Modular synthesis intro, part 5: Saving money
2018-01-21 intro electronics
This is Part 5 in a series that started with Part 1.
Back in Part 3 of the "intro to modular" series I said it would cost about US$1000 to US$2000 to get started in modular synthesis, and I said that although there are ways to economize, they are really ways to keep the price tag in that range, not ways to lower the low end further. This week, let's look at a couple of them.
Saving money by buying used
Modules, if kept in decent condition, usually retain most of their value on the used market. Some modules actually command a premium over their new price, either because they're no longer available new (for example, Modcan) or simply have limited supply (for example, Cwejman). On the one hand, the low discount for used modules means you're not likely to find many extreme bargains, and if you do, you should be careful that they're not too good to be true. On the other hand, it means buying a module you're not sure you will like, is safer - you can expect to sell it later and get back some of your investment, maybe even all of what you paid for it if you bought it used yourself. So buying used modules can be a good way to build up a system. It reduces the financial risk connected with whether you want to keep the modules. Modulargrid has a module marketplace. I don't really know much about it and can't give a solid recommendation for or against. So do other Web fora.
If you're in touch with other modular hobbyists in your geographic location, you can often buy, sell, and trade modules with them. That can reduce both the risks and the shipping costs associated with doing used-module sales over the Net.
Be wary of responses to "wanted to buy" ads; scammers have been known to respond using photos copied from eBay auctions as "proof" that they have the item you said you wanted, then they collect payment and disappear. Someone who often uses the alias "gearaudio" or some variation is well known for this scam - each time he tries it and it gets reported, they ban the latest alias and warn everybody, but he just seems to keep coming back and trying again. With a wanted to buy ad, you want to be very sure that the person you're dealing with can not only send you a photo, but also talk with you about the item they're selling in a way that shows they know what it is and have actually used it; genuine hobbyists can, and scammers usually can't.
Sometimes you'll see somebody's complete modular system offered for sale as a package deal. Be cautious with that. Buying a complete system means you lose the opportunity to choose all the parts individually, so unless you're sure you really want everything in the package, you should evaluate carefully how many of the modules you actually do want; how much the discount really is compared to the used prices of the modules individually (often, it's not much); and whether the discount makes up for the fact you may be paying for modules you don't want. Often a package deal on used modules comes from an estate, divorce settlement, or similar. The seller may not be the original owner, may not be a hobbyist themselves, and maybe just wants to liquidate the asset for whatever they can get. In that case, on one side they may be willing to settle for a low price because they just want it off their hands, but on the other side, they may well have unrealistic expectations and little knowledge of the market.
I don't think you should pay money for do-it-yourself projects that don't work. People sometimes try to recover losses on failed projects by selling them, with the suggestion the buyer can either fix the problems or tear the project down for reusable parts. It's seldom worth it. For that matter, you should be wary of any equipment (not just do-it-yourself stuff) that is being sold in less than working order. If you don't know exactly what's wrong with it, it's hard to make any fair estimate of how much it's really worth.
Saving money by building your own modules
Many beginners think they're going to save money by building their own modules instead of buying assembled. That's often less true than they imagine.
First of all, the only thing that makes sense is to compare quality for quality. If you spend US$200 to build a module that is really the same as a commercial module priced at US$400, great, you've saved US$200. That calculation ignores your time if you put a price on that, but it may be fair not to count time. If you enjoy doing electronics work, then the time spent assembling the module can be seen as a benefit instead of a cost.
But maybe your DIY module has a front panel made of circuit board material instead of aluminum, and maybe you used the cheap potentiometers from Thailand that wear out really fast and will be scratchy until they do, and maybe your module isn't calibrated accurately because you don't have the test equipment that the commercial factory would have, and so on... maybe you've just spent US$200 to build a module that is really worth US$200. All the savings came from settling for a lower-quality module, not from building it yourself. That's what I see happening with a lot of the people who claim to have saved significant amounts of money by building their systems from scratch. On a closer look, it turns out they've really only compromised on quality.
If we're going to compare quality for quality, there are several factors that make it difficult to save real money by doing it yourself. First, there's setting up your workshop. Starting at zero, it costs between US$1000 and US$2000 to put together all the tools and electronic equipment you need to build modules of commercial quality - maybe also including some expenditures on books, training, materials used up during practice sessions, and otherwise building your skills to be able to make modules. That price tag is about the same as a starter Eurorack system built with commercial off-the-shelf modules. So even if you assume (optimistically) that you can save 50% on module prices when you're building them yourself, and you're committed to only have modules you built yourself, you still won't break even compared to buying commercial products until you have a system twice the size of a starter system and you have paid the equivalent of full price for that.
Then, there are quantity discounts. Someone who buys parts in quantities of 1000 may pay half as much as someone buying in quantities of 10 or 100. That means commercial outfits making large batches of modules will be getting their parts cheaper than you can, partly compensating for the fact they have to pay for labour and you don't. There are also issues of whether you, as a small player, are able to source parts in small quantities at all; many distributors and manufacturers simply won't sell except in minimum quantities that are too big for an individual. They also may refuse to sell to anyone who isn't a business. So you face extra challenges dealing with those obstacles, and may end up paying for more parts than you need, having to organize group buys, or convincing someone who does run a business to act as a middleman.
Quantity discounts mean you'll sometimes end up ordering more parts than you immediately need, to have some left over for future projects. That's great if you really do use the extra parts, but if you don't, the money you've wasted on them will probably erase the discount. Using parts in future projects often doesn't work as well as you may hope, because there are so many fiddly details to sourcing the right parts - you may end up with a stock of parts that are almost right for your next project, but not quite, and so you still need to buy new ones. And even if you eventually will use all the parts you bought, having to tie up your money in inventory, store the parts and keep track of them, and so on, adds to the cost of running your workshop.
Some of the quantity and sourcing issues can be reduced if you build from kits instead of trying to obtain all the parts for a design all by yourself. Whoever's selling the kits will want to take a profit, but because they have access to quantity discounts you don't, the price of a kit may still compare favourably to what you would pay in a pure do-it-yourself situation.
Some manufacturing processes are completely impractical outside a large factory. You can kind of do reflow soldering with an ordinary kitchen toaster oven, but that doesn't work perfectly and may introduce reliability problems. Lots of people etch their own circuit boards at home, but when you look closely, you realise that almost all of those are single-sided boards, with bare copper, no solder mask, no silkscreen, and no plating. Quality for quality again: at just a slightly higher price than those homemade boards, you can order boards custom-made in China that will be double-sided, with plated-through holes, solder mask and silkscreen on both sides, and gold over nickel plating. But you have to order at least 10 of them. It's not realistic for a hobbyist to do all that on a single-unit basis in the garage; it wouldn't cost less than the Chinese factories with their huge volumes and robotic equipment; and if you attempted to do it anyway, you'd quickly attract the wrong kind of attention from your local regulatory agencies regarding how you planned to dispose of the waste. This again favours commercial production.
Let's not even get started on the safety issues related to handling peroxide-boosted acid etching solutions, which some hobbyists will cheerfully assure you must be perfectly safe because the granola people sell hydrogen peroxide at the health food store.
On top of all that, there can be issues regarding module designs. Most manufacturers of commercial modules keep the details of their designs secret. You can't just point at a module offered for commercial sale and say "I'll build one of those"; you're faced with trying to come up with a design for it either by yourself, from public sources and the hobbyist community, or by reverse-engineering the commercial one. That doesn't have to be a dealbreaking issue. Many high-quality designs are available in public sources, and some of the most innovative designs are born in the hobbyist community and only available as do-it-yourself projects. If you want a Turing Machine, for instance, you have to either build it yourself or pay someone to build it for you on a bespoke basis.
A few commercial manufacturers, including Mutable Instruments and my own North Coast Synthesis, make it a policy to publish all their designs. But these designs still may be optimized for commercial production - including the use of parts that aren't available in small quantities, or technologies like fine-pitch surface mount that aren't practical for hand assembly.
For modules with programmable digital parts (microcontrollers, FPGAs, CPLDs, and so on) there may be firmware required on top of the hardware design, with associated copyright issues, a need for specialized equipment to download it into the programmable device, and so on.
The bottom line is that it just doesn't make sense to turn to DIY construction solely as a way to save money. It'll cost so much in both time and money that you won't really save. Nonetheless, DIY construction may make sense if you have other reasons to do it beyond saving money. For instance, if you already have an electronics workshop as a result of some other activity (like ham radio, say), it brings down the cost of applying that to synth building also. And maybe electronic construction is your real hobby, a goal in itself, and it's to do that that you got into modular synthesis in the first place instead of the other way around. That's a big part of my own story, and in such a case, saving money is beside the point.
Continue to Part 6 of this series.
◀ PREV Random thoughts on negative resistance || Live-coding a permutation-based fugue NEXT ▶