GGG Tycobrahe Octavia build problems

[MrAstro]

I'm very sorry for asking this on my first post but I'm having some problems with this build and I don't know who to turn to. I'm fairly careful with checking components and soldering but my technical knowledge is pretty poor when it comes to debugging circuits.

I'm building the version of the circuit that uses two 3pdt switches and 2 LED's. So basically I think the way it is meant to work is when both LED's are off the effect is in bypass mode. When the red LED is on, fuzz is meant to work, when the red and green led is on fuzz and the octave effect is meant to work.

So I have the pedal sort of working but it's not working properly. Basically mine is doing the following:

Both LED's are off: The dry signal of the guitar is passed to the amplifier ie. bypass mode.

Red LED on, Green LED off: I hear a distortion/fuzz effect. Pushing the Green LED switch has no or maybe little effect the distortion stays on but the green LED light doesn't come on either.

Red LED on, Green LED On: I can't get the pedal to have both LED's on at the same time - not sure why...

Red LED off, Green LED On: This works but it is just passing the dry signal of the guitar - i.e. bypass mode.

I'm not sure what I've done wrong... I think the wiring looks per the diagram. I've checked the wires to the 3pdt switches and the polarity of the LED's so I'm baffled. It's also my first ever stomp box build.

[MrAstro]

The more I think about this it must be a switching problem of some sort - especially if both LED's will light up independently but not together. I think my brain is going to explode  

The wiring diagram is here: http://www.generalguitargadgets.com/effects-projects/octave/tycobrahe-octavia/

[MrAstro]

I've just gone through through and measured the transistors and input voltage. Sorry I didn't do this earlier - this is a huge learning curve for me.
As you can see there are some differences - I'm not sure what the variances mean but I guess there is power to the transistors.


   
     

Per GGG 9volt Supply: 8.8v Q1 C 5.1      B 4.5      E 2.3 Q2 C 8.8      B 2.4      E 1.8 Q3 C 5.7      B 1.8      E 1.1

Per MrAstro's Build            9volt Supply: 8.8v            Q1 C 4.1                 B 3.4                 E 2.8            Q2 C 8.8                 B 3.1                 E 2.3                        Q3 C 4.7                 B 2.3                 E 1.6

[kaycee]

Hi MrA,

It can be frustrating ironing out a problem, so stick with it. A good bit of determination will probably get you there in the end and you'll learn a fair bit on the way.

It's almost certain that your problem is in the switch wiring. If I understand it correctly The LEDs will light independently of each other when wired right, so you could have the octave light on but the pedal in bypass, it's completing the circuit that lights the LED not the signal.

Your going to need to use a multi meter to check the continuity (connection) between the switch lugs and their destinations. If you haven't got a meter and your going to continue building pedals it's an essential tool.

I can't see why a 3pdt is used for the octave on switch, a DPDT would cover it??

[MrAstro]

Thanks Kaycee - I can get the octave light on while it's in bypass. I can also get the fuzz light on with the octave light off. What I can't get is the fuzz and octave light on together ie. if I turn the fuzz light on, the octave switch doesn't do anything to the octave LED.

It's just so weird...

Both of the LED's light up really well when they are activated independently.

I have gone over the diagram so many times and I've been testing continuity. I just don't know what's going on... I think I should stick to my day job :-) but it was a fun process. It would be nicer if I could get the rotten thing working right though :-)

[MrAstro]

I might be closer to an answer...

I was given a red LED and green LED in the kit but the green one was dead when I tested it. So I bought a replacement at my local electronics store.

I just was reading that you can't run two LED's in parallel because they might be mismatched or something like that so out of curiosity I ran the new blue led and it had a voltage of 2.8v dc. The red one when it glows measured 1.8v. So maybe one LED is sucking the life out of the other LED so to speak?

Here's the thing - when both LED's are on and the red one glows and the blue one doesn't the blue one measures 1.8v - the same as the red one - but maybe that's not enough juice to make it glow?

Could this be the issue???

I did buy another LED as a spare but it is the same colour (blue). Would the circuit safely run two led's at 2.8v instead of presumably the originals at 1.8v???

[anotherjim]

Yes!
That wiring can only work if the LED forward voltages are matched. This is unlikely to happen with different colors. A quick fix would be to separate one of the 2 LED wires from the connection to the 1k and give it it's own 1k to the +9v.

That still leaves something wrong at Q1. Note that for NPN transistors, the base voltage should be about 0.6volts higher than the emitter voltage - which, in all the other readings, they are. Q2 & Q3 readings will hopefully correct themselves after fixing Q1.

[MrAstro]

Thanks so much for the great advice, I'll try and do something about those LED's. I get the general gist of what you are telling me - I just need work out where to connect this new 1k resistor.

Re: the transistor are you just saying that I probably just need to replace the Q1 transistor and with luck that will sort out some of the odd readings on the other transistors? Or perhaps the issue at Q1 is other components leading to a strange reading at Q1?

I really appreciated your reply - I just don't have that sort of knowledge.

[MrAstro]

That still leaves something wrong at Q1. Note that for NPN transistors, the base voltage should be about 0.6volts higher than the emitter voltage - which, in all the other readings, they are. Q2 & Q3 readings will hopefully correct themselves after fixing Q1.

I'm working through your advice but I just noticed something.

The Q1 voltage I listed shows the base voltage in my build is exactly 0.6v higher than the emitter so I'm confused about which transistor has the problem.

[EDIT] OK - I can see now that Q1 is a PNP - so you are saying that Q2 and Q3 are OK and Q1 differential is nothing like the GGG value Base/Emitter differential of about 2.2v

[anotherjim]

Oh sorry - my bad. I got the readings crossed over -
If Q1 is PNP, then what you have notated as the collector in the values must be the emitter - for PNP the base voltage will be about 0.6V BELOW the emitter.

Can you link to the schematic for the board? You values could just be out from the GGG owing to tolerances or some resistors swapped.

Solder a 1k resistor to the anode of one LED and wire directly to +9V at the top of the board. Insulate the resistor and joint with sleeving.

If you want the resistor on the PCB and you have a mini-drill, there's just room to drill 3 holes for the resistor and wire at the top left of the board just above the copper trace carrying +9V.

[kaycee]

Yes!
That wiring can only work if the LED forward voltages are matched.

Wow! Crazy idea.

Doesn't the octave off just lift the diodes to ground connection? That could be done with a DPDT sharing the ground for the LED on the other set of poles, pretty sure that's how it did it last I made one of these? I did see somewhere the idea of taking the fuzz output from before the transformer for the octave lift idea, never tried it though myself.

[MrAstro]

Tonight I rechecked all the component values and everything seems like it's in the right place.

I also put a new battery in turned the effects off (i.e. bypass mode) with no LED's on. Turned the pots to maximum and rechecked the voltages.

I ordered some new transistors and they should arrive in a day or two so I guess I should replace Q1 as advised?

I'm also going to try and order a matched pair of replacement LED's - I'm still sourcing those. I am also going to add the extra resistor when I get them.


 
     

Per GGG 9volt Supply: 8.8v Q1 C 5.1     B 4.5     E 2.3 Q2 C 8.8     B 2.4     E 1.8 Q3 C 5.7     B 1.8     E 1.1

Per MrAstro's Build Test 1           9volt Supply: 8.8v           Q1 C 4.1                B 3.4                E 2.8           Q2 C 8.8                B 3.1                E 2.3                     Q3 C 4.7                B 2.3                E 1.6

Per MrAstro's Build Test 2           9volt Supply: 9.32v           Q1 C 4.28                B 3.62                E 2.94           Q2 C 9.32                B 2.94                E 2.37                     Q3 C 4.97                B 2.37                E 1.72

[kaycee]

Okay, you won't find people selling matched pairs of LEDs, two the same colour should have near enough the same forward voltage, or you can get batch of 10 and use the diode function on your meter to match their forward voltage.

Myself, I'd take a look at the GGG wiring schematic for using a toggle switch and wire your foot switch 3pdt as that, then get a second DPDT foot switch, it will have 6 lugs. You wire the octave engage on one side using two of the three lugs, the pole (center) and a throw, upper or lower. You then wire any bloody colour you like LED to the other side of the switch

Run a wire from your power jack board power connection lug to LED positive. Run LED negative to the same throw as on the other side as your connection for octave engage. From the other pole run via a suitable value limiting resistor a wire to ground. Now when you engage octave you also connect the indicator LED to ground and complete the circuit.

The voltages problem is above my pay grade I'm afraid, but the switch I can help with

[anotherjim]

Ok - Q1 may be right for your build. It's actually meant to have collector and emitter swapped. Normal PNP operation would have the emitter on the top towards the positive 9 volt power supply and the collector to the 0 volt side. The schematic you showed me has normal connection of Q1, bu, the build instructions and test voltages are for the "reversed" connection with collector toward positive supply.
Reverse will "work" but with less gain - you can try it either way as the instructions mention..

Your voltages at Q1 only looked bad to me because I didn't realize it's a reversed PNP. Yes, they are off compared to the reference readings in the instructions, but yours are (I think) still in operational limits.

Play with it!

[MrAstro]

Thanks Jim and Kaycee. I am absorbing all of your feedback which is much appreciated.

I'm waiting for some bits and pieces as mentioned before. I have some new transistors coming as well.

When the parts come I am going to desolder the transistors. Then I'm going to install some sockets and socket the replacements. Then I'm going to test what was in there previously just out of curiosity.

I actually have a proper transistor test function on my multimeter but I only tested the transistors using the diode testing function before I installed them so I'll be interested to see what happens when I have them to retest.

[MrAstro]

OK Guys - I think I've solved the transistor voltage mystery.

Reversing Q1 which is optional (but not the way the GGG component layout is currently) changed the voltages significantly.

What I think this means is the reference voltages were for the previous version of this circuit (which GGG used to sell) and they haven't been changed for the new circuit layout with the Q1 in it's currently published orientation.

Do you guy's agree? Compare the GGG reference voltages to the reversed Q1 transistor voltages.



 
     

Per GGG 9volt Supply: 8.8v Q1 C 5.1     B 4.5     E 2.3 Q2 C 8.8     B 2.4     E 1.8 Q3 C 5.7     B 1.8     E 1.1

Per MrAstro's Build Test 1           9volt Supply: 8.8v           Q1 C 4.1                B 3.4                E 2.8           Q2 C 8.8                B 3.1                E 2.3                     Q3 C 4.7                B 2.3                E 1.6

Per MrAstro's Build Test 2           9volt Supply: 9.32v           Q1 C 4.28                B 3.62                E 2.94           Q2 C 9.32                B 2.94                E 2.37                     Q3 C 4.97                B 2.37                E 1.72

Per MrAstro's Build - Reversed Q1 Resistor (Optional)           9volt Supply: 9.28v           Q1 C 5.41                B 4.79                E 2.43           Q2 C 9.28                B 2.43                E 1.88                     Q3 C 6.14                B 1.88                E 1.25

[anotherjim]

Well, how does it sound?

Your "reversed" readings now line up very well with the GGG ones, given that you have a  higher supply voltage.

I suspect performance of Q1 might well vary according to individual transistors when the C and E are reversed. Q1 & 2 are directly dependent for bias voltage from Q1. Q1 alone has it's own bias network of resistors to it's base, so Q1 rules all. With a 9volt supply, Q1 B "should" be about 3.4volt (if my maths is right) for the values of resistors used. You pretty much had that in your first test with a lower battery!

[MrAstro]

It doesn't sound too bad when I use it in fuzz mode but there is a bit of RF and crackle... I'm hoping the enclosure might fix that when I mount it. I'm not sure about the octave mode because I have the wrong LED on it. It doesn't sound much different to the ordinary fuzz mode though but I'm trying not to use it until I fix up the LED's and mount everything in the effect box.

LED question...

I can't source low current (1mA) Vf 1.6 replacement LED's very easily in Australia. The original parts were 638-HLMPD150A  and 638-HLMP4740 from Mouser. They are cheap as chips but the postage would be $40  

So I was hoping to just buy some 20mA LEDs.

Can I just change the single the 1k resistor to say a single 390ohm resistor and would it work with two identical 20mA LED's?

[anotherjim]

Octave effect can be elusive. It's much more obvious when fed with quite a dark tone with a strong fundamental. Try neck pickup with tone turned down and play around 12th fret.

Just because a LED is 20mA doesn't mean it will actually take that much current for a reasonable brightness. The existing 1k might do for both of them assuming they are same type.

Don't somebody like Jaycar carry high efficiency LED's in OZ ?

[duck_arse]

geeze, mrastro, why dincha say you was in au?

don't worry about the maximum current ratings on leds, except to stay well below that figure. red leds have a Vf of about 1V6, it's part of the mechanish of them being red. same goes for the other colours, the internal chemistry determines the light and the Vf.

what leds do you have, water clear types or the diffused, coloured lens type? the diffused are generally much dimmer, hence they need a lower value resistor and more current. waterclear are usually high efficiency types, and can be as bright with resistors of as much as 10k~15k.

as much as I hate to say "jaycar", just get their cheapest HE red 5mm. use a 10k resistor in series with the led. if 10k is wrong, you know to go higher or lower, depending which way wrong it is. for a diffused type, maybe 1k instead. each led has its own resistor and mrastro has no led problems.

[edit: ] I woulda had this up before anotherjim, but my session timed out ....