My First Build - Noob Advice Required about components.

[steveyraff]

Thanks everyone. Yea, really proud of this. I think its the first thing I've ever really made properly from scratch - can't wait for the next build! I actually already did build a Super Hard-On as a little side project that was supposed to be much easier. It sounded like ass though so I scrapped it lol

Anyways - I gig every weekend and I can't wait to take this out on Saturday night and give it a run through its paces. Woohoo!

Cheers for the tip with the Alpha Pots - will make sure to snip them off next time.

[steveyraff]

Here's a video demo for y'all that I recorded in my studio.

http://youtu.be/HRra4xyzuik

[Jdansti]

Sounds great!  You're getting a good range of sounds from it.

[steveyraff]

Sounds great!  You're getting a good range of sounds from it.

Thanks very much John! Very happy with it!

[Kipper4]

Sounds good Stevie. Glad it all worked out man

[steveyraff]

Hey guys,

Digging up this old thread again - but thought I'd keep any queries related to it all in the same place.

Got a lot of musician friends asking me to make them one after hearing my own live. I've made about 3 or 4 of them now. There was a few off board mods and various other layout changes which I decided to put into a new circuit layout design of my own.

So, for the last few weeks I've been learning how to use Eagle. I successfully designed my own layout that incorporated all my changes. Got the first PCB printed out 2 days ago and just finished building it today.

Few issues. Firstly, I was delighted it actually worked. I presumed if I got the layout wrong, it wouldn't work properly at all.

There are a few errors and a few questions I'd like to ask you all about them. Firstly, I think I made my pads and tracks slightly too thin. Its very hard not to over heat them while soldering and two or three actually lifted up off the board which caused a few problems. At the start I was getting intermittent signal, I discovered one of these loose pads was the culprit.

Not its all patched up and working - but I A/B'd it to my prototype and there is definitely not as much distortion/fuzz saturation. The other ones I built previously seem to have more gain.

I am pretty sure my layout is sound, I have checked and double and tripled check it for errors. Here's what I am thinking - is it possible that over heating a capacitor can have this effect? I only ask as I worried about this with one or two of the caps that had pads lifting off the board. It meant a lot of messing around with the soldering iron trying to get them secure and they were getting really hot at times. I am unsure if a damaged cap would negatively affect sound like this, or if it would lead to the circuit just not working at all if that had happened? ? ?


I checked my diodes and transistors and they all seem fine. The only thing I had problems with were some of the caps. I also had issues with loose potentiometer hook up wire joints. I made the pads too close together and they lift and joined together and got all messed up. I sort of rectified it so that they are now secure and separated, but I wouldn't know if this could have anything to do with my distortion issues?

Any words of advice appreciated. I will try to amend my layout so that it doesn't happen on future builds.

[mth5044]

It's hard to tell what components might effect what if you don't supply a schematic.

Since you are having troubles overheating, you might want to practice soldering on empty perf before you start making pedals to give or sell to people. Wires, leads, extra components, etc. Are you etching these PCB's yourself or having them fab'd?

[steveyraff]

It's hard to tell what components might effect what if you don't supply a schematic.

Since you are having troubles overheating, you might want to practice soldering on empty perf before you start making pedals to give or sell to people. Wires, leads, extra components, etc. Are you etching these PCB's yourself or having them fab'd?

Well this is about the 12th build now - most of my builds are 100 + components. I've had a lot of practice. My soldering is good now. The only reason I was having problems over heating it was because of the pads lifting and then I was having to spend a lot of time on one component trying to get the thing soldered properly. The pads are lifting very easily because I've made them too small, much smaller than i normally do. This means the drilling is taking most of them out. Its an update for my next PCB print out.

I'm getting them professionally fab'd.

[steveyraff]

Sorry guys, don't mean to spam but I just want to bump this up again to see if anyone has any ideas as to why the latest build doesnt seem to have as much distortion as the previous builds? I A/B'd it to my prototype and it definitely seems a little weaker.

I presume its not the diodes or transistors as I checked their solder joints and everything is clean. I also presume its not going to be any of the caps that have a ground leg as they wouldn't allow any signal past at all. So could it be one of the other caps ? ?

Confused - not great with troubleshooting. Cheers.

[duck_arse]

stevey, nice to see you back. how the hell did we get this thread to 13 pages?

can we see your board layout please? there is literally nothing in eagle you can't set/reset to what you want/need. I only use it for circuit diagramming, but I have done some pcb's many years ago. you might be better to set the pad sizes (and track widths) as wide as can fit. the hole size can also be changed.

and not meaning to be rude, but there are some problems evident with your schematic-ing. I can't see the component values (not big enough) to comment on the workings, but I can see some parts are twisted about themselves, there are junctions where they shouldn't be, and something weird about your resistors. why do they have those brown lines either end?

the best way to repair lifted pads is to chuck the bad bit; push the component flat on the board so it can't shift about, wrap a bit of fine wire, like a resistor cut-off, around the poked through leg as a replacement for the pad, and trail it to the nearest still solid bit of track. solder carefully, just enough, no more.

[steveyraff]

stevey, nice to see you back. how the hell did we get this thread to 13 pages?

can we see your board layout please? there is literally nothing in eagle you can't set/reset to what you want/need. I only use it for circuit diagramming, but I have done some pcb's many years ago. you might be better to set the pad sizes (and track widths) as wide as can fit. the hole size can also be changed.

and not meaning to be rude, but there are some problems evident with your schematic-ing. I can't see the component values (not big enough) to comment on the workings, but I can see some parts are twisted about themselves, there are junctions where they shouldn't be, and something weird about your resistors. why do they have those brown lines either end?

the best way to repair lifted pads is to chuck the bad bit; push the component flat on the board so it can't shift about, wrap a bit of fine wire, like a resistor cut-off, around the poked through leg as a replacement for the pad, and trail it to the nearest still solid bit of track. solder carefully, just enough, no more.

Hey thanks  Its just the component package I was using I think. In the layout, there is nothing apparent that seems odd about the resistors (to me anyway! lol).

I mean, it should be the exact same as any other build I've done exact the layout is moved around a bit to make it easier. I started out with a layout that was an exact copy of the one I've been using up until now, and just changed a few parts around to make life easier. Nothing has actually changed, which makes me wonder if my loss of distortion is to do with the messing around I had with a few of the caps, or the fact that I repaired a broken track at the "Night" control with a piece of wire to bridge it?

CAPS:

C1
100n
C2
1n
C3
4u7
C4
22uF
C5
47n
C6
10n
C7
4u7
C8
47n
C9
470pf
C10
47n
C11
1n
C12
470pf
C13
100n
C14
470pf
C15
100n
C16
4u7
C17
10n
C18
100n
C19
22uF
C20
22uF
C21
220uF
C22
220uF
C23
100uF
C24
4u7

[mth5044]

What about all the things that aren't connected? Especially R28 on Q5. It is connected to an small pour of what I assume to be ground, but it is an island: isolated. Which that airwave to C23 is telling you. Same thing with the cap under the volume pads. And what about the other airwaves?

[duck_arse]

^ that's what I was going to say. looks like V3 is in the offing.

[steveyraff]

lol oh yea... erm, those. hah. Sorry guys, I know I must continually irritate wasting peoples time on this thread with my rookie errors, but I only learn when these things are pointed out to me - so I really appreciate your patience. It helps me understand. Yea, I seen these islands at the time and it crossed my mind that it couldn't possibly be enough to earth it - guess I should have went with that theory! Would something like that cause the problems I refer to with underwhelming tone/distortion ? ?

Also, what do you mean when you say 'airwaves'? Again, sorry for all the rookie questions. Do you mean the yellow lines that tell me something needs to be connected? I think some of them occurred when I disregarded resistor orientation - but they are still correct regardless (I think! :/ )

So, if the ground islands are the problem, is there a way for me to just quickly test to make sure? ie join the island up to the ground plane with a jumper wire? Or is this a ridiculous suggestion?

[duck_arse]

the yellow lines indicate that something connected on the schem still needs to be routed and therefore connected on the pcb. a redesign is needed on your pcb. you might be able to get normal operation if you add wire links betweeen the yellow-lined parts.

[edit: ]actually, don't just wire between yellows. around Q5 I can see they are crossing over. it might be easier to open the board in eagle and move the parts around until the connections are a bit clearer.

[steveyraff]

the yellow lines indicate that something connected on the schem still needs to be routed and therefore connected on the pcb. a redesign is needed on your pcb. you might be able to get normal operation if you add wire links betweeen the yellow-lined parts.

Hmm ok. I knew that, but I thought it might have just been something I did wrong in the schematic as the layout visually appeared the same in those areas as the previous working layout. Thanks guys. I'll keep trying.

Maybe this is why I don't have appropriate amount of distortion?

[mth5044]

'Earth it' doesn't really mean anything. All grounded items must connect together and, in this case, connect to the negative terminal of the power supply. R27, R28 and C10(?) are not grounded and will give you problems, especially since R27 is used to bias Q5.

Airwaves are the yellow lines. IF your schematic is correct (IF IF IF), then you can consider the airwaves on the layout correct. YOU MUST CONNECT THE AIRWAVES. Resistor orientation isn't a thing, so forget about that. If you still have airwaves on the layout after you've finished (unless it's something intentional, which it usually isn't) you haven't done the layout correctly. It appears you haven't done it correctly and it needs quite a bit of work.

Some things appear to be wrong in your schematic as well. What library are you using for those resistors? For some resistors (schem is too small to make out some of the labels, could you post a bigger one?), it looks like you tried to flip the component around without disconnecting it, such as the R in the top left schematic going to VB. You can see the green line through it, which means the trace is connecting to both sides and causing problems. Usually when you see a green dot on a single connection, something isn't right. Even less right when there is a green trace going through the component connecting the leads.

You can try to connect all the islands together. From what I see there are two of them. But that might not help you with all the other airwaves. I'd suggest starting from scratch. Look at Gauss Markov's Eagle tutorial and then start again. Those resistor packages also look strange with their pins, so maybe get Gauss's or Madbean's library.  

[duck_arse]

and R40 wants to have each of its ends connected together, so you have some problems with your schematic as well. looks like starting from scratch.

[mth5044]

One of each clipping diode is being bypassed in the schematic as well since the green lines are going through them. Too hard to tell in the layout, but it could cause clipping problems. The other components bypassed are the LED and the resistor in the power section.

[steveyraff]

Ill try to get a bigger schematic. Here is a pic of the layout which DOES work. This is what I've been comparing to. Its why I thought some of the airwaves could be disregarded. ie - you said R40 wants both of its ends connected, because the airwaves say so (probably an error on my schem?) - but if you look at this working layout, they are not connected?