Jawari do, Jawari don't

[moid]

Well come now all of you big smart men
Uncle M needs your help again
Got himself in a terrible jam
Way down yonder in vero-land
So put down your books and pick up a soldering gun
We're gonna have a whole lot of fun!

(with apologies to Country Joe and his aquatic friends)

Here's a story for you: I've been wanting to make a Tim Escobedo Jawari for years. Never got round to it. Then a little while ago mitchelr posted that he'd found a source for a transformer on ebay in the UK that wasn't expensive and worked for him on his build of it https://www.diystompboxes.com/smfforum/index.php?topic=124704.msg1184099#msg1184099 and I thought, yes, let's do this! Well guess what? it don't work! But it don't work in a weird way... if I power the circuit via a 9volt battery I can get some audio through (very quiet, I have to crank my amp to hear it) but I don't think there is any 'Jawari' sound in it; maybe a faint hint of something, and strumming / picking while turning the pot on the circuit seems to modify the fading harmonics of the notes, but also crackles madly while I turn the pot (like the ZVex Crackle OK circuit), but stops crackling as soon as I stop turning the pot. If I power the circuit through a 9 volt wall socket power supply, the LED comes on but no audio at all - just really loud hiss!

Weird eh? What say you, O enlightened ones?

I used Harald Sabro's vero layout:


The original schematic is here:


The only difference between my build and Harald's is my J201 is an SMT mounted on a mini PCB (I don't have any through hole J201), I'm using a 3PDT switch for true bypass and the original schematic uses a 2PDT. The transformer I'm using doesn't have a ground leg, but mitchelr used it and said ignore that... otherwise it should be the same. I assume my 1n34A diodes are correct (they were expensive so I hope they aren't fakes)

The J201 should be OK - I did a test of it using the technique listed here: https://www.allaboutcircuits.com/textbook/semiconductors/chpt-5/meter-check-transistor-jfet/ and it showed 0L in one position and really high resistance in the other position (23M - 56M!). It didn't read 0 though, so maybe that is wrong?

Should I buy the original 42TM018 transformer and try that instead? I've found a source for them in the UK; they're about £6 each so I can buy one if that's what is likely to be the problem. I've found a source for through hole J201s as well, they are expensive but I'll buy them if the SMT is the issue)

I know how much you love large photos of my amazing soldering skills, so here are some shots (before you tell me off about the legs sticking through the board; those are the transformer and I'll snip them as soon as I can prove the problem isn't the transformer - if it is the transformer I'll remove it with its legs so I can hopefully use it somewhere else in future). The 1N34's are socketed and I haven't trimmed them down yet - they don't touch each other / anything else so I'll do so as soon as the circuit is working - they are bloody expensive!). There is a 1n4001 hiding behind the capacitor on the right hand side




Thanks for any thoughts / ideas!

[duck_arse]

moidy! I think you need the centre tap on the transformer so's the diodes know what to do. and, if those germanium diodes are Russian types, they usually mark the Anode with a band, whereas the rest of the world marks Kathode/band. depending what you are feeding the input, you might want to fit an input cap - 10nF should be plenty big.

and yer pot should have three wires - hot to the fet, cold to ground [0V, common, earth], and the wiper to that cap, C2. otherwise, you'll get no DC through your fet, and those voltage measures you posted will .... ohh ... wait a minute ....

[Mark Hammer]

Think of the Jawari as an octave-up fuzz with very little "push" to it, such that the doubling fades away quickly, resulting in a quasi-filtering reminiscent of a sitar drone. 

I understand how one can perseverate to finally get a circuit working.      That said, if all you really want is the sound, the same drone is easily achieved by reducing the gain/sensitivity of pretty much any octave-up fuzz, and picking close to the bridge; one of the under-exploited perks of such fuzzes.

Personally, I loathe stripboard.  I'd much rather work from a schematic with perfboard.  MUCH easier to test and debug circuits when you know what part is doing what.

[moid]

Thanks for the fast replies chaps! Duck, I added that missing ground cable from pot 1 to power jack ground (d'oh, can't believe I missed that one). OK so now I can get some audio (quiet) to come in using the wall socket mains electricity 9Volt power, (not just on battery, but no Jawari sound alas). Luckily I socketed the diodes (it's almost like I expected trouble!) so  I could swap them round - that made the audio even quieter and more hissy... so I swapped them back.  I don't think they're Russian (well they don't move quick enough for that ), but there wasn't any data sheet where I got them from... so I don't know? Touching any part of the circuit or turning the pot leads to some ferocious crackles and pops... What parts do you want me to measure voltage wise?

Sounds like I should order that transformer as well... regarding input it's just guitars (no synths or nuffin', honest guv!) so should I put the 10nF with one leg to the audio in and the other leg to ground?

Hi Mark - OK I didn't know that about octave fuzzes... although I don't own an octave fuzz yet, so maybe i should look at building / buying one? I can see your point about stripboard, but I've never managed to make a perf board circuit work yet and have a higher rate of success with stripboard, so I tend to stick to it in order that I can make something I can play instead consigning more circuits to my junk box of despair!

[anotherjim]

So, that transformer has no centre tap? ...like Stephen says, it has to have that.

[Mark Hammer]

Hi Mark - OK I didn't know that about octave fuzzes... although I don't own an octave fuzz yet, so maybe i should look at building / buying one? I can see your point about stripboard, but I've never managed to make a perf board circuit work yet and have a higher rate of success with stripboard, so I tend to stick to it in order that I can make something I can play instead consigning more circuits to my junk box of despair!

Any port in a storm, as they say.  At least you don't have stacked bins of despair.   

[mitchelr]

So, that transformer has no centre tap? ...like Stephen says, it has to have that.

I have to disagree , I made the Jawari with the transformer I quoted ( without the center tap) which has the right 600:600R and the original circuit.
To my mind sounded pretty good ( see my video)
https://www.ebay.co.uk/itm/Audio-Transformer-EI14-600-600-1-1-Ohm-Isolation-Transformer-UK-Seller/184138688533?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649

[duck_arse]

what's with you and the missing of ground wires? anyhoo - the voltages on the fet drain and source would have been wrongsensical because no current would have been flowing without the ground, so, if you had measured those, you might have thought, funny, should be volts. and then when you posted those readings, we would have thort, funny, I thort. and then it would have been clear, missing ground wire. you can now measure the jfet source and drain, and volts will be there.

as for the input cap - no, one cap end to signal source, one cap end to the 1M gate resistor. you want to block the DC on the gate from the outside world's inherent biases. and if yer diodes were sold to you as D somethings, like D9's, they'd be Russians. if they were sold as AAxxx or OAxxx or even 1Nxx(x), they should be K banded.

and as for the centre tap transformer - I can't argue with someone who has built it working. so I won't argue.

[anotherjim]

You'll get a different/weaker sound without the centre tap - basically, the transformer secondary is then acting like a one-legged antenna picking up a signal from the transformer core. There is no circuit around through the secondary winding for proper transformer operation unless there is a little reverse leakage current through the Ge diodes. And that reverse leakage will be temperature dependant.

[duck_arse]

can that traffo work as intended with a 4 diode bridge at the secondary instead?

[anotherjim]

can that traffo work as intended with a 4 diode bridge at the secondary instead?

In principle, yes (good thinking!) - but it has to overcome x2 diode forward voltage drops in the bridge... but it then has the output of the total secondary winding instead of only half with a centre tap so should have similar performance.

Funny thing is, the sound this thing is meant to produce probably can be got by the leaky loose-coupled arrangement without the centre tap.

I'd expect Moid to approve!
Engineers look away!

[moid]

So, that transformer has no centre tap? ...like Stephen says, it has to have that.

I've ordered a 42TM018 just in case, so if that fixes it I'll go with that option!

Any port in a storm, as they say.  At least you don't have stacked bins of despair.   

Ouch! OK I suppose I should hope to one day have built that many circuits... but that feels like one of those midas touch sort of wishes! At the moment all my fails (bar the spring reverb) fit inside one large shoebox. I raid them for parts occasionally, but other than that I try to pretend that they don't really exist... but I swear I hear them muttering sometimes when I'm soldering late at night, begging me to open the box and fix them, to make them whole again... hmmm too many problems to sort out at work today I think  My mind is going...

I have to disagree , I made the Jawari with the transformer I quoted ( without the center tap) which has the right 600:600R and the original circuit.
To my mind sounded pretty good ( see my video)
https://www.ebay.co.uk/itm/Audio-Transformer-EI14-600-600-1-1-Ohm-Isolation-Transformer-UK-Seller/184138688533?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649

hi mitchelr would you mind me asking if you used a specific layout for your circuit, or was it something you designed yourself? Did you have J201s in through hole form or did you have to use SMTs? Thanks for chipping in, I'm just trying to figure out what I've done wrong!

what's with you and the missing of ground wires? anyhoo - the voltages on the fet drain and source would have been wrongsensical because no current would have been flowing without the ground, so, if you had measured those, you might have thought, funny, should be volts. and then when you posted those readings, we would have thort, funny, I thort. and then it would have been clear, missing ground wire. you can now measure the jfet source and drain, and volts will be there.

as for the input cap - no, one cap end to signal source, one cap end to the 1M gate resistor. you want to block the DC on the gate from the outside world's inherent biases. and if yer diodes were sold to you as D somethings, like D9's, they'd be Russians. if they were sold as AAxxx or OAxxx or even 1Nxx(x), they should be K banded.

and as for the centre tap transformer - I can't argue with someone who has built it working. so I won't argue.

It's because I love to leave something out so that you can find it! I know how much you love solving problems and if I didn't screw things up every now and then I'm sure you'd be bored Actually it might just be that making circuits last thing at night is not really the best thing to do... or maybe I'm just a bit stupid? A little bit from column A, a little bit from column B...

Anyways  I know you love numbers, so I grabbed my trusty DMM and stuck it into transistors that it has never gone into before and attached the ground probe to the power jack ground and the other probe to the transistor and this is what it says:


Drain - 0.166V or 0.003V depending on where the pot is swept.
Source - 0.003V whether the pot is one way or another.

Should I have a guitar plugged in at the circuit? ('cos I didn;t, and these numbers look very low to me which makes me think something is rotten on my transistor... I've ordered some through hole J201s in case the issue is my inability to solder SMTs, so at the end of the week we could swap that out - I socketed that SMT, see I learn )

I will try the cap thing at the weekend, just got back to work today and have read over 1600 emails so far... my brain is too fried to try soldering
The diodes were sold as 1n34As, no claims to any magical East European provenance at all. They do look different to any other diode I've owned; they look like they have a tiny thin hair inside them?

You'll get a different/weaker sound without the centre tap - basically, the transformer secondary is then acting like a one-legged antenna picking up a signal from the transformer core. There is no circuit around through the secondary winding for proper transformer operation unless there is a little reverse leakage current through the Ge diodes. And that reverse leakage will be temperature dependant.

Well we'll be able to compare notes/ fuzzy octaves,  I should get a 42TM018 this weekend and we'll see what happens. Is it possible to explain (in simple language that people like me might understand) what the transformer does in the circuit? I've never built a circuit with a transformer in it before (I've lived a sheltered life really) and I was wondering what it's magical purpose was amongst the rest of it? Does it remove the original pitch note leaving only the upper octaves? Or does it remove the sustain from the note? Or does it enable me to warm my fingers over it if I wire the power up wrong?

can that traffo work as intended with a 4 diode bridge at the secondary instead?

Whoa hold up there duckie me lad, there's no read to go all technical on me! Also, Que?

In principle, yes (good thinking!) - but it has to overcome x2 diode forward voltage drops in the bridge... but it then has the output of the total secondary winding instead of only half with a centre tap so should have similar performance.

Funny thing is, the sound this thing is meant to produce probably can be got by the leaky loose-coupled arrangement without the centre tap.

I'd expect Moid to approve!
Engineers look away!

Oh I love things that work when they aren't supposed to (I am pretty sure that's how my brain functions anyhow). Nobody will ever accuse me of being an engineer (much to the shame and disgrace of the rest of my family, who oddly enough are engineers... I'm the black sheep of the family;  I rejected my roots and went to art college - and look what a fine mess that's got me into!)

Thanks for all the advice chaps, very kind of all of you!

[11-90-an]

https://shopee.ph/2pcs-600-ohm-Audio-isolation-transformer-1-1-Toroidal-with-5-pins-i.92393770.1655444854

Sorry if OT, but will this transformer work? seems to have 5 pins, too...
Also its $1.2 for 2 pcs...
Dunno about shipping though, 

>I was wondering what it's magical purpose was amongst the rest of it? Does it remove the original pitch note leaving only the upper octaves? Or does it remove the sustain from the note? Or does it enable me to warm my fingers over it if I wire the power up wrong?

As far as I know, it's part of isolation? But I'm not really sure. Also really curious about this! 

[mitchelr]

""hi mitchelr would you mind me asking if you used a specific layout for your circuit, or was it something you designed yourself? Did you have J201s in through hole form or did you have to use SMTs? Thanks for chipping in, I'm just trying to figure out what I've done wrong!""

Didn't use anyone's layout , Just put it down on Breadboard from the schematic
In the video the circuit was on breadboard with through hole 201

If I was you I would buy a little breadboard , I ALWAYS try on breadboard first

Good Luck

[Sesh]

Looks very similar to the Death By Audio Octave Clang/White Business Card Fuzz. It has a transformer and does a mean sitar/ring mod-ish octave fuzz sound as well. Opamp instead of J-fet, tho.

The great Fuzzhead has a layout here: http://dirtboxlayouts.blogspot.com/2020/04/death-by-audio-white-card-fuzz.html

Schematic:
https://1.bp.blogspot.com/-7Rk3lzMfkTg/Xpr4tDDQfmI/AAAAAAAACXU/OzdHVuDRX10cr-HjHr6_Wnwa3QoEDDd-QCLcBGAsYHQ/s1600/dba-white-card-fuzz%2BSCHEMATIC.png

[anotherjim]

https://www.allaboutcircuits.com/textbook/semiconductors/chpt-3/rectifier-circuits
The rectifier is behind many octave-up fuzz effects (not all use a transformer). The circuit doesn't easily do clean so there's always going to be some fuzz going on.
There's a lot of projects out there...
https://www.instructables.com/id/Octave-Up-Pedal/

[duck_arse]

what's with you and the missing of ground wires? anyhoo - the voltages on the fet drain and source would have been wrongsensical because no current would have been flowing without the ground, so, if you had measured those, you might have thought, funny, should be volts. and then when you posted those readings, we would have thort, funny, I thort. and then it would have been clear, missing ground wire. you can now measure the jfet source and drain, and volts will be there.

It's because I love to leave something out so that you can find it! I know how much you love solving problems and if I didn't screw things up every now and then I'm sure you'd be bored Actually it might just be that making circuits last thing at night is not really the best thing to do... or maybe I'm just a bit stupid? A little bit from column A, a little bit from column B...

ohhh, are we doing a poll?

Anyways  I know you love numbers, so I grabbed my trusty DMM and stuck it into transistors that it has never gone into before and attached the ground probe to the power jack ground and the other probe to the transistor and this is what it says:


Drain - 0.166V or 0.003V depending on where the pot is swept.
Source - 0.003V whether the pot is one way or another.

Should I have a guitar plugged in at the circuit? ('cos I didn;t, and these numbers look very low to me which makes me think something is rotten on my transistor... I've ordered some through hole J201s in case the issue is my inability to solder SMTs, so at the end of the week we could swap that out - I socketed that SMT, see I learn )

Thanks for all the advice chaps, very kind of all of you!

bad news - something is still wrong according to your readings. the pot track carries the current through the fet from thru the traffo primary, and the cap on the wiper varies the amount of AC bypassing, hence gain. but varying the pot should not change the DC readings. no input signal, read the DC volts to the battery//traffo prim top, then the primary bottom//jfet drain, then the jfet source, I think both fet voltages should be higher - depends on the winding resistance, I guess.

that hair is a whisker.

[moid]

https://shopee.ph/2pcs-600-ohm-Audio-isolation-transformer-1-1-Toroidal-with-5-pins-i.92393770.1655444854

Sorry if OT, but will this transformer work? seems to have 5 pins, too...
Also its $1.2 for 2 pcs...
Dunno about shipping though, 

>I was wondering what it's magical purpose was amongst the rest of it? Does it remove the original pitch note leaving only the upper octaves? Or does it remove the sustain from the note? Or does it enable me to warm my fingers over it if I wire the power up wrong?

As far as I know, it's part of isolation? But I'm not really sure. Also really curious about this! 

Thanks for looking! I've got some on order, but my order is now coming from the USA (it was supposed to be in stock in the UK!) and it's now being held up by a hurricane (which is the first time anyone has used that as an excuse for not sending items on time!) so I maybe waiting until the end of next week to get my parts

Didn't use anyone's layout , Just put it down on Breadboard from the schematic
In the video the circuit was on breadboard with through hole 201

If I was you I would buy a little breadboard , I ALWAYS try on breadboard first

Good Luck

Thanks -  I did try it on breadboard first and it didn't work; but I just assumed that was me being crap at building on breadboard - I've built things on breadboard that didn't function before, but have worked on vero later so I hoped this was going to be one of those circuits. Nope. I've ordered some through hole J201s so hopefully that will be better than SMTs

Looks very similar to the Death By Audio Octave Clang/White Business Card Fuzz. It has a transformer and does a mean sitar/ring mod-ish octave fuzz sound as well. Opamp instead of J-fet, tho.

The great Fuzzhead has a layout here: http://dirtboxlayouts.blogspot.com/2020/04/death-by-audio-white-card-fuzz.html

Schematic:
https://1.bp.blogspot.com/-7Rk3lzMfkTg/Xpr4tDDQfmI/AAAAAAAACXU/OzdHVuDRX10cr-HjHr6_Wnwa3QoEDDd-QCLcBGAsYHQ/s1600/dba-white-card-fuzz%2BSCHEMATIC.png

Thanks very much, I'd never heard of that pedal before and there are some great odd sounds in some of the demos I watched. I'll add that to my list of things to make in future

https://www.allaboutcircuits.com/textbook/semiconductors/chpt-3/rectifier-circuits
The rectifier is behind many octave-up fuzz effects (not all use a transformer). The circuit doesn't easily do clean so there's always going to be some fuzz going on.
There's a lot of projects out there...
https://www.instructables.com/id/Octave-Up-Pedal/

Thanks anotherjim, that instructables link explained it in more simple terms for me; so the transformer and two diodes (1n34A) generate the octave up sound between them. Does that mean the original audio is also mixed in with that, or are they only generating the octave up? I ask because it might be interesting to output them to separate output jacks for mixing purposes... hmmmm


Duck thanks for your post, my brain is somewhat fried at the moment, I will look at the circuit this weekend and try to re test it, hopefully I may have a few minutes to think then!

[anotherjim]

No clean from the rectifier. You can take clean from the feed into the transformer primary winding. I would expect the clean signal to be much louder than the effect sound though.

[moid]

bad news - something is still wrong according to your readings. the pot track carries the current through the fet from thru the traffo primary, and the cap on the wiper varies the amount of AC bypassing, hence gain. but varying the pot should not change the DC readings. no input signal, read the DC volts to the battery//traffo prim top, then the primary bottom//jfet drain, then the jfet source, I think both fet voltages should be higher - depends on the winding resistance, I guess.

that hair is a whisker.

Hello again, some news, but weird news at that! I am assuming that the Primary part of the transformer is the left hand side part (the bit that is closest to the power and transistor), so the Primary Top reads 7.52V, the Primary bottom (connecting to the Drain of the J201) is 7.31V, and this time when I measured the Source of the J201 it was 6.62V!

My through hole J201s have appeared (and my new transformer!) Shall I switch to trying those out? One question about the new transformer; it has six legs, but the circuit needs five. I assume that a P on one side of the transformer means primary, so do I just cut the middle leg off that side? Or solder it in anyway? The new transformer is somewhat larger than the current one, so fitting it into this circuit is going to be interesting!