|
|
Post by young Protoboard on Sept 8, 2020 22:50:42 GMT
Tl;dr: I can't see and I hate patchwires so it's banana time. Also voltage ranges are relative and if you believe hard enough you can accomplish anything. Hey, all, Single supply/single rail environment is a LIE. Pull off about a quarter of all TW modules and you'll see a voltage divider creating a 2.5V rail. Go on, I'll wait. In each of these circuits, 2.5V can be thought of as GND, and the two power rails (0,5)V become +/-2.5V. This leads to noninverted behavior on only the 2.5-5V range. So, every time you DIY in AEM format, you must accept this duality and how it affects the circuit. A good example may be the CV summation module, which I do not own but by the description I imagine to be a summing amplifier circuit centered at 2.5V; rather than referencing 0V ground, the ground is "faked" to 2.5V. Such bipolar behavior characterizes the module. I'll skip a bunch of hand-waving explanation as to why negative voltages are but an illusion; if you're a circuits beginner, reverse-engineer and puzzle over the MIXER4-4 circuit. The confusion that may ensue informs the issues of negative voltage. For now, let's just say that GND is relative and the voltage value may be placed wherever is most useful. There's nothing WRONG with this; as long as you're comfortable with positive behavior in only half of the 5V CV range of the format at the design bench, then all is well. For me, I find that I prefer accepting the reality of negative voltage, which you'll see soon. Now, let's talk about patch wires. Ever had to pull out and then replace the input A2 from the MIXER4-4 when A1 and A3 are also plugged in? When breadboarding circuits I find that I need tweezers to remove and replace components, and I can't shake the feeling that tweezers would make modifying patches on AEM much easier. I have zero interest in making music with tweezers. And I can't help but feel that I would play my synthesizer a LOT more (it's easier to play more than 0 minutes per week) if the patching and user interface experience were improved. Now, I'm blind, or very nearly so, so this may just be a "me" problem, and if it is, then I'm perfectly content to be mocked and shouted down. But that won't change the fact that the banana cable patching experience feels more smooth than a properly chilled gelato. These, and many other DIY-centered reasons (can't build a damn thing in THT components on a tiny daughterboard, especially when the whole motherboard is taken up by pots, switches, and LEDs; the pin header positioning eliminates about 25% of routing and layout space when using THT components; and so on), have led me to create the following. Meet my newest format's power module:   It's a thicc boi, to be sure. This was a bit of a hodgepodge of components and power sources; it runs off of two separate center-positive barrel jacks that accept 9-15V. This was necessary to create a +/- 5V power supply while using only UL-listed (and thus legal to plug into wall power in USA) power supplies that I had on-hand. Version 2.0 will be more streamlined. In this format, audio signals are centered at 0V and CV signals run the entire 10V gambit in the +/- 5V supply environment. Reach all oscillator octaves without a toggle switch! Also note the switch in the upper-left-hand corner; this is a power switch providing global on/off for the entire synth. robertlanger I'M CALLING YOU OUT! GIVE THE PEOPLE THE POWER SWITCH THEY NEED AND DESERVE. I sincerely hope my format treason doesn't bar me from participating here; on the short list of DIY projects is several AEM-format banana-to-patchwire conversion modules. Cheers! yPb |
|
|
|
Post by young Protoboard on Sept 8, 2020 22:53:45 GMT
Oh, I guess I didn't state it explicitly in the OP: this format is all banana cable, all the time, in sweet stackable goodness. Further, the footprint of this particular module is identical to MASTER. This new format is physically the same mounting dimensions as AEM, and an integrated front panel allows for DIY modules larger than my minimum design footprint in the free version of Eagle allows for.
Cheers x2,
yPb
|
|
|
|
Post by slowscape on Sept 9, 2020 2:22:13 GMT
If you were making a whole new format, why did you choose to back mount the modules? Now, I have pretty good vision, and skinny fingers, so patching is not an issue for me. But if you want to talk about a limiting design convention, I think it’s the back mounting of modules. It does not allow for varying depths of modules, requires very specific standoffs, and introduces the possibility for stacking error.
|
|
|
|
Post by young Protoboard on Sept 9, 2020 3:37:24 GMT
If you were making a whole new format, why did you choose to back mount the modules? Now, I have pretty good vision, and skinny fingers, so patching is not an issue for me. But if you want to talk about a limiting design convention, I think it’s the back mounting of modules. It does not allow for varying depths of modules, requires very specific standoffs, and introduces the possibility for stacking error. Well, I have quite the stock of AEM-specific components, for one, standoffs included. (15mm F-F, 8mm M-F, and 10mm M-F only take up two drawers in my parts organizer.) There's the ease of removing and rearranging modules, as well; sliding nut systems look to be the worst option possible for this. (A discrete rail mounting as sort of a hybrid between the two might be a good bet to eliminate nut-juggling.) Primarily, though, is that I can only design a 100x80mm board size in a single file in Eagle, so I'd be limited to an 80mm depth maximum for a rail-mounted module. Of course, I could standoff/daughterboard another PCB to gain extra layout space, but I'd be limited by the module's horizontal width, as well as the depth of the panel-mounted components. Trim off, say, 20mm for jumper-wired jacks and an irritatingly-mounted (with standoffs) board parallel to the front panel for pot mounting, and I'm left with maybe one daughterboard of diminished size. Perhaps the front panel is huge on some sequencer and allows for multiple stacked daughterboards-- but then I have possible stacking tolerance issues, as you mentioned. As I've trialed and discovered, 100x75mm is not as much space as one would prefer with THT components, especially with circuits of similar or higher complexity than the most complex TW modules. Now, if all I made was basic mixers and other single-IC circuits, perhaps I could shave off some depth to the modules, but that's such a small slice of what can be realized, and doesn't therefore yield much benefit, as the case needs to be deep enough to accommodate the deepest possible module (80mm), which is thicker than this example module with little benefit. I can't help but observe that AEM modules don't vary in depth and the format gets along just fine, and so do many Eurorack skiff cases. (They also require very specific standoffs and need increased precision in manufacturing to combat stacking error.) Robert has the advantage of double-sided boards to utilize space as maximally as possible, whereas I can currently only produce single-sided PCBs. My current working solution is to increase the real estate dedicated to main-brain circuitry, which means a second PCB. While cumbersome (and not detailed in this module as it has no potentiometers), the topmost PCB would mount pots and the second electrically connects jacks. This extra space afforded by these parallel-to-front panels also allows for basic I/O processing, buffering, etc. to take place off the "main" two boards and instead on these "interface component" boards, further freeing up room and eliminating nasty trace crossings (or spaghettification on Protoboard) endemic to the single-layer AEM style. None of this would take place on a perpendicular-style mounting, and I'd be stuck with a graph theory problem every time I try to lay out a board, just like DIYing in AEM. I'd say, all in all, it boils down to the fact that this pseudo-skiff build best matches up with my production capabilities at this time; the solutions it provides to my difficulties with AEM are satisfactory, and what little added complexity apparent to the format are issues that I already battle in AEM. But mostly, I have six months+ of meager student savings in bulk AEM components and this is the best overall solution I've found so far that utilizes the resources available to me. This isn't a format for mass production, and I don't doubt that as my design and construction horizons expand, it'll undergo radical change. But for now, I'm just happy to be able to DIY a circuit that has a chance of fitting in a reasonable case footprint. I actually purchased 10mm standoffs in anticipation of creating two-board modules and a "double-deep" case. It would have been an easy matter to toss in a few standoffs onto every TW module to make everything flush. But getting away from the 1U "standard size," accepting the limitations of THT components, and capitalizing (via new patching hardware) on the increased front panel space caused by ever-larger PCBs for complex circuits, solves the problems I encountered with AEM DIY, and then solves some more. You might think the cost in patching components would be a deal-breaker for budget-minded me, but at a guestimated $3 per module (I have a spreadsheet buried in my computer somewhere) for the correct height and quality of standoffs (actually combination of Preci-dip and 3M standoffs) makes the banana jack edge out on small- and medium-sized modules. After that, it's a matter of patch cable price, which is where AEM wins wholeheartedly. Fortunately, they don't break often. Any suggestions to make manufacturing/design/the whole format easier to work on gets a big thank-you from me. I really appreciate the question; it made me rehash and reevaluate my design decisions. |
|
|
|
Post by young Protoboard on Sept 9, 2020 3:43:59 GMT
Now, there's one other thing that beats the pseudo-skiff-- soldering the electrical connections to those banana jacks was a pain. A perpendicular PCB style of module would make for easier air-wiring of those banana sockets. (It was a crazy reach to get the soldering iron on that green jack.) But now that I've built one module with this strange pin connection, I can say without a doubt I'll be running an air wire from socket down to a spot on the board near the edge, rather than directly underneath the jack itself. The pin connection was an interesting IDEA, but a very flawed and difficult execution.
|
|
|
|
Post by spacedog on Sept 9, 2020 13:39:31 GMT
That's some dense reading up above and I'm a simple soul at heart. Would I be correct in summarising as follows...? I like the concept and (some of the) designs, but I really can't work with the physical format.Hence, you're seeking to repackage existing designs into a more easily patchable format...? In the process, you're looking to rectify what you see as some fundamental design flaws (e.g. the power rail methodology)...? =================================================================================== OK, this next bit may be a hijack of the thread and may need to be moved, it came out of the thoughts provoked by your initial post... I had been considering the AE Modular physical format, especially in the context of third-party modules. When I say "considering", for me this is by way of running thoughts through my head about why I don't read much about the format, when many other formats are listed in various articles (I don't). Also, about how some basic design elements may be slowing down both overall mass acceptance and the arrival of third-party modules. Rather than write reams, I will just say that I think this may just be an interesting topic and could even generate ideas to kick this format up another level in the eyes of dedicated users of other formats. If it's got legs, it the topic can be moved. I'll say a couple of things though... The first part of which is the label often given to the AE Modular. I am not a fan of calling the AE Modular format a "cheap alternative to Eurorack". I say that for two reasons: - It's not an alternative to Eurorack insofar that you can't do everything that you can do in Eurorack. It's unlikely that this will ever be the case as the diversity of modules in Eurorack is going to be hard to beat, plus it's had quite the head start; and
- I don't like the adjective "cheap". AE Modular is definitely "value for money", I, and many others here, have put together something that meets their needs in terms of musical outcome, or sonic exploration, for a considerably smaller financial outlay than the equivalent Eurorack system would have cost.
So, I believe we have a value for money modular format that provides a good stand-alone system capable of allowing both musicians with an outcome in mind and sonic explorers of all abilities to achieve their desired outcomes - and an outcome may just be relaxing away from a hectic world for a few hours, making bleeps and bloops that no one else will ever here. Perfectly acceptable  On the other hand, the format doesn't necessarily draw in anyone who already has Eurorack equipment (remember the head start); yes, it can be interfaced, although it's not a simple cross-patching exercise, rather it's via 4I/O or the MASTER modules, and I am not sure the Eurorack cases really give good payback for the amount of HP they require (this is my opinion, not fact). Of course, someone designing a modular system from scratch can be very well served by the AE Modular, although if they want to branch out into something more esoteric, they may well have to move to Eurorack to achieve that, at which time it's a decision on whether to sell their AE equipment. Also, looking from the other side, module designers already bemoan the space they have in Eurorack, and many can't fit their designs into the less-deep cases. There is no hope in getting their precious designs into the AE format. We can hold out for ports of more software-based modules, possibly going into wide modules, but the depth is fixed. That can be quite a constraint. As a personal example of all of the above, I bought my AE equipment to give me "analogue" sounds that could be manipulated in real-time to generate ambient washes, incorporating any sequencing that was available and stable. I have been quite content to increase my use as the system has evolved - for me, it's one of a number of musical "tools". I have little desire to replace what I can achieve in AE Modular with the same in Eurorack, especially now that some of the really good utility modules are appearing. It's not an alternative to something, it's just an excellent piece of equipment. I have also been interested in setting up something small around sample manipulation/mangling for some ideas that I have and have implemented via a VST arrangement in Reaper, with lots of things mapped out to controller knobs - but I'd love to do it in hardware. Currently, that's probably a Make Noise Morphogene as a starting point - a module that alone costs almost as much as my entire three-row AE Modular has cost me, and I would need a lot of additional modules to make my ideas reality. I'd easily be looking at something 4-5 times the cost of my entire AE system, for a very basic, and dedicated, setup. Again, the AE Modular is not an alternative, it simply can't do this, and maybe it's always going to be beyond the format...? My point...? I accept the format for what it is, and with its limitations. There could still be a few improvements, but that's a very personal opinion based on my needs. I certainly wouldn't have wanted to spend what it would have cost me in Eurorack for an equivalent outcome, and again I accept any shortcomings as the value for money is astounding. Finally, I can't see third-party versions of some of the modules that interest me, or other more esoteric modules, ever becoming available and I accept that. A few things to chew on there and, as I said, maybe best done somewhere else. Sometimes it's good to shake up the groupthink a bit |
|
|
|
Post by young Protoboard on Sept 9, 2020 17:45:16 GMT
That's some dense reading up above and I'm a simple soul at heart. Would I be correct in summarising as follows...? I like the concept and (some of the) designs, but I really can't work with the physical format.Hence, you're seeking to repackage existing designs into a more easily patchable format...? In the process, you're looking to rectify what you see as some fundamental design flaws (e.g. the power rail methodology)...? That's a pretty good summary. I'll provide another (hopefully short) cheat sheet: As a synth player, I enjoy the concept of a small, simple, building-block format, but I really can't work with the physical format. As a module designer, I appreciate the concept of a "DIY-friendly" environment with simple, safe voltage ranges. But experience has taught me that the inherent limitations of AEM works against that. No no, this is perfect. I may have written novels, but what's already been posted was just the tip of the iceberg. I agree in that AEM is not an alternative to Eurorack. My limited understanding of modular synthesis history in regards to format is that Moog and Buchla kicked off with LARGE, EXPENSIVE modules. Then comes Serge, Eurorack, Modcan, and so forth, all getting smaller, cheaper, and more accessible. AEM, young, small, and inexpensive, is the polar opposite to Buchla, if you will accept the hand-waving generalizations. Buchla modules are complex, expensive, and powerful. There is no AEM module that can rival a single Buchla module; it is the nature of the formats. Thus, every attempt to make modular synthesis more accessible (cheaper, smaller) is, to some extent, a compromise in functionality. As you have said, Eurorack is shrinking to be too small for some designers. Well, imagine those same designers staring at AEM. They wouldn't be pleased. I contend that AEM's uniqueness and strengths prevent it from ever reaching the complexity and functionality of Eurorack. AEM is small in thickness and panel size, lightweight, and compact. Yet each of these design considerations limit the available board space for each module's circuitry. It is simply not feasible to port, say, a stepped-sloped-generator or Clouds/Tides/Typhoon in the cutesy AEM footprint; such a complex circuit NEEDS more real estate, and thus such a module would be large in U and more difficult to understand and use by nature of the placement of I/O headers and style of control layout of AEM. This problem is compounded when the DIYer is restricted to through-hole components; AEM uses just about the smallest SMD I've seen, and man oh man is any TW module with more than one IC tight-packed. (Mad props to Robert for his layout wizardry!) The designer of the Wonkystuff mm33, namke , bemoans this THT limitation, just as I have done so to the endless void at my soldering bench during my months of AEM DIY. Simply put, outside of the smallest, simplest circuits, AEM in my opinion is not DIY-friendly; any amount of complexity is immediately paid through the nose in rack space as the module grows in U. There exists also a notion that the "building-block" nature of the simple TW modules can be a benefit to the synth player, as it "increases flexibility" by 'simplifying and breaking apart,' if you will, some larger and more complex Eurorack modules. An imaginary example (I'm not well-versed in Eurorack offerings) is a Eurorack complex oscillator with attached wavefolder. AEM provides, say, 2OSC/d and WAVEFOLDER, which can, by this Lego building-block notion, emulate such a Eurorack module. Given the previously stated limitations of AEM module complexity, it's unlikely such a combination will ever rival this imaginary Eurorack module, let alone exceed its functionality. It is true that one can technically achieve the same absolute basic functions of such a Eurorack module, but it is no substitute, and thus no alternative; rather, a compromise. Lugia has, with Gargantua, demonstrated excellently another issue of the Lego notion-- approaching higher complexity means increasing rack space TREMENDOUSLY. A Buchla 258 (if my memory serves, this is what Lugia is trying to emulate two of) is far smaller in rack space than Gargantua, and while more expensive, is far more functional than any AEM facsimile could be. These "tiny Lego" blocks aren't so tiny if you need a dozen or so to emulate a module from another format. Sure, they may be repatched in ways other than that emulated module, but the fact still remains-- if the musician wants to use their AEM-facsimile Wogglebug, they'll still need to use a whole row or so of modules to do so, dedicating a large chunk of their instrument to one "module." This, simply put, doesn't save space. AEM is the "smaller" format only if one examines formats through the lens of module size and module count; regarding features and utility, Eurorack, by cursory examination alone, wins out. I tend to agree with all you've stated here, although I must give a bit of a chuckle at the quotes around "analogue." I'm saddened by the number of modules controlled by digital microcontrollers, but this is again a consequence of the restrictively tiny circuitry space in AEM; there's simply not a good way to fit many true analog circuits into such a small package. AEM is inexpensive; undoubtedly so. It's sticker price was the singular reason that I entered the wonderful world of modular synthesis, and for that I am ever grateful to Robert and crew. However... I have a Eurorack friend who is filling a standard 6Ux84hp (forgive my incorrect information; I'm not fluent in Eurorack) with modules that each cost under $150, and the vast majority are $100-$150. Six modules in, and their case is more powerful and fuller-featured than my similarly-priced (and larger-sized, not accounting for depth) three rows of AEM. A handful of their modules are pre-owned, sure, so it's not the strictest comparison. But, look at "acquiring a powerful modular synthesizer" and the race evens up considerably. And once they passed their sixth module (their rack is nearly complete now), my AEM system can never hope to compete in any parameter (size, price, looks, patching experience, features, availability). Let's look at some of the most complex and featured TW modules: WAVETABLES ($102), TOPOGRAF ($76), SOLINA ($79), SEQ16 ($87), ALGODRONE ($85), MULTIFX ($112). Certainly SOLINA (to my knowledge) is an emulation unique to AEM, and ALGODRONE is unique to AEM as well. And SEQ16 is an intriguing, useful, and unique design (at least, I haven't come across a similar CV sequencer in other formats). But as complexity and features increase (and the listed modules aren't by any means reaching the complexity of many Eurorack modules), so does price and rack space, and I predict a point in time where decently-featured AEM modules will be sold at similar price points to new Eurorack, which continue to get cheaper as time goes by. Accounting for the pre-owned Eurorack market, Eurorack seems to already be a better value per feature and rack space. So where does that leave AEM? I don't know, but it doesn't seem to fit my usecase as a person who can't afford AEM prices to begin with, and is left DIYing their own banana format. Simply put, I don't see any value in continuing with the AEM patchwire format, and my money would be better spent in components and design, or perhaps even switching to Eurorack and DIYing there. (3.5mm cables just feel a little too small and fiddly for me, though...) For all of the reasons above, I'm starting my new, easily-compatible-with-AEM banana cable format. Heck yes, fellow talkative modular friend! |
|
|
|
Post by young Protoboard on Sept 9, 2020 17:54:58 GMT
Hence, you're seeking to repackage existing designs into a more easily patchable format...? In the process, you're looking to rectify what you see as some fundamental design flaws (e.g. the power rail methodology)...? Ah, I forgot to address this. No, I'm not deliberately repackaging any AEM designs that aren't ubiquitous already. I see no need to; I get the most joy from designing modules from the circuit-up. In fact, I'm working on a universal slope generator that, if I can wizard it correctly, runs off of just one quad op-amp. Features include trigger-activated slope positive/negative generation, EOC trigger output, and slew limiting, as in any other format's universal slope generator base features. ("BUT ONE quad op-amp??? That's just one IC! Surely you could make that AEM compatible in 1U!" Lol nope, not enough real estate for DIY THT. Maybe in a 2U format, but there comes a point when trying to preserve the small footprint, etc. of AEM where one must simply choose the single feature of the desired module one is DIYing (eg trigger-activated slope generation), buy the closest module (2ENV), and put up with the lack of features. It's either compromise with cash, or make an ugly, sprawling, massive, spaghetti-infested PCB that takes up a huge chunk of your "travel-friendly" synthesizer.) Otherwise, yes; I'm looking to change the restrictive boundaries of the format I build in, power rails, circuit real estate, and so forth. The scope of this project, though, is small; I don't see myself producing these modules for sale or use for anyone other than myself. At most, I will build some adapter modules to mount in my AEM, and once my banana synth has it out-featured (I predict in about six modules), I'm likely to sell the AEM. All design and fundamental flaws aside, it's still a great instrument that I just... can't... use, because of my personal physical limitations. And that really sucks. |
|
|
|
Post by young Protoboard on Sept 9, 2020 18:32:59 GMT
Something that really helped enlighten my grievances with the AEM format is the following thought:
Flip the history of modular development on its head. Let's say AEM was the first ever modular format invented. And after many years, Eurorack hits the scene, with its larger, intuitively-spaced input and output 3.5mm jacks, larger toggle switches, and readable front panels whose design helps inform the function of the module. It has a larger CV range, and the complexity and features of each individual module surpasses AEM. It is more expensive but more powerful and easier to interface with as a musician.
Would Eurorack not then be revolutionary? Finally, a format for people who don't prefer identical grey rectangles and 0.1 inch connectors? You mean to tell me that I can patch between instruments with JUST the patch cables, because the mono 3.5mm cable carries and connects the two ground voltages? Look at all this space we have to design some really complex circuits!
This is all, of course, just from my perspective, as a musician who wants a fully-featured modular synthesizer; I'm not trying to physically emulate software, or old instruments, or toss my 48 square inch tiny instrument into my backpack to play somewhere in the park.
|
|
|
|
Post by rodney on Sept 10, 2020 11:33:10 GMT
Mighty thread you got going! bearheadwood converted his AE modular over to banana already for similar reasons. The Swedish take on the original Serge modules also looks nice. www.73-75.com/
I promised myself to get off screens and sleep but will read more thoroughly in the morning.
r.
|
|
|
|
Post by bearheadwood on Sept 10, 2020 11:53:02 GMT
Yeeep, slowly slowly chipping away at removing pots and LED's and replacing the switches with bigger ones. I'm just tapping the connections on the PCB and running wires to the parts that are attached to my phenomenally well-made panels. Nearly have the first third of my cases done, should be done by 2022 at this rate  My reasoning is more along the lines of really wanting a larger format system then any issues with the size of the AE, though.  |
|
|
|
Post by young Protoboard on Sept 10, 2020 16:19:55 GMT
bearheadwood This is fantastic! Have you documented this work elsewhere? I'm curious-- do you desolder and remove the AEM pots and remount them to a higher board? I seriously considered this kind of conversion for a while but didn't settle on an acceptable solution for the issue of the deep banana sockets on a shallow board. My closest working conceptual design was a 1:1 breakout module for each AEM module that carries every AEM patchpoint into banana, "permanently" patching (never removing the patchwires) the patchwire input and output rows to the breakout modules, and using the knobless AEM pots and switches on the AEM modules for user control. This would preserve the "individual" feel of each module in a way that an admittedly more practical build of just a big ol' banana patch bay linked to patchwire points loses in the conversion. Even so, the 1:1 breakout solution doubles the rack space of each module, and strange depth wizardry (or uninsulated banana jacks mounted in trickily-designed PCBs) to fit everything flush in an AEM case was needed, so it just wasn't the solution I was looking for at the time. Also thanks very much, Link Master rodney, always coming into the forum with pertinent references and vast knowledge, dropping URLs like they're California hillsides (read: on fire... again).
|
|
|
|
Post by bearheadwood on Sept 11, 2020 9:07:08 GMT
young Protoboard I have a few shots of panels, modules etc over on my instagram, but no real documentation to speak of. I'm just sort of doing it in dribs and drabs as time allows. The process is pretty simple, I'm using what I have on hand and aimed to spend the least amount of cash possible. I just replace the front panel of the AE with a banana one, and attatch the AE PCB to that. Basically I do this: - Plan out panel layouts on grid paper, glue them to the panels, I'm using 15cm height, and 3, 5, and 7cm widths because I want the bigness. - Panels are made of 3mm Acrylic, because I have that. Drill them next. Complain bitterly to myrself, because drilling panels is crap. - Remove faceplates from AE module and spacers/screws/allen screws. Put all the spacers and screws together into one chunky spacer for later. - Snip all the switches and LED's off the PCB, after marking the leg of the LED's closest to the edge of the board with a sharpie. - Snip the three legs off the front of the pots and remove them from the board by heating up the solder on the mounting legs and levering them out one side at a time. - Solder short wires from all the points you just removed to the pots and LED's, solder the backs of the pin headers to wires connected to the banana sockets. - Replace the switches with larger switches, soldering them with short wire as you did everything else. - Attach everything to the panel, switches and sockets will be fine with the supplied nuts, LED's fit in snugly if you drill the hole the right size, and pots you can mount however you choose. (I'm lazy and have put knobs over them and soldered them with wire. There's definitely a better way, but again, using what I have.) - Use the chunky spacers from before to hold the PCB to the panel. I'm mounting my PCB's with the jack connections at the bottom, as I've put all my banana jacks there for this design. - Test everything. Fix what's needed. - Repeat x50 and away we go! I think that covered your questions about it? I should shoot a video at some point, but what even is time?
|
|
Jihel
Full Member
knobs, knobs, and knobs !
Posts: 241
|
Post by Jihel on Sept 11, 2020 16:03:29 GMT
I like a lot your phenomenally well-made panels  |
|
|
|
Post by rodney on Sept 12, 2020 0:43:42 GMT
I imagine you and bearheadwood already considered this but I'm imagining a new front panel assembly, a bit longer and wider than your AE module. For each I/O connection the panel has a banana jack that is connected to a pin hanging down to line up with the female header on the original AE module.
In addition, in line with each of the AE knobs, there is an extension assembly that fits over the knob and feeds up through the front panel, expanding out into a nice tactile knob.
Ideally, the whole thing lines up and plugs straight into the AE module without soldering or screws. The total depth is about 4cm deeper so I assume we are forgetting about the AE case completely here.
my rough drawing, sort-of exploded see-through view. I imagine the little switches would be hard to do this way though and it would have to be a different for each module.
|
|
|
|
Post by young Protoboard on Sept 12, 2020 1:14:13 GMT
Thanks so much for the answers! - Panels are made of 3mm Acrylic, because I have that. Drill them next. Complain bitterly to myrself, because drilling panels is crap. TRUTH. I'm just glad I have a drill press that's PCB-sized and comfy to use. At 15cm height, are most of the banana sockets mounted over empty space, or do many hang over the AEM board? Is the extra-chunky standoff tall enough for either option? So your sockets (we might be using the same Cal-Test sockets) and switches come with mounting hardware, is that correct? Do your pots have them as well? Or do they kinda dangle with the knob resting on the front panel surface and the pots themselves are not mechanically mounted, per se? I think I'm getting a feel for your build here. Do you find yourself with a preference between stranded and solid wire for your connections to the front panel? I saw your banana cables on IG, and is was wondering the cost-per-cable (or bulk purchase cost, as I'd inevitably order in such a way) is. This thread should be renamed "The Banana Hammock," 'cause it's where all the banana boys come to hang out. |
|
|
|
Post by bearheadwood on Sept 12, 2020 10:20:27 GMT
I imagine you and bearheadwood already considered this but I'm imagining a new front panel assembly, a bit longer and wider than your AE module. For each I/O connection the panel has a banana jack that is connected to a pin hanging down to line up with the female header on the original AE module. In addition, in line with each of the AE knobs, there is an extension assembly that fits over the knob and feeds up through the front panel, expanding out into a nice tactile knob. Ideally, the whole thing lines up and plugs straight into the AE module without soldering or screws. The total depth is about 4cm deeper so I assume we are forgetting about the AE case completely here. my rough drawing, sort-of exploded see-through view. I imagine the little switches would be hard to do this way though and it would have to be a different for each module.
I had definitely considered the "hard patching" angle to connecting to the sockets, but when I realised I'd need to mount up to 12 of them I figured it would be easier to just run lines from the back, tbh. Love the idea of running extenders for the pots, they're the biggest pain I have at the moment. Replacement switches were pretty cheeeeeeeapp on ebaaaayyy. :DDD Thanks so much for the answers! - Panels are made of 3mm Acrylic, because I have that. Drill them next. Complain bitterly to myrself, because drilling panels is crap. TRUTH. I'm just glad I have a drill press that's PCB-sized and comfy to use. At 15cm height, are most of the banana sockets mounted over empty space, or do many hang over the AEM board? Is the extra-chunky standoff tall enough for either option? So your sockets (we might be using the same Cal-Test sockets) and switches come with mounting hardware, is that correct? Do your pots have them as well? Or do they kinda dangle with the knob resting on the front panel surface and the pots themselves are not mechanically mounted, per se? I think I'm getting a feel for your build here. Do you find yourself with a preference between stranded and solid wire for your connections to the front panel? I saw your banana cables on IG, and is was wondering the cost-per-cable (or bulk purchase cost, as I'd inevitably order in such a way) is. This thread should be renamed "The Banana Hammock," 'cause it's where all the banana boys come to hang out. I spent a day just drilling panels so I don't have to do it any more (Lies, there are always more T_T ) The sockets and switches all have their own mountings, yes. I'm reusing the pots from the PCB's so no mechanical mounting at this point. The "knobs on top and wire them together" method is actually working pretty well. I may just end up taping them to the panel with some gaffa, tbh. I'm not trying to make anything pretty here. Per cable was pretty cheap. $22 for 100 jacks and another $13 for (25m?) of wire, chop, strip and screw them in because they don't need soldering, so it's half a day's work for 45 cables, so about 80 cents each. I'm fine with stranded wire for the most part, I've started using dupont cables for the LED's, as the PCB holes are so small, and there's not enough of a pad to solder stranded wire to consistently. I'm deliberately drilling slightly small holes, so the sockets actually sit quite proud on the panel. There's plenty of space between the end of the jack and the PCB when they poke through. I'm doing it this way mostly because drilling large holes into acrylic has been frustratingly inconsistent. |
|
Jihel
Full Member
knobs, knobs, and knobs !
Posts: 241
|
Post by Jihel on Sept 12, 2020 11:25:49 GMT
How did you fix the AEM case on the wood case ? With tapes or velcro-like system ?
Attachments:
|
|
|
|
Post by bearheadwood on Sept 12, 2020 11:35:34 GMT
How did you fix the AEM case on the wood case ? With tapes or velcro-like system ?
Angled the case backwards, haha! Nothing really holding it in other than gravity there. |
|
Jihel
Full Member
knobs, knobs, and knobs !
Posts: 241
|
Post by Jihel on Sept 12, 2020 11:39:43 GMT
Angled the case backwards, haha! Nothing really holding it in other than gravity there. Yes gravity can help... |
|
|
|
Post by tIB on Sept 12, 2020 21:19:40 GMT
It's funny - back when I had my first pin header synth (the tinysizer) I set about a similar plan. These days I'm happy to accept while bananas are undoubtedly better (for me) I'm happy with the functional density that I can pack into the much smaller package the AE patch system allows.
That said it has occured to me one could make a very nice banana system with a slight redesign - make wider faceplates to accomodate a row of however many banana jacks were needed each side. It would require a different way to mount (or a system comprising of half as many modules if you added bananas in the current width format - or make all modules one space wider to allow for a row of banana either side. I won't be doing this.
|
|
|
|
Post by young Protoboard on Sept 12, 2020 22:33:58 GMT
I imagine you and bearheadwood already considered this but I'm imagining a new front panel assembly, a bit longer and wider than your AE module. For each I/O connection the panel has a banana jack that is connected to a pin hanging down to line up with the female header on the original AE module. In addition, in line with each of the AE knobs, there is an extension assembly that fits over the knob and feeds up through the front panel, expanding out into a nice tactile knob. Ideally, the whole thing lines up and plugs straight into the AE module without soldering or screws. The total depth is about 4cm deeper so I assume we are forgetting about the AE case completely here. my rough drawing, sort-of exploded see-through view. I imagine the little switches would be hard to do this way though and it would have to be a different for each module.
I had definitely considered the "hard patching" angle to connecting to the sockets, but when I realised I'd need to mount up to 12 of them I figured it would be easier to just run lines from the back, tbh. Love the idea of running extenders for the pots, they're the biggest pain I have at the moment. Replacement switches were pretty cheeeeeeeapp on ebaaaayyy. :DDD Sorry I missed rodney's post when I last commented-- Yes, I toyed with basically this concept at first. I'd agree with bradaddle in that a wire (even a DuPont wire with the banana socket end chopped and stripped) would serve better than a pin, and you'd want to solder the pin to the banana socket anyways (unless you can dig up a socket sized just so to make a nice, snug fit with a header pin). Also, as bradaddle mentioned, "hovering" the banana plug over the input would ge very crowded, very quickly (imagine a 'banana breakout' panel for the 2ATT/CV). Mounting bananas around a "perimeter" overhang of sorts, then solder wiring the inputs where they need to be, may prevent crowding as well as the knob height issue. Every other solder-free solution would work just fine; I think the custom pot knobs would get a little pricey, but if the front panel height from PCB allows, one might be able to cut down any cheap plastic knob until it fits without dragging on the front panel. It would make for a shorter knob, but it might just work. As for switches? The best solution I can posit would be a long plastic shaft with a small bore for the switch lever in one end; glue it onto the toggle lever and you'd be good to go. Such a solution may not work well for certain modules, such as the broken-out steps on the TRIQ164; that's a LOT of banana plugs. As TW modules are sparse in features to begin with, I would personally hesitate to cut any out. There's a cost issue as well; I could DIY copy the 2ATT/CV module to my banana format for cheaper than I could purchase the 2ATT/CV module and apply this "breakout panel." But... this would get the patch points over to banana, and if I could work this into an impact-free (no glue or solder) solution from the perspective of the TW module, then I could potentially utilize my current instrument while still preserving its resale value... Hmm... Oh, and I bought 1000 large switches for like, $40 from Aliexpress before COVID hit. Not AEM-board compatible, but great for DIY. The TW switches just look and feel so... dainty, that I worry about dropping a module when its front panel is off and bending it. That said it has occured to me one could make a very nice banana system with a slight redesign - make wider faceplates to accomodate a row of however many banana jacks were needed each side. It would require a different way to mount (or a system comprising of half as many modules if you added bananas in the current width format - or make all modules one space wider to allow for a row of banana either side. I won't be doing this. I feel like this would be my personal choice for adapting AEM. I also laughed out loud at "I won't be doing this."-- I won't be either, but it's a good idea nonetheless. With the sort of front panel "overhang" you describe, the AEM front panel height might just be able to be preserved. |
|
|
|
Post by spacedog on Sept 14, 2020 13:39:45 GMT
[...] So where does that leave AEM?I don't know, but it doesn't seem to fit my usecase as a person who can't afford AEM prices to begin with, and is left DIYing their own banana format. Simply put, I don't see any value in continuing with the AEM patchwire format, and my money would be better spent in components and design, or perhaps even switching to Eurorack and DIYing there. (3.5mm cables just feel a little too small and fiddly for me, though...) For all of the reasons above, I'm starting my new, easily-compatible-with-AEM banana cable format. [...] With apologies, I pulled everything back to this statement as it seems to sum up where your head (and your wallet) sits. My head is in a different space, which I will amplify and then take to a different spot, if appropriate. As for my wallet, it's always open to persuasion. In case it wasn't clear in my earlier discussion, from my perspective the AE Modular represents an excellent way for me to add to my palette of sound design. It's not so much that I can't do what it does by using other equipment, although there are a few things that I really struggle to do any other way; rather, it's that I can set up some (i.e. more than one) voices (sub-patches, perhaps) within one live large tweakable setup and then I can play it and record it live. I usually do this through having an initial seed of an idea and I start patching - for me, this usually involves trying to edge out modulation(s) that I can't always get with more fixed architecture. From there, I seek to add an appropriate amount of extra sauce to the flavour, sometimes with modulation(s) that I can sweep in, and sometimes with extra voices that can be brought in and out via the mixers. I do also add all of my effects outside of the setup, although not every voice has to go through the same chain of effects as I can send different outputs through different setups, all of which are gathered togther via an external mixer. Nothin mind-blowing there, but more to explain my playing. Whilst I have a modest setup compared to some, I can still do quite a lot and the key point is that I tweak live, I don't patch live. All patching happens during sound design. OK, if I need it, I will employ a switch, remembering to mute that voice on the mixer when I do it - but still, I'm not patching. Now, I do admit that I have brushed patch wires and created irritating noises, and I've brushed hard and yanked them out. Personally, I blame me for that and I try to act slowly around my patched AE Modular, but I can admit that in the heat of the moment something can come adrift. I do also have a combination of new and old "holes" and new and old "wires". So some combinations need to be avoided. My point was really that I accept the shortcomings and I work with them to achieve my desired outcome of live tweaked sounds. I don't really want anything bigger in format as I can deal with that as I separate the process of patching from that of tweaking and recording. I was aiming more at the view as to whether third party designers are coming to the format soon, and what stops them, and also why the format doesn't get discussed that much outside of this closed group. Hence my note that we need to recognise potential groupthink. I have discussed the AE format with a friend who considers his Eurorack setup to be value for money, When we bat modules and costs backwards and forwards, he puts up a decent argument on features vs costs for his value-for-money modues that he has. Soundwise, he's doing quite a lot with a little and I feel that I am doing the same. The price points are different, but not as much as some might think, mainly as he buys basic modules and always looks for a bargain. I would struggle to find the physical space for his setup and that's a key factor for me. He struggles with what he sees as a constrained diversity of modules. It is these discussions that prompted me to raise it here. Ultimately, in my opinion, it comes down to how much you are inspired by the interface and the sounds. I am and so I use my system a lot to add those peculiar flavours to my various musical noodlings. I use other equipment as well, of course, all to produce the final manifestation of what's in my head |
|
|
|
Post by young Protoboard on Sept 15, 2020 19:49:15 GMT
With apologies, I pulled everything back to this statement as it seems to sum up where your head (and your wallet) sits. My head is in a different space, which I will amplify and then take to a different spot, if appropriate. As for my wallet, it's always open to persuasion. In case it wasn't clear in my earlier discussion, from my perspective the AE Modular represents an excellent way for me to add to my palette of sound design. It's not so much that I can't do what it does by using other equipment, although there are a few things that I really struggle to do any other way; rather, it's that I can set up some (i.e. more than one) voices (sub-patches, perhaps) within one live large tweakable setup and then I can play it and record it live. ... My point was really that I accept the shortcomings and I work with them to achieve my desired outcome of live tweaked sounds. I don't really want anything bigger in format as I can deal with that as I separate the process of patching from that of tweaking and recording. I was aiming more at the view as to whether third party designers are coming to the format soon, and what stops them, and also why the format doesn't get discussed that much outside of this closed group. Hence my note that we need to recognise potential groupthink. No need to apologize. I think that perhaps I should have prefaced my posts and opinions with a big bold disclaimer of "FOR ME AND MY PERSONAL USECASE:". Nothing I wrote was intended to tear down or run contrary to what you wrote; rather, it was my own perspective, and a rather narrow, limited, specific perspective at that. In fact, I agree with much of what you said, especially in regards to the format's limitations and its effect on third-party designers. I think you made that second quoted point excellently, and I appreciate the perspective of live playing with a small setup. And further, I think together we laid out the format's strengths and weaknesses both in general and in specific usecases. As you've said: for you and what you use it for, AEM works great. And as I've said: for my usecase, it doesn't. This is true for many tools and many formats; Buchla, for example, likely doesn't fit the usecase of anyone with a thin wallet and enough time to learn the complexities of those behemoths. This doesn't make Buchla a "bad" format, just better suited for some than others, and I think the same is absolutely true of AEM. It's not bad because I can't use it, and it's not good because you can; it simply is what it is, with its quirks and uniqueness and pros and cons. |
|
|
|
Post by funbun on Sept 16, 2020 12:21:03 GMT
Do you think you might solder up your own Eurorack modules, or build a hybrid system? I'd love to see a cost comparison of a big AE modular vs someone who soldered their own Eurorack rack system like the Afrorack guy did.
Let me ask, how much time does it take to solder up a basic VCO module? Time to build and test is another cost to consider. Just like you said about test cases: for an engineer, well, time may be what you have the most of.
|
|
My reasoning is more along the lines of really wanting a larger format system then any issues with the size of the AE, though. 
