Twitter   Patreon WIGGLING 'LITE' IN GUEST MODE

 Goto page 123456789101112131415161718 Previous  1, 2, 3 ... 5, 6, 7 ... 16, 17, 18  Next [all]

Are you:
Interested in some panels?  87%  [ 124 ]
Or not?  4%  [ 6 ]
Too vague?  7%  [ 11 ]
Total Votes : 141

Author [IC] 4U Single Function 'Sergesque' panels
LetterBeacon
 I think this is a fantastic idea, and so today I decided to lay out a panel for some of my already built modules in this new format that the Bad Producer is proposing. I used the measurement guide he posted earlier in the thread to design the panel. The panel is 9UMW (242.9mm) wide and the spacing between each horizontal element is 13.25mm. However, you can see that the left-hand edge of the VCO module is much further to the edge of the panel than the right-hand edge of the Wave Multipliers, and I can't work out why. Is there something off with the measurements, or have I aligned something incorrectly? What do you guys think...?
 Hi Letterbeacon, nice panel! I would first say, that although the actual measurement 'should be' 242.9, I round it down to .5 (or in reality probably 242mm) for 'tolerance' - this approach to 'HP' I copied from Doepfer's specifications on euro HP widths. Another thing, why not make that into 4 panels? One for each module!? What I do for 1, 2, 3, 4 UMW is put a centre line on the panel and then space the columns in a 1" spacing from that, this will create an even and equal spacing between the columns of knob/sockets on the 'grid', it will also leave a slightly larger space at the edges of the panel ( 0.0625"/2 extra per UMW panel width). I like this as it gives a subtle 'break' to the grid which delineates the separate modules in the panel or case, it is actually a result of the division of a 17" panel into 16 UMW or columns. On larger panels, like this one, I split the panel into several smaller modules (here, 3UMW and 3 x 2UMW sections). In this case I would design each panel seperately and add them together to get the combined larger panel width - I often have each module designed separately anyway, so this is easier. This seems to space the columns out nicely, each 'module' has it's border and the knobs/sockets spaced correctly, and between each section is a nice even gap. What do you use to draw on? I use inkscape, if you want and it would be useful I could send you a svg, it might make more sense! If you want PM me your email address (But it should be noted that I do all my panels in inches! ) You can of course place your columns evenly but not on that plan, nor the 1" grid, in which case you would simply divide the panel width by the number of columns, here that would be (@242.9mm wide) 26.99mm*, well 27mm, or between each horizontal element 13.49mm... Starting from a centre line might stop the problem of the columns 'drifting' and ending up not even at the right hand side? *That's why I design in inches - it is all in 1/16"'s ...! Also, aplogies for the digression and waffling, got a bit of panel-design-cabin-fever!
LetterBeacon
Thanks for the reply -very informative! And thanks for the compliments on the panel.

The reason this is a collection of modules on one panel is because the ARP VCO is a very big PCB, plus it has a stripboard to bring the levels from 0-5v to 10vpp. This means I have to lie it horizontally parallel behind the panel which is why I've got the DC mixer next to it. The mixer's PCBs are tiny to the VCO can use up some of the mixer's panel space. I intend to do the rest of my modules more in the style of your idea, and have them as separate panels.

 Quote: On larger panels, like this one, I split the panel into several smaller modules (here, 3UMW and 3 x 2UMW sections). In this case I would design each panel seperately and add them together to get the combined larger panel width - I often have each module designed separately anyway, so this is easier. This seems to space the columns out nicely, each 'module' has it's border and the knobs/sockets spaced correctly, and between each section is a nice even gap.

In fact this is how I did it with mine -I designed each module in a separate file and then pasted them all into a new file which was the size of the large panel. I positioned each module according to the top left most element in the module.

I've been using a mixture between imperial and metric to work out my sizings. Because of this I may have gotten slightly confused and made a mistake somewhere along the line!

 Quote: What I do for 1, 2, 3, 4 UMW is put a centre line on the panel and then space the columns in a 1" spacing from that, this will create an even and equal spacing between the columns of knob/sockets on the 'grid', it will also leave a slightly larger space at the edges of the panel ( 0.0625"/2 extra per UMW panel width). I like this as it gives a subtle 'break' to the grid which delineates the separate modules in the panel or case, it is actually a result of the division of a 17" panel into 16 UMW or columns.

This sounds like a good idea -I think I'm going to try this. But will the knobs and jacks line up vertically if you have another module above it in the case? I'm annoyingly anal when it comes to stuff like this!

I've been using Illustrator to design the panel, but I think it does take .svg files. I'll send you a PM!

Don't apologise for waffling -it's all good stuff!
 I think with BIG PCB's I would do exactly the same, tho I do have a tendency to start going totally off grid! With the method I describe, you will have slight discrepancies vertically across your rack - I think i would like this, I get the impression you wouldn't! Well, I guess you would if you always placed a 2UMW panel above a 2UMW etc... I actually found that most of anything I make is either 2 or 3 wide.
LetterBeacon
 I had another go at spacing the borders as I realised they weren't quite the same on each module. Now the border on each module is exactly 1/8" from the edge, meaning that they're all evenly spaced now. This panel has been made slightly smaller to take into account tolerances, which means that there's slightly less space on the right-hand side, but I don't think that's going to bother me... too much!
LetterBeacon
 I'm thinking about the mounting holes, and something's just occurred to me... I'm using Schroff rails that I've previously used in my Euro DIY. The threaded inserts have 5.08mm between each threaded hole (as per the Euro standard). But because the Serge panel width isn't the same as the Euro HP, aren't the holes going to be off from the second panel onwards? I think Bad Producer said he was using Vector rails in his build, but doesn't that also follow the Euro standard of 5.08mm?
lordofthebored
This would be a problem when using tapped strips like those found in your Schroff rails, but won't matter if using sliding nuts. With nuts, they can be positioned wherever they are needed. Personally, I find these rails to be an unjustified expense, and I'll probably just screw mine into a wooden frame.

 LetterBeacon wrote: I'm thinking about the mounting holes, and something's just occurred to me... I'm using Schroff rails that I've previously used in my Euro DIY. The threaded inserts have 5.08mm between each threaded hole (as per the Euro standard). But because the Serge panel width isn't the same as the Euro HP, aren't the holes going to be off from the second panel onwards? I think Bad Producer said he was using Vector rails in his build, but doesn't that also follow the Euro standard of 5.08mm?
LetterBeacon
 Ah I see! I had no idea those things existed as I've always used threaded inserts! Thanks very much.
 I've started a page here: http://loudestwarning.tumblr.com/4Umodular With some more info and pics, including the rails in a boat thing It's a beginning as yet, but more to come - including panels hopefully!
Monobass
 Really keen to get Thonk involved in this format
LetterBeacon
 Great page, BP! I see you've countersunk the screws into your panel -how thick are the Bud boxes? I've been thinking of doing that to my boxes, but they're only 1.5mm and I worry they'll be too thin.
ElectroSteam
 I would suggest you add a few changes and recommendations to the standard, in addition to what you have already put in there: *) Mounting holes. You *really* ought to think ahead here. Not everyone will be using those shallow Hammond boats, so you might consider how people could use alternatives, now you are at it. A pair of M3 machine screws in steel nuts will hold a lot of force and torque, yet wood screws fitting a hole for an M3 machine screw will either have to be really long, or won't hold much of anything. Upping the hole diameter to 4.4 mm for M4 screws would be much better for anyone contemplating fitting heavy modules, like PSUs (open chassis backs ensuring adequate cooling, of course ), in DIY wood cabinets, made from pine or similar low cost material. M4 screws also looks better on panels of this size, though that point of view is pretty subjective. By the way: The standard mounting hole for an M3 screw is 3.3 mm, as far as I recall. *) You might suggest that people making mounting rails on metric, could use mounting holes 2.7 cm apart, evenly spaced around the center. The total error across 8.5 holes to a spacing of 17/16", 0.1 mm, is negligible. *) Panel thickness. I'd suggest not having a standard here may lead to a mess, aesthetically speaking. I did a bit of informal testing. 3mm (or 0.125" - the MOTM standard - for our friends on the other side of the pond) of a good quality machinable Aluminum alloy seems plenty strong enough, even for a 1 UMW panel with lots of mounting holes. *) You ought to specify the details for the electric signals, as there appear to be some confusion on the details. Dotcom format seems self consistent, unlike Euro/Doepfer: Signal voltages (VCO, LFO, VCF etc.): 10Vpp (-5 to +5V). Trigger signal inputs: +1.5V threshold, triggers on rising edge. Digital outputs: 0 to +5V. CV: 1V/oct, going up to at least +5V, possibly higher. All inputs accept 20Vpp (-10 to +10V) without any odd side effects. Envelopes (input/output): 0 (off) to +5V (VCA fully open). I couldn't easily find what the standards are for CGS and Serge modules.
synthcube
 Monobass wrote: Really keen to get Thonk involved in this format

SynthCube too.... Great stuff!
J3RK
 I don't know if a full set of electrical specs can be established here. It's not only CGS designs going into these. A stiff guideline is probably a good idea, especially for things like signal level acceptance on inputs. The things I see being more difficult to set a rigid standard for are logic and envelope levels. There seem to be just as many modules out there that like 10V logic/envelope signals as there are 5V. (I prefer 10V myself) and then there are some like older Buchla designs that might even expect 15V. In this case, I personally would take the middle ground of 10V, as it would likely still work ok. If the inputs for logic are designed in a way (such as a forward comparator) with a lowish threshold, then really anything could be used within reason. As long as it's specified for each design, the builder can decide whether to implement as-is, or perhaps make modifications like added reference voltages, zeners, comparators, etc. to level out the system being built. I agree with many of the levels you have in mind, just not the logic ones. I would definitely agree that -5/+5 (10V P2P) is desirable for audio and bipolar modulation signals. It's definitely a good idea, but I think this format is going to act as a big funnel concentrating designs from too many different designers, cloners, existing PCB libraries, etc. I also see a big divide on power rail specs. Some designs have built in voltage references, some don't.
All good points Elektrosteam! I had expected the mounting hole question to come up as I hadn't 'really' nailed it down, I think some discussion is certainly necessary, on that and many other aspects of this format.

I believe there is some importance in having the possibility to use the vector type rails, as - evidenced by the preponderance and popularity of 'skiffable' euro modules - lots of folks want the capability to do that. We have at this moment for sure the opportunity to think ahead!

Interesting thoughts on centring the mounting holes like that - are you imagining there the use of a rail similar to the MOTM rails? I should do a test with 4.4mm mounting holes on a budbox with the rails, I guess it could be okay with a washer of some sort. FWIW I've used 8mm wood screws into pine (M3 ish size) and MDF/ply for years without any problems at all, and that's a heavily gigged system...

I've found 2mm aluminium fares very well (even at 1 UMW) in the bend test, I agree though about not setting panel thickness standards. In fact i do feel like the last person to set standards at all, beyond common sense! Your list of electronic signals look to be about what I have mine working on - either running at 12v or 15v - re CGS signals, the trigger/gates can all generally be set on the output buffers, 5v is standard I think. The trigger inputs on some operate at +2v, others a difference of 1.2v...

Overall I set these to work with the rest of my synth when I build, in an attempt to have some cohesion in it, I reckon that the specs are actually all over the place though!

 ElektroSteam wrote: I would suggest you add a few changes and recommendations to the standard, in addition to what you have already put in there: *) Mounting holes. You *really* ought to think ahead here. Not everyone will be using those shallow Hammond boats, so you might consider how people could use alternatives, now you are at it. A pair of M3 machine screws in steel nuts will hold a lot of force and torque, yet wood screws fitting a hole for an M3 machine screw will either have to be really long, or won't hold much of anything. Upping the hole diameter to 4.4 mm for M4 screws would be much better for anyone contemplating fitting heavy modules, like PSUs (open chassis backs ensuring adequate cooling, of course ), in DIY wood cabinets, made from pine or similar low cost material. M4 screws also looks better on panels of this size, though that point of view is pretty subjective. By the way: The standard mounting hole for an M3 screw is 3.3 mm, as far as I recall. *) You might suggest that people making mounting rails on metric, could use mounting holes 2.7 cm apart, evenly spaced around the center. The total error across 8.5 holes to a spacing of 17/16", 0.1 mm, is negligible. *) Panel thickness. I'd suggest not having a standard here may lead to a mess, aesthetically speaking. I did a bit of informal testing. 3mm (or 0.125" - the MOTM standard - for our friends on the other side of the pond) of a good quality machinable Aluminum alloy seems plenty strong enough, even for a 1 UMW panel with lots of mounting holes. *) You ought to specify the details for the electric signals, as there appear to be some confusion on the details. Dotcom format seems self consistent, unlike Euro/Doepfer: Signal voltages (VCO, LFO, VCF etc.): 10Vpp (-5 to +5V). Trigger signal inputs: +1.5V threshold, triggers on rising edge. Digital outputs: 0 to +5V. CV: 1V/oct, going up to at least +5V, possibly higher. All inputs accept 20Vpp (-10 to +10V) without any odd side effects. Envelopes (input/output): 0 (off) to +5V (VCA fully open). I couldn't easily find what the standards are for CGS and Serge modules.

 J3RK wrote: It's definitely a good idea, but I think this format is going to act as a big funnel concentrating designs from too many different designers, cloners, existing PCB libraries, etc.

Rather like eurorack then! I think the Serge triggers are generally similar, but there is that output from the SSG which is really hot!
 LetterBeacon wrote: Great page, BP! I see you've countersunk the screws into your panel -how thick are the Bud boxes? I've been thinking of doing that to my boxes, but they're only 1.5mm and I worry they'll be too thin.

Yes, I think it would be preferable to use panheads - which I did on the first - but these seem okay (it's 1.5mm too roughly) and I wanted to try it out so they could be racked without any spacing for the screw heads between the boxes. It does appear sturdy enough screwed nice and tight!
CLee
 ElektroSteam wrote: *) You ought to specify the details for the electric signals, as there appear to be some confusion on the details. Dotcom format seems self consistent, unlike Euro/Doepfer: Signal voltages (VCO, LFO, VCF etc.): 10Vpp (-5 to +5V). Trigger signal inputs: +1.5V threshold, triggers on rising edge. Digital outputs: 0 to +5V. CV: 1V/oct, going up to at least +5V, possibly higher. All inputs accept 20Vpp (-10 to +10V) without any odd side effects. Envelopes (input/output): 0 (off) to +5V (VCA fully open). I couldn't easily find what the standards are for CGS and Serge modules.

I think it's great that you all are creating a standard for a front panel system in 4U but I have doubts about setting electrical specs. For instance, no mention of 12 or 15 volt supply. When I release boards (designed for 12v) I specify changes needed for 15. Should LFOs be bipolar? Some are some aren't. The original DUSG isn't. I know of at least 1 CGS module that triggers on the falling edge. Though what you have for specs are generally the norm, keep in mind that these are modular systems. And that builders will end up getting boards/kits from lots of different folks.
J3RK
 Agreed. I've started using voltage references on some of my newer designs, but many of the older ones, and even a few new ones are typically designed with 15V supplies in mind, and later adapted to 12.
ElectroSteam
 J3RK wrote: The things I see being more difficult to set a rigid standard for are logic and envelope levels. There seem to be just as many modules out there that like 10V logic/envelope signals as there are 5V. (I prefer 10V myself) and then there are some like older Buchla designs that might even expect 15V. In this case, I personally would take the middle ground of 10V, as it would likely still work ok. If the inputs for logic are designed in a way (such as a forward comparator) with a lowish threshold, then really anything could be used within reason.

The things is, if you make envelopes 0-10V, then a standard LFO cannot immediately be used for driving a VCA to full output, at least not without amplification.

Additionally, according to Mr. Doepfer, then Euro is apparently supposed to use 0-8V envelopes, with LFOs being 5Vpp (-2.5 to 2.5)...

So even if you DC level shift a Doepfer LFO in a mixer, the resulting signal can either not fully turn on -or- it cannot shut off an Euro VCA...

Not to mention ... Maths. Is that an EG or a LFO? Etc.

I would suggest not reinventing the wheel, where it isn't needed. From an engineering standpoint Dotcom's electric format does make sense to me as a suggested standard to aim for, if you truly wish to be able to mix and match signals, regardless of type. All you need is attenuators, mixers, a few DC voltages and maybe inverters (attenuverters....).

Not sure you gain anything by using 10V envelopes and digital signals. The reduction in (S+N)/N ratio doesn't seems troublesome. Additionally, if you have the odd module generating a 10V envelope, then it is easy to attenuate it before wiring it to the rest of your modules.

(Snip.)
 J3RK wrote: I also see a big divide on power rail specs. Some designs have built in voltage references, some don't.

I had some thought on that as well, but wanted to hear people's reactions to the signal issue.

If you recommend +/-12V rails *and* decide on 10V envelopes, then you miss a golden opportunity IMHO.

If you had decided on 15V rails, then you'd frequently have a neat solution to the problem with designers using the rails as voltage references, or when a design doesn't have a good PSRR. If the design is intended for 12V 'reference rails' (no internal reference), then you can easily add a pair of voltage regulators to the 15V rails, giving you locally regulated - and very low noise - 12V rails. Run the module off this, and you have a nice, stable module without having to modify anything.

Similarly, a module with the same problem, built for 15V, would frequently only need a few resistors swapped to be able to enjoy the same advantage: Locally regulated, low noise 12V rails. No redesign needed, nor a new PCB.

So with 15V rails you can enjoy *both* all the 5U modules, the world of Euro Frankenstein'ed into 4U, the discrete world of Bob Moog (on +12/-6V rails). Plus everything in between. All your kits and PCBs are belong to us!

Going with 12V rails plus 10V envelopes would generally throw this advantage out of the window. There would not be enough headroom to account for both the opamps and the local voltage regulator to allow you to generate a 10V signal.

I'm a bit surprised the habit of using local rail regulators from 15 down to 12V on an individual module isn't used more. You don't even need an expensive LDO precision chip, as you can make a half decent linear 'post regulator' from about a handful of discrete - and dirt cheap - components per rail.

Note on M4 mounting screws: I realized you cannot use decorative flat washers under the head of an M4 machine screw on your panels. A standard M4 washer is 9 mm in diameter, so would have to either be offset from center, or go over the edge.

Another note: Might be worth having a suggested 'clear height' on the back of the panel, for instance 1 cm in from the top and bottom edges, to allow room for the chassis/boat mounting rails.

End of soap box for now.
ElectroSteam
Yes, I mixed the identities of you and J3RK. Sorry...

 the bad producer wrote: (Snip.) Interesting thoughts on centring the mounting holes like that - are you imagining there the use of a rail similar to the MOTM rails?

Yes, in addition to using those metal insert thingamabobs with a threaded hole mounted along the edge of a wooden chassis. The M3 type is a bit on the small side IMO.

 the bad producer wrote: I've found 2mm aluminium fares very well (even at 1 UMW) in the bend test, I agree though about not setting panel thickness standards.

Didn't have any 2 mm plate to test with, but 1 UMW in 1.5 mm was definitely on the thin side IMO.

I was actually suggesting that having a recommended panel thickness is a good thing. A mix of, say, 2 and 3mm thick panels would look very odd IMO. 8_)
NV
 I've been contemplating working in a 4U banana DIY format for awhile with a number of similar ideas. There are a few things I think warrant some careful thought before solidifying standards. As an important point that I'm basing all of my thoughts off, I think it's crucial that a new DIY format be as builder-friendly as possible. This means allowing the format to be affordably constructed, built/mounted with hardware that can be easily accessed in a variety of locations and without effort or tooling that may scare a less-experienced or time-permitting DIYer away, and mounted/presented in a variety of ways to suit the individual (varied widths of cases, 19" racks, etc). Standards are necessary and ultimately beneficial, but it's important that the standards not be too stifling and inhibit the growth of the format. ======================================= Module Sizes and Mounting The idea of incorporating Eurorack mounting is a solid one, since the rails used for Eurorack have become very common and affordable and allow for mounting in a variety of enclosures as well as 19" racks. However, I think there are some compatibility issues with the proposed standard of 1.0625" widths. As of now I count 6 manufacturers of Eurorack rails, each providing their own rail lengths and mounting sizes. Here's a listing of the various rails and the sizes they accommodate: SRS (Elby) - 84HP, 168HP - M5 nuts or tapped M2.5 strips Rittal - 84HP - M6 nuts or M2.5 tapped strips Schroff - 84HP - M2.5 tapped strips TipTop - 84HP, 104HP, 126HP, 168HP - M3 tapped strips Gie-Tec (Doepfer) - 84HP - M3 tapped strips Vector (Erthenvar, Goike) - 42HP, 84HP, 104HP, 150HP, 300HP - M2.5 nuts or M2.5 tapped strips Of the above 6 rail distributors, 3 accommodate sliding nuts but all 6 accommodate rails tapped to 1/5"-5.08mm increments. In those sliding nut rails the sizing varies from M2.5 up to M6 (clearance variation of 3.85mm), but the tapped strips vary only between M2.5 and M3 (clearance variation of 0.55mm). When it comes to the standard length of the rails, none accommodate a module width of 1.0625" without hand-cutting the rails or inserting blank spacing panels but all accommodate widths in 1/5"-5.08mm increments without modification. Additionally when using tapped strips as most of the rails support, a standard of 1.0625" would require that each mounting hole be about 8mm wide to accommodate the variation that would be encountered as the modules fluctuated to and from the 1HP increment. In short, these rails are all built around the HP standard of 1/5"-5.08mm increments so it seems problematic to force another sizing standard inside of an existing standard, with neither of them actually matching the other without restricting the rails to a particular brand and placing responsibility upon the builder to cut the rails to a custom size. Module Installation I understand the idea of creating the format around easily installing them into Bud/Hammond boxes, but I think restricting the format under that notion presents problems in both customization and accessibility. Many people prefer to have their modules mounted into standard 19" racks alongside their studio equipment. Additionally many people prefer to have their systems spread out rather wide - the "show your system" threads in all the formats reveal that the widths of systems vary as much as any other factor. Miniature systems also have benefits for portability, control, or a particular system focus (noiseboxes for example), in which case something like a 42HP implementation would be appealing. Eurorack rails come in a large variety of sizes to accommodate those elements. There are multiple 84HP 19" rackable options with pre-built cases you can purchase (all rail manufacturers offer this option), and for customized systems there are pre-fabbed rails available from a variety of distributors from lengths as small as 42HP (8.4") to as wide as 300HP (60"). I feel this level of potential customization would greatly appeal to any DIYer, whereas restricting it to a concept of 17" boxed systems may turn people off from the format as they would be limited in the ways they could present their system. ======================================= I think the standard of 1.0625" presents some aesthetic benefits but at the cost of a number of technical and practical complications. Since the format would be utilizing mounting that is based upon the HP standard of 1/5"-5.08mm increments, I think it would be beneficial to embrace the HP standard and build from there rather than push a differing standard to work within one that is already established. My overarching point is that I think the concept of module sizing and mounting needs to account for the variety of users that would potentially pursue the format and the variety of ways they may prefer to house and present their system. Using euro rails allows for a great level of customization by cutting the rails by hand but I think pressing that responsibility into the builder as a requirement for entering into the format may turn a number of people away, since rather than being something you can buy and throw together with some screws into an existing rack it becomes something you need to machine yourself before even stepping into the format. The popularity of Eurorack took off when a variety of cases started to become both affordable and available and things became much more user-friendly - embracing that from a DIY standpoint will likely prove equally beneficial, and the more users the more appealing it would be to design for the format.
CJ Miller
 Some interesting perspective, NV. What you seem to be saying, in essence, is that for 4U to be DIY-friendly, it needs to use Eurocard rails and widths. IMO, mixing inches and millimeters in the same simple standard gets unnecessarily complicated. I do my Euro work in mm, my 4U work in inches - and I definitely prefer to keep them distinct. But, a person can still use metric mounting rails anyway. Just use sliding nuts instead of tapped strips, and the problem is solved. Same goes for Vector T-struts. The size of 1.0625" comes from fitting 16 modules in a 17" boat - but this in no way limits people to using boats. Anybody can still mount as many modules together as they like.
I think the preamble I gave in the OP set me up at the beginning to be paddling against the stream somewhat, as demonstrated in yours, and other, recent interesting posts!

To be honest I did re-ignite the '4U format' discussion - which we've had here many times over the years - in an effort to 'nail it down' and have a general consensus, I think we may now be moving towards that:

I now feel it boils down to neither re-inventing the wheel, nor creating an unusable and future-fragile (ie unthought through) format which in the beginning might appear suitable and/or ideal, but with retrospect suffers from a lack of insight into what may happen - who may use it, design for it, build it etc - in the future. There are lessons to be learnt everywhere, one that springs to mind is the 'red-stripe' imbroglio of euro!

In view of that, it does indeed make sense to become more involved in the actual hardware of the euro system ie the use and total adaptation of panel dimensions to an HP subdivision enabling the use of - as NV points out - a whole plethora of racking options. Given that one particular application that was/is dear to my own heart (the racking of a bud box) is actually; 1) possible and 2) not disallowed in your suggestion (multiple HP widths based around 1/5") NV, I get the impression that we may be getting to the point where this becomes the system to which common sense adheres us:

* As suggested on many occasions, by many people: 1" widths (HP's)

 NV wrote: ...Standards are necessary and ultimately beneficial, but it's important that the standards not be too stifling and inhibit the growth of the format....

This is an important statement, for me the absolute emphasis for a usable format is a mechanical aspect that enables as much as possible within it's framework. It should allow questions of module construction to be limited to "what colour LED's shall I put on my Burst Generator?" not "Where the hell do I drill a mounting hole?". I would include questions like "12v or 15v" in the first type of question. I'm certainly the last person to discuss electrical specifications with as I have boxes and modules in my working system that range from 5V DC to 230V AC, supply AND control...

I'm going to go and think with my hands, do some sketches, rethink, regroup!

One thing I just thought though, I think 5HP works better than 6HP ( ie 1" min width) certainly for budboxes, as this would be 17 'columns' on a Serge grid, useful for adding that extra row on a 16-step programmer...
kons
 Some really good points raised here... Just want to point out though, 84hp is the 'standard' euro rail that fits in a 19" instrument rack. If a 5hp Euro is equivalent to 1" ...