True Bypass Pros/Cons or My Bypass is Better Than Yours

[MoonWatcher]

I'm not really in a position to get boxes cast for my home builds, so if I want to use tiny little momentary switches (which we know DO fail, because people post here asking how to fix it sometimes; probably far fewer failures than 3PDT but let's be honest here; unless you're using £7+ vandalproof switches, the mechanical connection between circuit and human is always the weak link) I'd also have to get boards fabbed and find an effective way to use standoffs. So if I want to do electrical switching, it has to be with momentary panel mount footswitches, which start at £1.65 on rapid; then two logic ICs and, say, a handful of boards from a cheap fab because if I was going to go this route I'd probably do that. I think that by this point, electronic switching and 3PDTs would both cost about the same at around £2.50 (probably more depending on the boards for the electronic switching), at which point one should decide on which will be more effective for the application and not decide based on cost.

Bingo!

Along with that, what does time cost? It takes about 15 seconds to drill a 13mm/.5" hole. The labor involved with wiring up a 3PDT is not as bad as it is made to sound. I'd rather deal with 9 big lugs than either a pair of 16+ pin logic chips, at least two transistors or add'l op amp stages, etc.

How does the argument of "not durable enough for stage use" pop up, when folks use pedals with 3PDT switches in them all the time?

I guess it comes down to how you perceive a switch. Does it have to be designed to last for a thousand years, or can you live with it being a maintenance item, and just periodically replace if necessary? I mean, tubes wear out, speakers wear out.

I find it hard to believe that larger manufacturers went with buffered bypass for any reason other than the stupid cheap cost of a momentary switch and the scale that they build pedals on. It just makes sense for HOW they have to manufacture them.

I don't claim to have a monogamous relationship with 3PDT switching, but I also don't have Roland/Maxon tooling crammed into my garage.

[mictester]

So I do not feel that DIYers should use electronic switching to save money, because it doesn't, at our scales. Where appropriate (Eg if they're making 10 and getting boards made/making boards themselves) it's well worth considering, but what's cheaper for Maxon is not necessarily cheaper for most of us.

I find that with really high gain pedals (like the 21st century Big Muff), it's a good idea to mute the input and disconnect the output completely in bypass mode. It's sometimes a good idea to short the effect output to ground, too (as far as AC is concerned), so that any noise generated in the circuit doesn't find its way into the output. I can also use the CMOS switches to change the mode or between preset gains or between tone control settings... The comprehensive bypass switching I need could require several poles on a footswitch - there are three changeover switches in a 4053 and four SPST switches in a 4066. I can do so much more with cheap CMOS!

Now, should the guy convert his tubescreamer pedal to true bypass? Probably not. Should we all be using electronic switching for one off vero projects in hammond cases? Probably not.

I entirely agree with you, but when you want complex switching, or when you want sequentially selected settings, or you want a "tap tempo" system, CMOS wins every time! It's also good to have an understanding of the electronic switching options, too.

[Hides-His-Eyes]

So I do not feel that DIYers should use electronic switching to save money, because it doesn't, at our scales. Where appropriate (Eg if they're making 10 and getting boards made/making boards themselves) it's well worth considering, but what's cheaper for Maxon is not necessarily cheaper for most of us.

I find that with really high gain pedals (like the 21st century Big Muff), it's a good idea to mute the input and disconnect the output completely in bypass mode. It's sometimes a good idea to short the effect output to ground, too (as far as AC is concerned), so that any noise generated in the circuit doesn't find its way into the output. I can also use the CMOS switches to change the mode or between preset gains or between tone control settings... The comprehensive bypass switching I need could require several poles on a footswitch - there are three changeover switches in a 4053 and four SPST switches in a 4066. I can do so much more with cheap CMOS!

Absolutely; and this is what I mean by deciding on the context.

As for running different gain settings; I guess that involves (let's take an op-amp overdrive) running two drive stages and just repeating yourself with an op-amp half and a handful of passives on each side?

Now, should the guy convert his tubescreamer pedal to true bypass? Probably not. Should we all be using electronic switching for one off vero projects in hammond cases? Probably not.

I entirely agree with you, but when you want complex switching, or when you want sequentially selected settings, or you want a "tap tempo" system, CMOS wins every time! It's also good to have an understanding of the electronic switching options, too.

Oh, absolutely; and it's dangerous to have people encounter an electronic switching system and instantly assume "I can improve this with my solder gun and a clunky switch!" and for them not to be taught about why that was done.

By the time you've got TT, bypass, and multiple options... Fuck 4xxx, get a PIC in there and save yourself a headache

[mictester]

As for running different gain settings; I guess that involves (let's take an op-amp overdrive) running two drive stages and just repeating yourself with an op-amp half and a handful of passives on each side?

No - it's easier than that!

GainSwitch.png (2.6 KiB) Viewed 1833 times

You just have to switch the gain-setting pots.

[Hides-His-Eyes]

shit, I forgot the whole loop would be at V_bias anyway,

[mictester]

I forgot the whole loop would be at V_bias anyway,

Also, you can do something very similar with level and tone controls. I build quite a lot of gear with internal preset controls. Many players find that with (for example) a chorus pedal they want an on / off switch and (perhaps) a fast / slow and a shallow / deep switch. They can set up the sounds they like with the case open, and then leave the pedal preset for gigging.

The 21st century Big Muff is a good case in point. Again, I find that players have (perhaps) three settings that they like to use - I even know some guitarists who deliberately break the knobs off their pedals to stop them getting changed! With the Big Muff, I also found that it was interesting to have the option to bypass the tone controls and also another sound is to switch off one pair of clipping diodes. All this switching can be done very effectively with the cheap CMOS switches.

Oh, and by the way, Dirk - I was there. I didn't see you at the design meetings in Kobe, Japan in 1980 and 1981. They were interesting days.

[mictester]

Oh, and by the way, Dirk - I was there. I didn't see you at the design meetings in Kobe, Japan in 1980 and 1981. They were interesting days.

That's awesome to hear from someone who was there when possibly the all time classic circuit was designed..
Can you tell us who was the main engineer who designed the Tubescreamer as I've never seen it stated before ?

Like with all designs at Maxon (where I worked for a while and most of my friends worked), KME (where I mostly worked) and the other companies I knew, design was always by committee. The lead designer at Maxon at the time was a guy called Ito, but his assistants were the really clever ones.

The concept of the diode clipper in an op-amp feedback loop came out of radio communication gear - it was first used in the 60s to limit the incoming audio level in SSB receivers, to protect users' ears from static crashes. There were also attempts to use similar circuits to provide additional sustain to cheap solid-state guitar amps, with low distortion. Someone had pointed out years before that it sounded "a bit like a tube" and so the Tubescreamer was a natural progression. There was no moment, it was more an evolution than a revolution. For example, I remember endless experiments going on with delay lines and various oscillators to modulate the delay - they were trying to find out what modulation waveshape worked best (hypertriangular in most cases!), and "listening groups" who had to decide between different frequency responses.

I can put hand on heart and say that I provided the values for a five or seven band graphic equaliser filter so that they would use standard values and provide musically useful frequencies for the sliders. That same basic design has existed from the early 80s to today, and uses the same values that I calculated! The two prototypes were built on perf board with rotary pots....

[Greg]

Interestingly.. Maxon now claim True Bypass as a modern upgrade to their better effects:

"Some modern day utility features were also added, including mechanical true bypass via a 4PDT switch"

[mictester]

The Boss OD-1 seems to be the predecessor to the TS... it has the same frequency response, same IC, same tone control, etc.
Was there one of those lying around at Maxon ?

It's substantially the same circuit, because it came out of the same board manufacturing plant! At the time, the "Panasert" machines (automatic, programmed component inserter machines) were expensive and rare. There were just two plants that had them at the time, and only the KME plant made guitar effects! KME was a subsidiary of Matsushita (Panasonic), which is why they developed all those MN-series BBDs - they had the IC fabrication plant and the various "manufacturers" had use for their chips. You have to realise that, at the time, almost all the Japanese pedals were made by one company (Maxon) and then badged Boss, Ibanez, and all the rest.

In 1984, a lot of the Maxon pedal production was "out-housed" to a subsidiary of Hung Nien Electronics in Kowloon. The quality fell sharply - a lot of pedals were shipped with poor quality soldering (usually because of wrong temperature flow baths), and a good few had parts missing! I remember one batch of a pedal that had the bias resistors for the op-amps missing! Strangely, the quality of the cast cases increased so you could either get a properly working pedal in a cheap and nasty Japanese case, or a faulty pedal in a fabulous quality Chinese box!

At the time of the move to China, I made my move to the telecommunications subsidiary of Panasonic, so lost touch with what was going on in the pedal world for a few years.

[mictester]

Interestingly.. Maxon now claim True Bypass as a modern upgrade to their better effects:

"Some modern day utility features were also added, including mechanical true bypass via a 4PDT switch"

Well.... there you go. They obviously think that fragile and unreliable mechanical switches are good enough for gigging musicians. Either that or they're just acceding to the vagaries of fashion....

[Hides-His-Eyes]

At the end of the day they're a business; if someone will spend £20 extra on a pedal for a £2 part, so it goes.

[Dirk_Hendrik]

In the words of a Scottish comedian...

"I smell Shite!"

with above quote,

Sorry Mic.
Tha odds are against you. Although I do want to congratulate you with the claimed calculation of a series of gyrator frequencies. Job well done! I did similar calculations at the age of 16 without effort. Just punch the values into the calculator. Excel makes it even faster! The well chosen graph eq frequencies at ether an octave, 1/2 octave or 1/3 octave have been industry standard for years so determinig these should have been pretty easy!

No Mic,
with the few Japanese FX manufacturers that made it past approx '82 one can, continuously, see very clear design marks. While the basic circuits may be sort of similar in building blocks there's a signature in various brands that sets em apart (and makes the difference between a Boss, a Maxon, Pearl Yamaha or Korg so easily distiguisable). I've spent a little too much time in japanese 80's effects to accept a "one company" and then "badged". All the odds are against you.

so, since you so clearly claim hands on and I wasn't there and have to work from an analysis perspective.
Could you provide an example that support these "Maxon made Boss" claims, and better then the "certain similarities"


(eeh... How come that the Boss EQ's were made in by the Maxon gents but have different component values )

[Duckman]

AMZ TB w/relay... best of both worlds?

http://www.muzique.com/schem/bypass.htm

Not really... just a silent TB, but ... momentary switches last forever?

[Hides-His-Eyes]

What about the cheap world?

[Duckman]

What about the cheap world?

You're right Sorry

[DrNomis]

They are usually guaranteed for so many thousands of cyclical operations before they do fail...

"Guaranteed" or "Rated" ?

Both...


They are mechanical in nature,it's a given fact that anything mechanical will wear out,and or fail eventually...

Of course, and all switches are tested and rated for a given amount of cycles.. but I've never seen a guarantee and I'm not sure how you'd prove the amount of cycles a switch had been subjected to before failure.
Can you post an example.

I have read somewhere that,to rate the maximum number of cyclical operations of a switch,what they do is a "random" sample of a batch of switches,then they put the switch in a special testing machine,which counts the number of operations the switch goes through till it fails,I've got a failed switch amongst my bits and pieces,that came out of an old DOD Grunge pedal,it became intermittent in operation,and so,the bypassing circuitry was switching erratically...

I'm just looking in my Jaycar Electronics Catalogue for a good example of this,of course,there is actually no way to prove how many cyclical operations a switch has undergone,usually the mechanical parts just wear out from metal fatigue caued by constant flexing,or just friction between moving parts,or the electrical contacts which touch and complete a circuit,develop an oxide film which increases the contact resistance,or the contacts simply corrode from environmental factors...

I believe that those are the usual fail modes of switches...

The number of cyclical operations rating,is just a measure of how many cyclical operations (on/off) that a switch can be expected to last for before failing....

I'll have a good look around for a good example of cyclical operation ratings,if and when I do find it,I will post it here straight away,since I think all freestompboxes.org members may be interested to know this,so please stay tuned...

[mictester]

I have read somewhere that,to rate the maximum number of cyclical operations of a switch,what they do is a "random" sample of a batch of switches,then they put the switch in a special testing machine,which counts the number of operations the switch goes through till it fails,I've got a failed switch amongst my bits and pieces,that came out of an old DOD Grunge pedal,it became intermittent in operation,and so,the bypassing circuitry was switching erratically...

I'm just looking in my Jaycar Electronics Catalogue for a good example of this,of course,there is actually no way to prove how many cyclical operations a switch has undergone,usually the mechanical parts just wear out from metal fatigue caued by constant flexing,or just friction between moving parts,or the electrical contacts which touch and complete a circuit,develop an oxide film which increases the contact resistance,or the contacts simply corrode from environmental factors...

I believe that those are the usual fail modes of switches...

The number of cyclical operations rating,is just a measure of how many cyclical operations (on/off) that a switch can be expected to last for before failing....

I'll have a good look around for a good example of cyclical operation ratings,if and when I do find it,I will post it here straight away,since I think all freestompboxes.org members may be interested to know this,so please stay tuned...

Yes yes yes.. as I said I understand that they are tested to failure, and a rating is given as to how many cycles you can expect from them.

What I originally asked about, and am still asking about, is that you said they were "guaranteed".
A guarantee means they would refund or replace a switch that failed prematurely, and I've never seen that and don't know how it would work..
hence my question: "did you mean a rating or a guarantee" ?

I'm not trying to be pedantic.. I'm just interested, and I think you've missed the point of what I was asking.

The "number of operations to failure" is usually pretty arbitrary - some samples will last a lifetime, others just a few years. If you're really concerned about mechanical wear in the switches, go for optical switching. I saw a couple of prototype switches that had a "rubber dome" for the footswitch that operated a sprung plunger inside the box. The plunger then allowed light from an LED to a photo-transistor, which operated the logic. You could argue that the spring or the rubber dome could fail....

[Crazylead]

Problem starts when you used too many Fet bypassed stomps in your signal chain .

My best way is using a quality bipolar powered at maximum rating quad opamp buffer at the begining of the chain. Others True bypass mechanical switches.If i have a fuzz containing rig i place fuzz at begining guitar directly connected to it then buffer.

This is my way of bypassing anyway you maybe like yourself s choice .

[Lonkero]

What annoys me regarding the truebypass pros/cons is that Visual Sound has started their own quest against true bypass as switching method. Youtube has many videos regarding this, but judged by scientific method they don't offer anything really revolutionary, just another opinion to support Visual Sound's style to build their pedals.

Of course when you hook up many true bypass pedals together in chain you lose something (often a bit of the highest freqs) but the worst situation is when a single pedal (like good example is boss turbo distortion) alters your signal and cuts out all the highest freqs and then your tone is completely fucked up after that. To another thing...Buffers are great but may often cause some oscialltion with certain pedals.

Until someone develops a better way to retain original signal than true bypass i'll stick with it, no matter how the Visual Sound guys promote their non-true bypass pedals.

Well, just my 2 cents for this matter. Please continue

[Dirk_Hendrik]

As much as I respect VS's head of R&D....

Bob Weil never ever should have made those youtube video's, let alone make the claims he makes.

take one?

"It's a Video Opamp!! I'ts not even made for Audio!!"

(A video opamp is an opamp optimized for a high bandwidth and will therefore work just as well for audio)