why are boss pedals so complicated?

[jamiefbolton]

wow! lots of comments! cool!

yeah i see what you guys are saying about consistency, reliability, and the FET switching thing. that does add a bunch of parts to the list. however, i would like to see Boss try their hand at a more "boutique" line. just to see what they come up with....

[newfish]

I'm guessing the FET switching doesn't add any more 'extras' than well thought-out polarity protection.

If you're making a pedal for your own use, you *know* which way round the connector goes.

The issues start when other people (who haven't spent hours lovingly making your pedal) start borrowing / using them.

3DPDT / DPDT switches are a great place to start though - a simple, sturdy way to get your stomp boxes up and running.

+1 on the Danelectro Tuna Melt.  If it had a wave mix function, it would still be on my board.
Homebrew kit took its place (with a DPDT switch, in case you were wondering...)

0.02

[Fender3D]

...The issues start when other people (who haven't spent hours lovingly making your pedal) start borrowing / using them....

lol 
back in '70, I saw the schematic for a coiless wha in an electronic magazine, I never used a wha pedal 'till then, so I borrowed a spare volume pedal from my keyboard guy and mounted the circuit into.
I discovered I mounted the pedal coarse reversed several years later when I borrowed it to a friend...
I'm still using reversed wha 

[DougH]

however, i would like to see Boss try their hand at a more "boutique" line. just to see what they come up with....

Okay, so you want to see BOSS build yet another version of the tube screamer too?  

Why? What's the point?

I guess my point is what does "boutique" mean to you?

To me, in the majority of cases it just means a pretty box with some internet-buzz fed hype about "tone". Usually some solder-jockey is too busy in his garage trying to meet a production quota based on orders (or the infamous "list") to even think about coming up with any new ideas. So they usually don't, even if they are capable.

A big company like Boss or EHX can afford to innovate new products and ideas. They can afford to do real R&D and hire personnel with the right qualifications to do that. Why do you think EHX is coming up with new (and IMO exciting) products every year? Even Boss has come up with some interesting ideas, feedbacker and acoustic simulator come to mind, in addition to their relatively new DSP stuff.

Not all of it is a success but they can afford to take chances. And having the environment where you can afford to take chances is what keeps everything moving ahead. That's why we have new and cool stuff like the Freeze, HOG/POG, even the Whammy pedal and etc. Without that environment, they are just rebuilding and perfecting the "perfect" tube screamer or overdrive- which is the kind of thing most of the boutique industry seems to be obsessed with.

[jamiefbolton]

i don't know. i would just want to hear what their engineers can do again in the drive department. no DSP stuff, but simpler analog circuits. just to hear it.

[Mark Hammer]

Whoops.  Check my last post here:  http://www.diystompboxes.com/smfforum/index.php?topic=86288.msg722755#msg722755

[DougH]

i don't know. i would just want to hear what their engineers can do again in the drive department. no DSP stuff, but simpler analog circuits. just to hear it.

SD-1, DS-1, Blues Driver, the overdrive with the discrete op amps. Lots of people like those, even/esp pros.

[earthtonesaudio]

I keep promising to prepare a photo-augmented tutorial, but now I think it really needs to finally get done.

"Remote" switching of Boss-like pedals that use a momentary to actuate an electronic switch is a simple thing to retrofit. It involves putting a hole in the box for a mini-jack, but is non-destructive in that that it changes nothing about the normal functionality of the pedal.  What it DOES do is permit you to stick the pedal anywhere you want or need to, and run a slender cable to another spot on your pedal board, or wherever, and actuate the electronic switch remotely.  For that matter, you could run a sort of snake up to your pedalboard, and have a soundman do the pedal-switching for you from a little pushbutton unit by the soundboard.  Doing that with TB-based pedals would be a big nuisance, not to mention a potential source of noise and tonesucking, when all the needed cabling is considered.

So, the complexity of the switching circuitry actually can help to make other tasks much simpler and easier to implement.  I might point out that remote switching capability via a simple phone jack on the back, and an SPST momentary situated elsewhere, is something you often find with rack-mount equipment.

Stay tuned.

That sounds really cool!  Most Boss pedals have one stereo and one mono jack, so you could avoid drilling a hole for another jack by swapping the mono to stereo and using a splitter cable.

[Mark Hammer]

That sounds really cool!  Most Boss pedals have one stereo and one mono jack, so you could avoid drilling a hole for another jack by swapping the mono to stereo and using a splitter cable.

pros and cons of that.

Yes, it involves no drilling or other installation.  But, it require special cables to work and may compromise plain vanilla operation with a battery as a standalone pedal.  What I illustrated (in the other linked-to thread) lets you leave the pedal working as before, but with an added optional feature.  As long as its not messy and the soldering job doesn't compromise the momentary switch, I can't see the mod lowering the resale value of the pedal.

[slacker]

So is it just me or are these guys more or less designing/building complex preamps that often have no practical use but for a very limited number of amplifiers.

Worst commercial pedals I've owned to date would be the Danelectros.
Best would be Sansamp and of course, my tube screamer.
Fact is, the few I still have all collect dust now.
I think in many cases, it's been a matter of giving someone the option to buy a 40-60 dollar pedal to use on a really scummy sounding amp and step up the performance to a respectable, usable level when you can't afford a new amp.

I guess that depends on why you're using pedals, if like you say you're using them to make a "bad" sounding amp sound better then that probably means what you actually want is a screaming Marshall stack or something not a pedal. Once you get a decent amp with whatever sort of sound you really wanted all along you're probably going to be disappointed with your pedals compared to it.
That's not why I use pedals though, I use distortion/overdrive pedals because I like the sound of them. I've got an amp with a decent dirty channel if I want it, but a lot of the time I still use pedals.

[DougH]

I have a screaming "marshall" (half) stack and I still like driving it with a pedal sometimes.

[MoltenVoltage]

Those of you who haven't yet might want to read "What are all those parts for?" at GEO. The simple answer is that they're there to get (a) consistency in the face of tolerances and (b) long term reliability; at least to the end of the warranty period. Consistency is to be valued much more than a small percentage of amazing products. Having every unit perform as well as every other one is extremely valuable.

And a lot of things go into what I call making a circuit "play well with others".  You want to be sure that no pot/switch setting produce shrieking oscillation or dropouts. You want to be sure that simple things like opening the case doesn't produce the possibility of circuit damage. You want to be sure that you don't put controls or settings on there that could cause inadvertent nastiness for a slightly-drunk musician in front of several hundred or thousand people. You want to be sure that when the bass player plugs his speaker cable into the pedal input, the pedal is not destroyed. Think of every possible way the pedal will be misused. You can be sure that at least one of your customers will do that.

All of these things can be removed from a DIY pedal to make it simpler. That may be a good thing because while there are some really good practitioners in the DIY world, the majority of DIYers are still learning to solder, learning where to buy parts, learning where the wires go, learning which way round a polarized capacitor goes. For these folks, the simplicity is good; I'm relatively certain that went into shaping the original poster's comment.

Albert Einstein said "everything should be as simple as possible, but no simpler". Commercial pedals are more complex because they have to do more and different things than DIY pedals.

This got me thinking about how Boss makes their Phaser with JFETs without having someone sorting them all day long to match them.  I assume they need to bias them so it becomes a no-brainer.

RG has drawn a Boss Phaser PH-1r schematic and, from what I can tell, each JFET is biased:

http://www.geofex.com/FX_images/Boss%20PH-1r.pdf

[Mark Hammer]

If you're a OE manufacturer and get an order of 1000's of FETs in from Fairchild or whomever, I would think that each order is reasonably consistent, and you CAN just grab a fistful, smoosh them into a PCB, bias away and be done with it.  The problem would seem to be batch-to-batch variation, where the odds of substantive discrepancies are higher, and the need to take steps to match that much greater.  If hobbyists like us order from a smaller distributor, that distributor may have a bin box with a pile of 2N5952s and when the pile starts gettng low, they order some more.  The order comes in and the bin box is replenished.  But it contains stuff from this order and the previous order.

(P.S.:  The extra components between gate and drain are intended to make the FET more immune to distortion.  See this now-classic drawing:  http://www5b.biglobe.ne.jp/~houshu/synth/PhaseFet0205.GIF )

[Gus]

Mark I bought  a bag of 1000 jfets for condenser microphone circuits same brand.  They are all over the place.
Assume JFETs need to be tested and sorted.  

[Mark Hammer]

I stand corrected.

Did you get them from the manufacturer or from a retailer/distributor?

[Gus]

From mouser.  Same brand and I GUESS batch.  I built a circuit fragment and a simple IDSS tester to sort.  I now have a curve tracer that I hope one day to add something like a USB A/D pod from the curve tracer X,Y output to the A/D input and then to the  computer.

[greaser_au]

If you're a OE manufacturer and get an order of 1000's of FETs in from Fairchild or whomever, I would think that each order is reasonably consistent...

Mark,
Following on from your suggestion, I'd imagine that as an OE manufacturer ordering parts in the 1000s & 1000s you could ask the manufacturer to pre-sort batches of parts to a specification at line test for very little extra cost...



david

[Mark Hammer]

Let's consider the case of Behringer, which uses the venerable 2N5952 for their PH9 pedals.  Given the volume of sales they aim for, and the low production costs they aim for, I find it hard to imagine they would sink the time, money and effort into testing and sorting the trannies themselves.  They gotta come presorted/matched from the manufacturer.

[R.G.]

Let's consider the case of Behringer, which uses the venerable 2N5952 for their PH9 pedals.  Given the volume of sales they aim for, and the low production costs they aim for, I find it hard to imagine they would sink the time, money and effort into testing and sorting the trannies themselves.  They gotta come presorted/matched from the manufacturer.

Maybe. In big quantities, the maker will do it. But sorting/binning is only a problem if you can't afford to do it automagically, and even a little time spent on fixturing and semi-automated testing would speed things up and dumb down the task so that good results could be had from the only self-programming, adaptive logic, general purpose computers that can be assembled with totally untrained workers: humans.

I would use something like Sculpy or Fima poly molding clay to make a fixture that just fit the rounded back of a TO92 case, and had indentations just fitting the lead wires. I'd put the spring contacts out of an RJ11 connector one per leg-channel so that connecting the device itself would just be dropping it into the depression and pressing down to hold it in place.

That taken care of, I'd make a circuit card that used the built-in A-D converters in a programmable controller to do what amounted to either my quick and dirty JFET test or as complicated as you want to go. The controller just needs to be told what to put on the pins and what the measurements are. That sounds complicated, but a single $2.00 uC programmed with something as simple as BASIC will do it fine. I'd also provide a small two-digit seven-segment display driven from the controller to give the results out to the operator. Finally, I'd set a number of bins/buckets under the workstand corresponding to how finely I wanted to sort the devices.

The uC also honks and squawks if the device is no good or didn't get pressed in place correctly. It could even send a message to the overseer with the whip if there wasn't X devices per hour going through it.    But I digress.    I can see all of one day of design time to get the uC set up and running and maybe another day or two of skilled work to make the fixture if you're doing something like molding.

The task per test station is (1) insert a device and press it into place. An insulator on the index finger of the operator lets this be tool free. (2) the uC reads the device, decides which bin it goes into, and puts the number on the display (3) the operator drops the device into the correct bin, then grabs another part.

I can't see that taking more than a few seconds per part. Other than being mind-numbingly dull, it's something that a human can do for minimum wage, which is kinda what mind-numbingly dull means, I guess. A phaser needs four JFETs, so the sorting content is, call it 24 seconds per pedal. A human working an 8 hour day will get probably 4 hours of useful work done at such a task. The current US minimum wage is $7.25 per hour. The day costs you $58 and you get from that 4*3600/6 = 2400 sorted JFETs, so the cost per pedal for sorting your own JFETs is ($58/2400)*4 = $0.096666... . Call it a dime a pedal for four sorted JFETs. One sorter can keep up with a production rate of 600 pedals per day. If you need more production, use more sorters.

Since the test is so fast, one operator per tester is about all you can do. If the test was longer, you'd use multiple stations per operator, but this test is effectively instantaneous to humans. The only difference between this and what would be done at a factory is that you have substituted in a human for the capital machinery to move a device into place, orient it, and then drop it into the appropriate bin. Machines can do this much faster once you build the machines, but humans are much more flexible to put into place and get going, as well as to re-use other places when the job is over.

[MoltenVoltage]

What about reels of surface mount parts?  Or reels of through-hole parts for that matter?

Suddenly your pick and place robot becomes horribly inefficient if you need to manually place parts on a PCB after hand-testing them...and of course your manufacturing cost jumps way up.

I figure the large scale manufacturers must be able to get JFETs pre-sorted within a certain tolerance.

Manufacturers can test surface mount resistors that are .4mm across, so surely they can test other parts.

That being said, I haven't ever found pre-sorted JFETs from Mouser.

My point?  That companies that mass-produce phasers for $29.99 retail probably don't care if its "dialed in".  I haven't used any of those pedals but I have a hard time imagining that they can even begin to compare with their more carefully manufactured cousins, except by accident.

Anyone ever bought one of the cheap ones?  If so, did it sound good?