain in some parts of the world, moreover, it may be overkill for the intended purpose. A humble CMoy headphone amp (a minimalistic design created by Chu Moy of HeadWize might do equally well. Best of it all is that its components are commonly available for the most part, and that it's cheap and easy to build (on a piece of strip board). I'll not be discussing the headphone amps in further detail, there are loads of information on sites like Headwize, etc. Building a project like this foremost requires at least some search engine skills

I think I'd be happy using just one headphone output on my monitor controller, and use a pair of balanced outputs to route a cue feed to the musicians who're using their own headphone amps. However, your mileage may vary, and you may want to power multiple headphones using a single headphone amplifier. A buffered PIMETA amp will be capable of doing just this without too much problems, but a CMoy will probably not like it, especially when using high-Z headphones.

OK, so we're now working on designing a routing switch with volume attenuator on the main amp outputs, an engineer's headphone output, three inputs, three outputs, and a talkback mic that needs to be switched in by pressing a non-latching 'TB' switch (otherwise it'd be open all the time, and although this can create a hilarious atmosphere in some circumstances, I'd prefer to be able to have some sort of control over muting/unmuting it).

But... employing a pre amp and talkback microphone, as well as a headphone amplifier also brings in the requirement for a power supply. I figure a power supply capable of supplying a voltage of around 12VDC at a rate of 1A current would be capable of supplying both a feed to the pre amp as well as one to the headphone amplifier, thereby leaving a nice power reserve for future expansions

OK, that's a power supply then as well. A Velleman K1823 board, which employs a LM317 regulator allowing its voltage output to be adjusted over a range of 3-28VDC, has decent specifications for audio, especially if you'd replace the onboard pot, power regulator and big electolytic capacitor with something higher spec'ced. Ripple rejection is more than 80dB in its standard configuration, far better than most off-the-mill wall-warts and many other power supply kits. But there's another option coming soon, slightly more expensive, but it seemingly has even better specs than the Velleman. I won't post the information here yet because the design of the PSU board is currently in the experimental phase and the first trial version isn't available anymore, so we'll have to wait a little longer (or build our own PSU).

Important Household Messages
Whilst on the subject of PSUs: remember electrical power, especially AC, is potentially lethal. If you don't feel confident of what you're doing, then please have someone else doing the building for you. In this article, I assume the builder has basic understanding of signal flow and electronics, in much the same way as I've myself. If anyone feels I'm wrong in any aspect, or if my explanations or details are incorrect or insufficient, please don't hesitate to reply here or PM me - I'd like to make this a project for all those interested, not just a project of my own. If you feel you've anything useful to contribute, please do so as well!

Back to the drawing board
OK, what do we have so far?
- A monitor routing switch, capable of switching 3 balanced inputs into 3 outputs;
- Main amp volume output is adjustable by means of an attenuator;
- Talkback facility with integrated pre amp and back-electret mic capsule;
- Engineer's headphone output;
- Active circuitry powered by a simple, yet good quality integrated power supply.


(To be continued...)
User avatar
Frank Eleveld
Moderator
Posts: 1167
Joined: Sun Aug 29, 2004 11:00 pm
Location: NL

 


Re: Project: monitor controller

Postby PWGLE » Thu Apr 21, 2005 10:11 pm

Its somthing I'm also considering... Most of the time all i need is routing because the Firewire interface has most the in/outs I use day to day...

I was thinking of using a simple PIC system I have experience in, to switch passive components and control analougue parameters, and then a nice LCD to really infomation back on... The idea is the advantages of being able to control things easily and program things.

I was thinking keep the audio passive which goes to monitors, etc. But have a active headphone amp, and a few other useful bits and bobs... I recently designed a test tone generator which I would definatly put in because it is so so useful..

Maybe some forum collaboration is in order?
laurence
PWGLE
Regular
Posts: 100
Joined: Sat May 03, 2003 11:00 pm

Re: Project: monitor controller

Postby Frank Eleveld » Thu Apr 21, 2005 10:19 pm

Let's have a look at the currently available commercial products, just to get an idea of what functionality we could employ in our own project.

Many of the monitor controllers available today have:
- A 'Mute' switch (muting the output to the control room monitors);
- A 'Dim' switch (cutting the output to the control room monitors by a certain amount, usually -20dB or slightly more);
- A 'Mono' switch (summing the stereo signal sent to one or more outputs into a mono signal);

Some, slightly higher-end models also have:
- A 'Phase Reverse' facility (allowing to present the listener with a phase-reversed version of one signal for checking problems with phase coherency)
- More flexible routing options.

Fitting it all in
Although my project is still on the drawing board (I haven't even made a full parts-list yet) one of the questions arising sooner or later is what enclosure to fit the project in.
Hammond and Isel are two manufacturers making enclosures that would suit smaller devices like headphone amps and power supplies quite well, but since I'm planning to employ Neutrik XLR-connectors throughout, the physical size of the connectors will restrict me from using a compact case, and I'll have to resort to a 19" case.

Metal work can be a huge pain, especially when not particularly skilled in this area (like me), so I prefer to have an enclosure of aluminium - which is softer and thus easier to work with.

The best 19" enclosures I've seen so far are obtainable from
Elgee in Australia - not exactly cheap (I recall around AUS70 or thereabouts, excluding VAT and shipping), but beautifully modular in construction, and with aluminium front- and end panels.

But... having the main unit in a rack isn't always the most ergonomic solution, some of us would like to have the unit right in front of them without having to reach for the rack to make the necessary routing and/or volume adjustments. The obvious solution would be to incorporate some form of remote control.

Remote control
Adding a remote control to the project will immediately raise a few problems. The original idea was to make a passive controller, which means the basic signal flow going through the unit isn't touched by electronics that alter the sound (which includes amplifier circuitry etc.), apart from the headphone power output which will be active.

Employing a passive design puts up requirements on the impedance of cables and connections, signal routing should be as short as possible as to avoid impendance related problems. Ideally, the signal path in the unit should not be longer than around 50, 60 cm. (according to a few designers who've already made similar units). By adding a remote control, this rule is obviously broken. Until today, I've not yet decided as to implement a remote control facility (this basically means I'll have to abandon the passive idea and employ line buffers for buffering the output to the remote) or leaving the remote idea and keeping the whole design passive. Ah, decisions...

(To be continued... soon I hope )
User avatar
Frank Eleveld
Moderator
Posts: 1167
Joined: Sun Aug 29, 2004 11:00 pm
Location: NL

 


Re: Project: monitor controller

Postby PWGLE » Thu Apr 21, 2005 10:24 pm

I was thinking if you keep the audio line passive, and use audio grade transistors, gates, etc to switch in different elements (phase, pads, etc)

You could control this digital thought an interface and even use a remote, but the signal would stay passive. Remote controls are quite easy I build one a few months back, but that was using IR which is not completly idealy as the range is a bit sucky. I'm currently looking into proper wireless remote systems, such as the one the ATi Remote Wonder is based on..,.

All interesting stuff Frank.,.,
PWGLE
Regular
Posts: 100
Joined: Sat May 03, 2003 11:00 pm

Re: Project: monitor controller

Postby Frank Eleveld » Thu Apr 21, 2005 10:26 pm

Pirate Giant with Laser Eyes wrote:I was thinking of using a simple PIC system I have experience in, to switch passive components and control analougue parameters, and then a nice LCD to really infomation back on... The idea is the advantages of being able to control things easily and program things.

That would be a nice option indeed - it would be great to be able to control and program routing and being able to read it off an LCD display.

Pirate Giant with Laser Eyes wrote:I recently designed a test tone generator which I would definatly put in because it is so so useful..
Hmmm... that's something I also have still lying about... Never could put it to use so far, but it's definitely worth considering employing it in the controller.

Pirate Giant with Laser Eyes wrote:Maybe some forum collaboration is in order?

That's what I'm aiming for - I'm no expert on electronics and there really are a lot of considerations to make when building a project like this. For example: passive/active, how many I/Os, what facilities, volume pot or expensive stepped attenuator, etc.

I thought it would be nice to put my thoughts and considerations on-line, even though I'm still at the drawing board stage and will probably not able to start building until I've returned from holiday this Summer.

It's good to have inputs and insights on this project, I think the DIY community on the SOS forums is a bit behind compared to some other forums. Let's try to change that

Thanks for your input!
User avatar
Frank Eleveld
Moderator
Posts: 1167
Joined: Sun Aug 29, 2004 11:00 pm
Location: NL

 


Re: Project: monitor controller

Postby Scottdru » Thu Apr 21, 2005 11:04 pm

Frank Eleveld wrote:
Pirate Giant with Laser Eyes wrote:Maybe some forum collaboration is in order?


That's what I'm aiming for - I'm no expert on electronics and there really are a lot of considerations to make when building a project like this. For example: passive/active, how many I/Os, what facilities, volume pot or expensive stepped attenuator, etc.

I thought it would be nice to put my thoughts and considerations on-line, even though I'm still at the drawing board stage and will probably not able to start building until I've returned from holiday this Summer.

That's a great idea, guys! I might be up for building something like that . . . maybe toward the fall or something. We should definitely start some DIY collaborations to get things moving a bit in here.
When are you coming back from holiday, Frank?

The other, more urgent thing for me is to build something with which I can run a couple of headphone mixes out to musicians, or possibly even having a four input thing that will allow the musicians adjust their own headphone mixes from four submixes sent to those inputs.

I have a few other projects on my plate at the mo, including refurbishing and modding my Rhodes 73 (among other things, I've got a mod kit to improve the action from a guy here in NYC who works on Rhodes and Wurlitzers for a lot of big names (Norah Jones et al.) and some big studios, etc.). I also need to sort my acoustics in my studio and to rack up a couple of channel strips I've got coming. Those three things are first on the list and need to be ready by around the first of June. We'll see what money is available after that, because I suspect my hours at work may be cut again in the summer.

I'll start thinking and looking around on both the monitor controller thing and the headphone thing. I'm guessing these two things could either be optionally combined, or we could get enough people interested in these two projects to get something going. I think these two things are probably things that many people need for their DAWs, since we get questions about them fairly often in the other forums. It would be nice to have some better options than to buy the Samson and Behringer crap.

Perhaps we could look at building any of this stuff in a more or less modular sort of way (using existing designs, as Frank suggests), so that people could add or delete features as needed, or build and use in sections as their needs and/or budget dictate?

It could also be good having guys on both sides of the pond working on this, so that we can offer helpful information regarding sourcing the parts.
User avatar
Scottdru
Frequent Poster
Posts: 1160
Joined: Tue Dec 17, 2002 12:00 am

Re: Project: monitor controller

Postby Frank Eleveld » Fri Apr 22, 2005 9:06 am

Scottdru wrote:
When are you coming back from holiday, Frank?


I'm away for the best part of June, and my schedule is pretty filled up until then.

Scottdru wrote:The other, more urgent thing for me is to build something with which I can run a couple of headphone mixes out to musicians, or possibly even having a four input thing that will allow the musicians adjust their own headphone mixes from four submixes sent to those inputs.


I built a passive splitter box last year, designed to split one stereo feed into three stereo outputs. It lacks individual volume controls on the outputs, but other than that I've used it successfully on a number of occasions. The biggest disadvantage is that the headphone impedances need to be close to eachother, otherwise there's a considerable difference in loudness between the headphones.

To overcome this, and to increase flexibility, building a multi-output headphone amp is probably the best option. Perhaps I can start another topic here about building headphone amplifiers as to avoid getting off-topic here.

Scottdru wrote:
We'll see what money is available after that, because I suspect my hours at work may be cut again in the summer.


The nice thing about the project as I've in mind is that it can be as esoteric as one would like. If you've already some parts lying around, or if you'd accept to implement cheaper parts, the building costs of a simple passive monitor controller can very low indeed.

Scottdru wrote:Perhaps we could look at building any of this stuff in a more or less modular sort of way (using existing designs, as Frank suggests), so that people could add or delete features as needed, or build and use in sections as their needs and/or budget dictate?


Exactly my idea. What works for me, doesn't necessarily have to work for someone else, I'd like to outline a basic feature set, and ask for input from those interested for additional features.

Scottdru wrote:It could also be good having guys on both sides of the pond working on this, so that we can offer helpful information regarding sourcing the parts.


This is exactly the way it works in many other DIY-forums and it would be nice to see a similar cooperation in the SOS forums as well. BTW: a passive controller doesn't need many parts, just a few input and outputs, low-capacitance wiring, a few switches, a pot or stepped attenuator, some resistors and a case are basically all that's needed.

What I'd like to implement are the BBC features Hugh is frequently raving about, like individual channel mutes (Left Mute, Right Mute, L+R Mute), sum the stereo output to mono with one channel phase-reversed, etc.

The best thing about DIY is that you can make your project anyway you like it, following your needs and budget
User avatar
Frank Eleveld
Moderator
Posts: 1167
Joined: Sun Aug 29, 2004 11:00 pm
Location: NL

 


Re: Project: monitor controller

Postby Frank Eleveld » Mon Apr 25, 2005 9:47 pm

OK, some theory for those interested in participating in the 'passive volume controller' project.

Following are the (basic) specifications of the unit I'm planning to build:

- At least three (minimum) stereo inputs, on balanced Neutrik XLR females (balanced jack sockets are a cheaper option for those interested);
- Rotary switch for routing input signal;
- Passive attenuator or high quality volume pot (depending on budget);
- Rotary switch for selecting mute L, mute R, mute L+R (optionally, 'dim' at which the signal is cut by 20dB);
- Phase reverse option;
- At least three output options on stereo balanced (Neutrik) male XLRs;

Niceties:
- At least one headphone amplifier output (active) with a link option to distribute signal to external pre amp;
- Tangent's 'TREAD' power supply (more about this later);
- Talkback circuit


Motivation for the choices listed above
I'll be choosing quality Neutrik NC3 series XLR connectors. These don't come cheap (at around four quid each, prices may vary) but I've found them to be very reliable. Some cheaper (Neutrik imitation) XLR jack sockets literally melt when soldering wires to them, the metal busses inside will come loose pretty quick when the terminals are heated at too high a temperature or just a little too long. Obviously, this doesn't contribute to reliability. The Neutriks are much better in this respect.

I'm not yet sure what attentuator I'll be using. A reasonably low-impedance stepped attentuator is the best option for several reasons. Pots usually have tracking errors (differences) in the left/right paths and are not fully linear (for audio work most pots are actually logarithmic, otherwise most of the volume change would happen at the first millimetres of travel of the pot), and with stepped attenuators, the output can be controlled much more precisely. There are two options when choosing stepped attenuators; making your own (employing an empty switch and soldering high quality resistors to the switch's terminals). Depending on the number of gain steps you'd want to implement, this process can be very awkward and long-winded, but it saves a lot of money. The other option is to buy a nice DACT C-series attenuator, but these are expensive! A passive attenuator like the aforementioned DACT will easily account for 75 per cent of the total building costs.

Since I'm aiming for keeping the entire signal path balanced throughout the unit, we'll be needing a stepped attenuator that has four 'decks' (layers) to which we'll be connecting the signal terminals. Namely, we'll be connecting the positive left signal to deck 1, the negative left signal to deck 2, the positive right signal to deck 3, the negative right signal to deck 4 (just an example). We won't be switching grounds, these are star-grounded in my design.

Signal switching needs to be carried out in much the same fashion; we'll be needing four terminal switches for switching just a stereo signal, hence my choice for rotary switches. Lorlin's for example, carry cheap rotary switches that are available in a number of pole-configurations. More of this later when we'll be getting closer into the actual design. For now, it's pretty all preliminary stuff...

Then over to the PSU section. If we'd be integrating an active headphone stage (just a simple headphone amplifier circuit like the famous CMoy amp will do fine at not too much cost) and a talkback system, we'll be needing a power supply of some sort. I've praised the Velleman K1823 PSU kit on a number of occasions because these are commonly available, and fairly cheap (At around £ 8). The only additional components necessary are a heatsink and a transformer and you've got a basic power supply with decent specs for audio.

However, Tangent (not the guy's real name obviously ), one of the designers of the PIMETA and PPA headphone amplifier projects which get a lot of praise amongst headphone lovers, has designed a simple power supply that seemingly outperforms the Velleman kit considerably. I won't be going into further detail about the merits of these PSUs, because there is sufficient information on Tangent's website about it. The TREAD (current revision is v1.1) can be obtained as a kit from Tangent for USD 15, the PCB is USD 5. I'd recommend to buy the entire kit unless you've already most of the parts.

Just to finish this update, I'll be providing a link from DACT with more information on attenuators. I consider this an essential read, especially since there are a number of caveats with passive projects like this one - caveats we only will avoid by knowing our theory. I recommend visiting the above links and reading some of the theory provided on the sites.

Cheers for now,
Frank
User avatar
Frank Eleveld
Moderator
Posts: 1167
Joined: Sun Aug 29, 2004 11:00 pm
Location: NL

 


Re: Project: monitor controller

Postby Scottdru » Mon Apr 25, 2005 10:41 pm

Frank, with regard to building our own stepped attenuators, is there any liklihood of inconsistency between resistors, to the point of needing to go through and hand match them all?

If so, would this add to the cost enough that it would be approaching the cost of simply buying a stepped attenuator? Of is this less likely to be an issue with resistors than with other components?

I don't have any real experience with that kind of thing, other than that I have read about certain components needing to be checked and matched in various other projects.
User avatar
Scottdru
Frequent Poster
Posts: 1160
Joined: Tue Dec 17, 2002 12:00 am

Re: Project: monitor controller

Postby PWGLE » Mon Apr 25, 2005 11:03 pm

If we are going to collaberate, maybe we should start building a list/specification we can work to...

I could set some ftp access up on a folder on my server? So its nice and easy to access?

Talk soon chaps,


Oh I'm very interested in that power supply module, I normally design and build specific power suplies depeding on application, but that looks quite suited. Another thing I think should be added, is a powered usb socket. These sockets are damn useful for lights, or if you need to quickly charge a usb device. I've added usb sockets on to all my mixers so I can use those LED flexible lights with them (so much cheaper than goosenecks!).

I remeber reading about a audio IC a few weeks back, which converts any varible resistor signal/pot into a very accurate 2 channel volume control.

Talk soon, Laurence
PWGLE
Regular
Posts: 100
Joined: Sat May 03, 2003 11:00 pm

Re: Project: monitor controller

Postby Hugh Robjohns » Mon Apr 25, 2005 11:05 pm

A couple of suggestions from me.

If you want to keep the signal path passive and balanced, then I would suggest using a variable transformer arrangement for the level control -- Sowter do a reasonably good system, and 0VU has often mentioned another more expensive (but apparently even better sounding) make. Can't remember it's name right now but it'll come back to me I'm sure. it's expensive, but offers far better balancing than the switched attenuator ideas, and is not that much more expensive.

I'd also suggest inorporating a balance control, ideally with a click-stop centre position. Surprisingly useful facility to have when checking the monitoring arrangement in an unfamiliar environment -- like when recording on location.

Finally, I'd suggest staying well away from LCD screens and any kind of processor-controlled switching systems. Keping the processor clocks and switching noises out of the passive analogue circuitry needs very careful designs and board layouts indeed.

Hugh
User avatar
Hugh Robjohns
Moderator
Posts: 16761
Joined: Thu Jul 24, 2003 11:00 pm
Location: Worcestershire, UK

Technical Editor, Sound On Sound


Re: Project: monitor controller

Postby hughb » Mon Apr 25, 2005 11:21 pm

Is a balanced internal signal path really necessary/feasable? I would have thought that having balancing transformers or other circuitry at the inputs and outputs, and a reasonably well shielded enclosure would sort out any significant interference problems.

There's a strong possibility that I'm talking rubbish.
User avatar
hughb
Poster
Posts: 29
Joined: Thu Feb 20, 2003 12:00 am

Re: Project: monitor controller

Postby hughb » Tue Apr 26, 2005 9:43 am

Double-posting I know, but I had some more thoughts on the subject of balancing last night. Anything to avoid revision!

Do I understand you correctly Frank, when you say 'keeping the signal path balanced through the unit' you mean connecting both legs of your input connector directly to the circuit, and carrying them both through the unit to the output?

My concern is that for a balanced circuit (and by circuit I mean a single XLR cable or the suchlike) to reject noise, both legs have got to be electrically identical, and, ideally, occupying pretty much the same axis. If the monitor controller is completely in line, I would imagine that it will be nearly impossible for you to keep both legs electrically identical, what with different track lengths and, more importantly, paths on the PCB, and the possibility that they may pass through non-identical components.

This would mean that any noise introduced in the connection between your source and the controller, or between the controller and the monitor/amp/defunct TV star and German rock legend David Hasselhoff will no longer be common-mode, as the circuit now includes the monitor controller, and as such will not be rejected by the balancing circuitry at the monitor/amp/hoff.

Much better (IMHO) to electrically divorce the input and output circuits with balancing transformers, and let the input transformer reject any noise in the circuit between source and controller, and the monitor balancing circuitry to deal with any noise that turns up between controller and monitor.

I hope that's mostly right... I've got an audio electronics exam in less than two weeks...

Hugh
User avatar
hughb
Poster
Posts: 29
Joined: Thu Feb 20, 2003 12:00 am

Re: Project: monitor controller

Postby Frank Eleveld » Tue Apr 26, 2005 4:13 pm

Hugh Robjohns wrote:A couple of suggestions from me.

Many thanks for your input Hugh - always nice to have your opinion for projects like these. I regard your contributions as very valuable indeed; passive designs like these are fairly easy to build, but there are so many insights on the pros and cons of passive audio projects that it's hard to make out what's true and what's not.

@Scott - of course we could hand-match the resistors (IF you'd build the stepped attenuator yourselves), many hi-fi boffins would make you believe it's absolutely necessary in order to maintain consistency between the left and right channels.

I think that if we'd be using 1% resistors, the individual differences between the values are already much better than we would get using the best pots available. So basically, I'd say it isn't necessary to hand-match the resistors. It may well be some mastering engineers have other ideas on this.

If I can find some time this evening I'll continue working things out and posting the odd theory here.

Cheers,
Frank
User avatar
Frank Eleveld
Moderator
Posts: 1167
Joined: Sun Aug 29, 2004 11:00 pm
Location: NL

 


Re: Project: monitor controller

Postby cc. » Tue Apr 26, 2005 5:48 pm


(I'm playing devil's advocate here to get you thinking )

I don't think you should think of passive designs as being perect and not colouring the sound at all. If you look at the attenuator, the and the input capacitance of the monitors you've basically built a filter. This is without even considering all the other capacitances and inductances involved. Plus you are changing the load on the output as you change the attenuation.

It seems to me that 4 gang attenuators are so expensive that you could certainly afford to do a really good active design for the same price if you chose to go that way. Certainly a better stage than is likely to be in the output of your soundcard!

Also, I don't think the idea that a passive design is much simpler holds water. Sure it's simpler if you're just building an attenuator, but given you're doing mono summing and headphone amps etc you've got nearly everying you need to do an active design already.

Just some thoughts...
User avatar
cc.
Regular
Posts: 96
Joined: Tue Mar 11, 2003 12:00 am
Midipicks - the all new MIDI Guitar forum...

Re: Project: monitor controller

Postby Frank Eleveld » Tue Apr 26, 2005 8:06 pm

Thank you for your input CC!

cc. wrote:
I don't think you should think of passive designs as being perect and not colouring the sound at all. If you look at the attenuator, the and the input capacitance of the monitors you've basically built a filter. This is without even considering all the other capacitances and inductances involved. Plus you are changing the load on the output as you change the attenuation.

Yes. The effect that output load changes when changing attenuation is the biggest drawback with passive stages. This is basically happening because there is no buffer stage between the attenuator and the output. If I understand it correctly, a well-designed stepped attenuator is preferable over a pot because its load can be made constant, regardless of its setting.

cc. wrote:It seems to me that 4 gang attenuators are so expensive that you could certainly afford to do a really good active design for the same price if you chose to go that way. Certainly a better stage than is likely to be in the output of your soundcard!

Hmmm, not convinced. I've looked at various active controllers, and the 'home studio grade' designs such as the Mackie Big Knob are quite horrible in that the audio passes as many as six gain stages. The DACT CT2 four gang 24-stepped rotary switch is indeed expensive (I've found an unconfirmed retail price of € 200 including VAT here in the Netherlands), but I'm still unsure whether I'll choose that expensive option.

cc. wrote:Also, I don't think the idea that a passive design is much simpler holds water. Sure it's simpler if you're just building an attenuator, but given you're doing mono summing and headphone amps etc you've got nearly everying you need to do an active design already.

Very true. The idea in this project is that it's modular, people who're not needing a headphone section can omit it, same for the mono summing and all other bits. The version I've on the drawing board now is basically derived from monitor controllers such as the Coleman Audio MH-series - these are passive, but do contain a powered headphone buss.

Your points are all valid, and there are still some considerations to be done - which is why I've posted my thoughts here - everyone is free to shoot at them
User avatar
Frank Eleveld
Moderator
Posts: 1167
Joined: Sun Aug 29, 2004 11:00 pm
Location: NL

 


Re: Project: monitor controller

Postby Frank Eleveld » Tue Apr 26, 2005 8:25 pm

hughb wrote:Do I understand you correctly Frank, when you say 'keeping the signal path balanced through the unit' you mean connecting both legs of your input connector directly to the circuit, and carrying them both through the unit to the output?

Yes.

hughb wrote:My concern is that for a balanced circuit (and by circuit I mean a single XLR cable or the suchlike) to reject noise, both legs have got to be electrically identical, and, ideally, occupying pretty much the same axis. If the monitor controller is completely in line, I would imagine that it will be nearly impossible for you to keep both legs electrically identical, what with different track lengths and, more importantly, paths on the PCB, and the possibility that they may pass through non-identical components.

First of all, I'm not planning to use any PCBs, apart from the PSU and headphone amp board (though the latter also may be built on a piece of stripboard if it's a simple design like a CMoy.

In a passive circuit, different cable lengths will have an effect on the impedance load. However, the differences we're talking about with regard to this project are negligible - and not of importance anyway, since in a balanced circuit, any differences between the signal carried on the hot and cold terminals are cancelled out - an advantage over an unbalanced system.

hughb wrote:Much better (IMHO) to electrically divorce the input and output circuits with balancing transformers, and let the input transformer reject any noise in the circuit between source and controller, and the monitor balancing circuitry to deal with any noise that turns up between controller and monitor.

Hugh R also brought this up - I think it's a sensible idea and I'll definitely consider it.

Thanks.
User avatar
Frank Eleveld
Moderator
Posts: 1167
Joined: Sun Aug 29, 2004 11:00 pm
Location: NL

 


Re: Project: monitor controller

Postby James Perrett » Wed Apr 27, 2005 11:48 am

Frank Eleveld wrote:

hughb wrote:Much better (IMHO) to electrically divorce the input and output circuits with balancing transformers, and let the input transformer reject any noise in the circuit between source and controller, and the monitor balancing circuitry to deal with any noise that turns up between controller and monitor.

Hugh R also brought this up - I think it's a sensible idea and I'll definitely consider it.

Thanks.

I'd vote for no transformers - good transformers are expensive and could push the cost of the project out of reach of some people. They will also add unnecessary complexity in many situations and some would argue that even the best transformers degrade the sound.

Cheers.

James.
James Perrett
Moderator
Posts: 5760
Joined: Sun Sep 09, 2001 11:00 pm
Location: The wilds of Hampshire

JRP Music - Audio Mastering and Restoration.
http://www.jrpmusic.co.uk


Re: Project: monitor controller

Postby cc. » Wed Apr 27, 2005 12:07 pm

I thought Hugh (R) was talking about using a variable transformer as the level control (instead of a resitor based attenuator) rather than transformers on the input and output.
User avatar
cc.
Regular
Posts: 96
Joined: Tue Mar 11, 2003 12:00 am
Midipicks - the all new MIDI Guitar forum...

Re: Project: monitor controller

Postby Frank Eleveld » Wed Apr 27, 2005 12:32 pm

I already expected different insights on this project - I've noticed similar differing opinions on another forum where passive 'mastering consoles' and 'monitor controllers' were subject of discussion.

The ultimate goal of this project is to make a monitor controller that's versatile, reasonably easy to build, even for the electronic novice, and not overly expensive.

I've given CC's suggestion on making an affordable active design a thought. We could implement an active version by employing a low-gain amplifier circuit (once again, the CMoy headphone amp would be suiteable due to its simplicity, reasonable performance and low cost) but such circuits are mostly unbalanced. This would mean that we'd have to drop the balanced concept, or make the necessary balanced-to-unbalanced, and unbalanced-to-balanced conversion stages to fit within the circuit.

So far, we've discussed a lot about theory and before starting out, some considerations would have to be overtaken. As an example, I'd like to point to a link where 'bblackwood' showed his passive 'mastering console' on the PSW Recording Forums. This device is a more complicated version of the idea I had in mind.

There are some excellent photographs on this page. As you can see, there is not a single circuit board in sight, the whole circuit is point-to-point wiring using high quality switches.

Just to give anyone an idea of what's possible using a few switches and some wire...
User avatar
Frank Eleveld
Moderator
Posts: 1167
Joined: Sun Aug 29, 2004 11:00 pm
Location: NL

 


Re: Project: monitor controller

Postby forumuser729356 » Wed Apr 27, 2005 1:45 pm

Frank
Watching this with interest - I spend more time over at the lab nowadays amassing informtation that I hope one day I will make sense off
I do believe that Kev from the Lab - has made a schematic of a talkback unit available through the DIY factory site
Is the box starting to look a lot like Hughs bespoke BBC type monitor box - as pictured and reviewed in a previous SoS? - If so...mmmhh count me in...
(Hugh - any chance of a piccie of the insides???)
forumuser729356
Regular
Posts: 51
Joined: Wed Nov 02, 2005 12:00 am

Re: Project: monitor controller

Postby Hugh Robjohns » Wed Apr 27, 2005 2:21 pm

If I was building my own monitoring controller today, I think I would base it on a variable tranformer. The advantage of this approach is that

a) relative simplicity in wiring and components
b) passive and balanced connections throughout
c) can be operated with unbalanced or balanced circuits on either side without problems
d) inherently prevents ground loops

The Sowter transformer(s) maintain extremely good sound quality -- even 0VU commented on it. There is a better version for the real golden eared from another company, but it is a lot more expensive.

It would be easy to incorporate a simple active talkback circuit, cue light switching and anything else you might want, without compromising the main monitoring path.

I'm quite happy to put up some piccies of my bespoke controller and piccies of the innards... but it's not that pretty As I've said before, it was constructed using a bunch of standard cards left over from other Audix projects, so there is a lot of circuitry in it that is not actually being used. It uses active electronics to unbalance the various inputs, perform all the level control functions etc, and rebalance everything for the outputs -- all pretty standard stuff.

It uses relatively expensive components and the circuit designs are good, but it certainly isn't the last word in total transparency -- it's akin to the monitoring section of a good console (which is what it really is), rather than the ultimate hifi tweakery oxygen-free, rolled gold and dunked in unicorn spit bollox!

For me, though, the practical advantages outweigh any inaudible theoretical loss of fidelity, especially given that I use it mainly when working in untreated rooms on location -- so the whole moitoring chain is inherently compromised anyway.

Having said all that, the operational features and facilities are excellent and I would certainly try to mimic those in whatever I chose to replace it with, wether active, passive or whatever.

I'll try to get some piccies sorted for tomorrow.

hugh
User avatar
Hugh Robjohns
Moderator
Posts: 16761
Joined: Thu Jul 24, 2003 11:00 pm
Location: Worcestershire, UK

Technical Editor, Sound On Sound


Re: Project: monitor controller

Postby hughb » Wed Apr 27, 2005 3:28 pm

James Perrett wrote:

I'd vote for no transformers - good transformers are expensive and could push the cost of the project out of reach of some people. They will also add unnecessary complexity in many situations and some would argue that even the best transformers degrade the sound.

Cheers.

James.

I just mentioned transformers as a balancing option - there are several transformer-free electronics based balancing options. Just don't ask me to draw you a circuit diagram!

Hugh
User avatar
hughb
Poster
Posts: 29
Joined: Thu Feb 20, 2003 12:00 am

Re: Project: monitor controller

Postby Frank Eleveld » Wed Apr 27, 2005 4:19 pm

Hugh Robjohns wrote:
I'm quite happy to put up some piccies of my bespoke controller and piccies of the innards...

That would be great!

Hugh Robjohns wrote:but it's not that pretty

As long as it serves the intended purpose(s) as it's designed to do well, then it can't be ugly in my opinion

Hugh Robjohns wrote:
Having said all that, the operational features and facilities are excellent and I would certainly try to mimic those in whatever I chose to replace it with, wether active, passive or whatever.

I must admit that some ideas of employing certain functions (muting individual channels, phase-reversing one channel, etc.) of the controller project are based on comments made in your (or Paul White's) reviews and Studio SOS articles in which aforementioned features are highly praised. And so my original plan to replace my Quad 34 pre amplifier with a simple attentuator has developed into a more extensive project...

I'm considering to put the building process of a headphone amplifier in another thread here in the DIY section - just to avoid clogging this thread with a myriad of different topologies and components, because I consider the monitor controller to be a modular project; one can add any features (s)he'd need. For some, the full feature set will prove a little too daunting, for others, the basic attentuator/input switch idea will prove too limiting. That's the biggest advantage of DIY, one has the ability to construct a unit exactly to the desired needs...

More soon!
User avatar
Frank Eleveld
Moderator
Posts: 1167
Joined: Sun Aug 29, 2004 11:00 pm
Location: NL

 


Re: Project: monitor controller

Postby Scottdru » Thu Apr 28, 2005 1:34 am

rather than the ultimate hifi tweakery oxygen-free, rolled gold and dunked in unicorn spit bollox!
LOL! I don't know about you guys, but I'm dunking my whole unit in unicorn spit when it's finished.

(hehe . . . he said "unit")
User avatar
Scottdru
Frequent Poster
Posts: 1160
Joined: Tue Dec 17, 2002 12:00 am

Re: Project: monitor controller

Postby Hugh Robjohns » Thu Apr 28, 2005 1:18 pm

As promised, piccies of my bespoke monitoring control unit can be found here

They are quite low res just to save on web space and upload time, and I've just realised that the jpeging has made the labelling unreadable. So the controls from left to right are:

top row: balance (blue cap with a calibrated centre detent), source select (black lever key selecting two replay machines and desk out), cut A/cut B (maroon lever key), mono both/mono A (maroon lever key), Dim/phase (white key), talkback mic. With all switches in the centre positions the system monitors 'desk out' in normal stereo.

The mono both and phase switch positions are momentary (non-latching) to prevent accidents (!) but all other positions are latching.

Bottom row: Volume control (green cap with a calibrated centre detent), red and green cue light buttons (red and green lever keys), talkback switch (yellow), and talkback mic level (yellow cap).

Both cue and talkback switches all operate associated colour-coordinated LEDs when active, and the switches provide momentary actions if pulled down, and latching actions when pushed up.

The empty panel space to the right was intended to house remote controls principally for my Genex recorder, but I have yet to get around to that...

The rear panel shows the XLR I/O connectors. From left is an IEC mains inlet, monitoring outputs (on latching Neutrik balanced quarter inch sockets), three pairs of line level electronically balanced inputs (desk and two recorder outputs), a metering feed (see below), cue light (DC) control output and line level talkback mic output.

The metering output is a buffered balanced output taken after the source selection switch, and wired via a 5-pin XLR, which I hook up to the analogue inputs of a DK-Audio MDS600 meter.

The various internal shots show the PSU and a couple of circuit cards, with all the wiring. The best part of this unit is the use of the lever key switches, which are fantastic to use but sadly no longer available

Hugh
User avatar
Hugh Robjohns
Moderator
Posts: 16761
Joined: Thu Jul 24, 2003 11:00 pm
Location: Worcestershire, UK

Technical Editor, Sound On Sound


Re: Project: monitor controller

Postby Frank Eleveld » Thu Apr 28, 2005 1:49 pm

Hugh Robjohns wrote:As promised, piccies of my bespoke monitoring control unit can be found here.

Many thanks for that Hugh!
User avatar
Frank Eleveld
Moderator
Posts: 1167
Joined: Sun Aug 29, 2004 11:00 pm
Location: NL

 


Re: Project: monitor controller

Postby tomas » Sat Apr 30, 2005 4:06 pm

What about using opto-electronics for the attenuator? I am thinking about taking a simple mono potentiometer (but with a logarithmic curve) to control a voltage going to several (4 for balanced stereo) opto-couplers having plain photocells on the output. Something like this should fit the bill. If calibration is required, this could be done by placing multi-turn trimpots in series with each LED, and trimming each one to provide equal signal coming through the photocells.

I think it could be cost effective, and provide a good solution from the audio purity point of view, as long as the opto-couplers all have the same linear response.

As an added bonus, adding remote control would be no probs at all, since the potentiometer is controlling a DC voltage.

Cheers,
Tomas
User avatar
tomas
Regular
Posts: 152
Joined: Tue Feb 04, 2003 12:00 am
Location: Luxembourg

cheers,

t-:


Re: Project: monitor controller

Postby Frank Eleveld » Sat Apr 30, 2005 6:01 pm

tomas wrote:What about using opto-electronics for the attenuator?

Tomas,

Many thanks for your suggestion.

Perhaps some people may want to implement the option you're suggesting, but I feel it'd make the project considerably more complicated - my original intention was to outline a simple DIY project that can be built by skilled and starting DIY-ers alike, and adding sophisticated features like using opto-electronic controlled attenuation etc. would bring this project to a level where the average DIY-er interested in building wouldn't consider it feasible anymore. Of course, it's always possible to add additional features to the basic building blocks, so if one would be interested in adding opto-electronic driven attentuation, would you be prepared to explain the workings of a simple opto-electronic circuit in greater detail here?

Cheers,
Frank
User avatar
Frank Eleveld
Moderator
Posts: 1167
Joined: Sun Aug 29, 2004 11:00 pm
Location: NL

 


Re: Project: monitor controller

Postby Frank Eleveld » Sat Apr 30, 2005 7:13 pm

I've ordered two Tangent 'TREAD' PSU kits of which I'm planning to use one as a benchtop PSU, the other one would be powering the active electronics in the monitor controller project. Given Tangent's usual fast shipping (by US Airmail), I assume I'll receive the kits before the end of next week.

I'll post my findings about kit assembly and performance (mind I've only limited testing equipment available, so for in-depth detail on performance I'll have to rely on Tangent's words) some day soon.
User avatar
Frank Eleveld
Moderator
Posts: 1167
Joined: Sun Aug 29, 2004 11:00 pm
Location: NL

 


Next

Who is online

Users browsing this forum: No registered users and 1 guest