New Triad Clips! Featuring Jackson Ampworks Britain

[LyleCaldwell]

Courtesy of Martin Lavallée, a great player up in Canada, who happened to get the Triad prototype on the same day he received his new Jackson Ampworks Britain amp. So he was kind enough to let them play together, along with a '58 Tweed Deluxe, a '63 Vibroverb, a '65 JMI AC30, and a recent Marshall 2061X handwired 20W. Going along for the ride was a '59 LP RI, a '53 Tele, and a '73 Strat.

Check it out:

http://psionicaudio.com/main/triad_audio.htm

[Alex C]

That's simply amazing.  I downloaded and thoroughly enjoyed the earlier clips you posted.  That sounds really good.

[LyleCaldwell]

Thanks!  I'm very pleased with how it came out.  It's been fun going from "I wonder if..." to this point.

[$uperpuma]

I go way back with the Jackson brothers REALLY nice guys...nice amps too

[LyleCaldwell]

Sorry, didn't see your reply, $uperpuma.  Brad's a great guy, and his amps are built to perfection.

And now, the gut shots some of you have been asking for...

The board:




One side view:




The other side view:

[syndromet]

Man, that thing is amazing! It's stuff like this that make me realize I'll never be able to make it as a Butique-builder.

[LyleCaldwell]

Thanks, but I'm a long way from "making it."  This first run of 50 was to see if there is ienough nterest to justify my financial gamble.  <g>

So far it looks like the answer is "yes" so my next step is to make some changes for future runs that will cut my assembly time.

First, I have to eliminate as many wires as possible.  So the board will be growing down over the switches so the switches can be attached with shorter wires that can all be attached with the board and switches out of the enclosure.  It would be easiest to put the switches directly on the board, but I'm not sure that's a good idea in terms of vibration and stress - still pondering that, plus it would complicate the angle needed to get the board in the enclosure.  The pots will all be soldered to a daughterboard that will attach to the main board with a multipin connector.  That will not only eliminate a lot of wires but will speed up pot alignment.  I'll probably have to go to D shaft pots instead of the solid round shafts, but that will also speed up the knob attachment, as I currently have to align each one by eye.

Doing all that will take about 2 to 2.5 hours off the assembly time of each pedal, and if I do the new layout correctly and choose the right D shaft pots will sound the same as the current method (better in theory, actually, as I can shield traces from each other).

Or I could just pull the recent "boutique" trick and charge $800 for one (not a slam on Clay Jones, who sold his ODs at a reasonable price - the current price idiocy is not his fault).

[Phorhas]

MAN, she sure is beautiful!

Good luck with that Lyle

[BDuguay]

Great job for sure!
I'm envious of not only your pedal creation, but the fact that you've got the cojones to go ahead and make a go at this.
I wish you the best of luck and sincerely hope this really takes off for you.
AFAIC, you've won half the battle already by developing a well designed, well built, versatile peice of gear.
B.

[LyleCaldwell]

Thanks.  Something you guys will appreciate:

That one lonely diode on the bottom of the board is a sacrificial diode that will burn out when someone (inevitably, despite the labels and my many warnings) plugs incoming 9vac into the outgoing 9.6vdc jack.  The diode will fry, releasing its magic smoke, but the rest of the circuit will be untouched.

So I put that diode on the bottom so it will be easy for a regular tech to replace without totally disassembling the Triad.  It's just a 1N4002.

The second batch of boards will use a 1N5403 there, which should protect the circuit without burning out, but the first version of the board won't take a 1N5403 in the board holes.

Part of the learning process, both in terms of design and of bringing things to market.

[Mark Hammer]

First, kudos.  Nicely done.

Second, smart idea about the extra diode NOT on the main board.

Third, why the heck did you put the white washer on the inside?

Fourth, I've become convinced that at least SOME of the switch failures from stompswitches occur when there is sufficient heat buildup across contacts during soldering to cause reflow of the otherwise thick goop inside the switch.  That goop would normally just be sitting on the middle pivot point of each rocker contact inside the switch so as to hold the rocker contact momentarily until the two switch halves are combined and the side tabs squeezed.  When excessive heat is applied, the goop seems to flow outward from the central pivot point along the rocker contact, blocking contact with the side lugs.

Now, somebody must have figured out a way to reliably solder these things in large quantity without causing this form of switch failure because, let's face it, places like Dunlop and Fulltone could not continue to exist if they couldn't avoid such switch failures.

But these places are cranking out larger numbers of such pedals, and more importantly, only use a single switch per pedal most of the time.  So, I'm wondering.  Not to underestimate your fabrication chops (which clearly exceed my own), but I gather you aren't wave soldering in some Chinese factory, so how the heck are you going to assure that you don't find yourself with a board that has itself secured to three stompswitches, and one of them is bad?  I can easily see the advantages of mounting the switches to the board, but how does a guy who turns out units in the dozens, rather than thousands, arrange to have soldering consistent so as to avoid such production issues?

Again, I'm not questioning your chops.  This is more of a practical "how do YOU do it?" sort of question.

[LyleCaldwell]

First, kudos.  Nicely done.

Second, smart idea about the extra diode NOT on the main board.

Thanks, and thanks.

Third, why the heck did you put the white washer on the inside?

The washer isn't needed on top as the finish is extremely durable.  But in playing around with the height of the switches, I found that the washer in combination with the nut and the lock washer ends up being the perfect height.  So I just tighten the nut down over the washer, slide on the lock washer, and then each switch is exactly the same height.

There's no mechanical downside, so I'm happy with it.  Using a second metal nut instead of the plastic washer might look nicer, but no one will be looking inside that much.

Fourth, I've become convinced that at least SOME of the switch failures from stompswitches occur when there is sufficient heat buildup across contacts during soldering to cause reflow of the otherwise thick goop inside the switch.  That goop would normally just be sitting on the middle pivot point of each rocker contact inside the switch so as to hold the rocker contact momentarily until the two switch halves are combined and the side tabs squeezed.  When excessive heat is applied, the goop seems to flow outward from the central pivot point along the rocker contact, blocking contact with the side lugs.

Now, somebody must have figured out a way to reliably solder these things in large quantity without causing this form of switch failure because, let's face it, places like Dunlop and Fulltone could not continue to exist if they couldn't avoid such switch failures.

But these places are cranking out larger numbers of such pedals, and more importantly, only use a single switch per pedal most of the time.  So, I'm wondering.  Not to underestimate your fabrication chops (which clearly exceed my own), but I gather you aren't wave soldering in some Chinese factory, so how the heck are you going to assure that you don't find yourself with a board that has itself secured to three stompswitches, and one of them is bad?  I can easily see the advantages of mounting the switches to the board, but how does a guy who turns out units in the dozens, rather than thousands, arrange to have soldering consistent so as to avoid such production issues?

Again, I'm not questioning your chops.  This is more of a practical "how do YOU do it?" sort of question.

I'm very far from wave soldering in a Chinese factory.  I'm using 25W and 35W Weller pencil irons with a Radio Shack iron holder on my kitchen table.  Though I might order some Chinese takeout for lunch as I work, that will be as close as I get.

The answer so far is just trial, error, consistent soldering approaches, and really nice Belden 22AWG silver stranded wire.  I strip each wire just enough to go inside the lug on the switch, and then use a really hot iron very very briefly, causing the solder to flow.  I use really thin solder for this, so it flows quickly but not a lot at once.  Doing it this way, each lug is only touched once for a very short time.  I get nice shiny solder joints but the switch doesn't get that hot.

I tried this with some 24AWG wire from smallbear, and while it was nice for general purpose wire, in the Triad (small spaces, high gain) it was microphonic and the smaller gauge had more space around the wire inside the switch lugs.  The Belden 22AWG isn't microphonic and it exactly fits inside the lugs, so there's more metal to metal contact and it takes less solder as no solder is needed as glue/filler.

I've had one of these blue switches fail, in one of my first prototypes where the switch in question saw repeated resoldering.  The plastic holding the lug went soft and the lug could move around.  So I think your concern is absolutely correct, but so long as I'm consistent and quick I don't foresee problems.

(well, eventually I will have pedals returned to me to have switches replaced due to mechanical wear and tear.  That's inevitable) 

[BDuguay]

Lyle

I solder for a living and I'd like to offer a suggestion. While your current method of soldering the switches is working fine, I have method that will further decrease chances of over heating the switch lugs.
If I may.
When I wire up switches, or any off board wiring, I strip the wire, apply a drop of flux on the exposed wire and touch it with a dab of solder. This process is called tinning.
The tinned portion can now be bent into a hook that can be secured in the switch lug. Once inserted into the lug I pinch the hook tight against the lug creating a mechanical bond that keeps the wire in place for soldering. This method greatly reduces the amount of time the iron needs to be applied because the soder flows quicker to a pre-soldered wire.
I believe the extra step of pre-tinning is worth it for the advantage of being able to pinch the wire in place leaving you both hands to apply the solder.
Keep in mind though, I am about 1 extra hand wash away from being completely O.C.D.

Solder on Soldier

B.

[LyleCaldwell]

Thanks B.  I tin just about everything except the switch lugs, as there isn't room for the wire to go through the lug hole if the wire is tinned.  The Belden wire I'm using is partially tinned already though - the solder flows fast.  And this insulation just does not melt - I love that.

[Mark Hammer]

Okay, so if I'm understanding you correctly, you will be soldering little solder core "legs" to the switch lugs, then inserting the feet of those stubby little legs through holes in the PCB and then soldering the "feet" to the board?

[LyleCaldwell]

Not sure if I will be doing that (with some flex built in) or having the PCB come down to surround the switches with square cutouts, and holes in the PCB surrounding the switch lugs.  Either way would decrease the wire length and speed up assembly time.

I was also pondering making little square PCBs for the switches with solder pads around the edges, so any strain would be on the solder pads instead of on the switch lugs.  This would also give me the ability to use small ribbon connectors for the switch connections, eliminating the individual wires entirely, if I can buy suitable cables of the right length.

All a matter of budget - a more expensive part is more than justified if it eliminates enough labor.

At the end of the day I want to be a designer and maker, but not an assembly worker.

[Mark Hammer]

It's precisely those little cutouts for the switch lugs and the heat required to fill those gaps up with solder that I was initially so concerned about  I'm going to recommend some sort of intelligent heat sinking during their installation, just for reliability's sake..

[LyleCaldwell]

Thanks Mark, I will definitely keep that in mind when I got to the next board design.  It's always something.