Welcome to Solid State Guitar Amp Forum | DIY Guitar Amplifiers. Please login or sign up.

March 28, 2024, 06:26:09 PM

Login with username, password and session length

Recent Posts

 

DIY amp directions

Started by Lauren, August 25, 2014, 07:10:38 PM

Previous topic - Next topic

Roly

Quote from: galaxiexit doesn't match our circuit boards exactly. (sorry Roly  :))

Wah?  I screwed up?  Oh noze... must correct.

Huh?  Just checked against your board pic and they look he same to me.  One side of the board has two rows of donuts, the other side three, my drawing shows the amp built on the 3-row side of the board, vertical strict, horizontal ad lib.

Going with a different layout is an interesting development, but meh, we'll see how it works out; still think it's a good idea to roll both builds into one thread.

Quote from: galaxiexEdit; took the day off work today to play and have fun...  :)

Now there dedication for ya.  I took the day off too (but wait ... I don't have a job:o   I'm retired but I often say that "I should get a job so I can have some time off", I still get a stream of musicians at my door with some ill waif under their arm "I know you're retired, but I was hoping...", and I'm a sucker for a bird with a broken wing).

Quote from: LaurenI am a bit intimidated by this whole thing

Look, your have set yourself a birthday deadline which doesn't help considering you haven't done this before, BUT the build itself is only a few bucks worth of parts (and a lot of your time).  If the worst really came to the worst you could just dump your first attempt and do a second and, as we say here in Oz, "there is no sheep station hanging on it".  This is exactly what Jerri was talking about, "giving yourself permission to fail", it's okay, this is just a fun thing, it's not going to lead to the collapse of Western Civilisation if you happen to smoke a chip or something; after 50 years at this I can assure you it goes with the turf, humans make mistakes.  And as any experienced musician will tell you, we all make mistakes, the important thing is how we recover from them.

{I'm still reading down the (exploding!) thread commenting as I go, so forgive me if I double up}

Quote from: Laurenbasics

Some terminology for more effective communication.  The plain side of the board is called the "component" side, the other, obviously I hope, is called the "copper" side.

The copper itself is called a "pad" or "doughnut" for single holes, and "traces" where they connect more than one hole. (but yes, they do have a layer of gold plating to stop the copper oxidising and keep it easily solderable, about a molecule thick)

The layout is strict vertically, but needs to be positioned horizontally on the board to taste/fit.  There are 23 full height vertical (numbered) columns of holes with a shorter one at each end between the mounting holes.  I suggest that you skip the short column and one long column and call the second longest column your first, most left-hand, column - that will be the column where your power and input wires go, "1" on the layout = "3" on the board.

Quote from: LaurenWhat's the difference between the individual circles and the ones grouped by the gold bands? 

That's it, some are interconnected and some aren't.  I haven't made a lot of use of that in my layout, treating it mainly as if they were all isolated.

Have a look at my layout and note where C3 is.  It runs between pin 7 of the IC and the ground "rail" or trace.  You will see that it passes over a vertical group of three.  Now if it turns out when you are actually stuffing components into the board to be easier/neater to take C3 from pin 7 down to the top of that triplet trace, then link from the bottom of the triplet to the ground rail, then do it.  Electrically it's the same, just physically a wangle, or easier.

Similarly with R3 and C4 where the triplet is used as a tie point for the connection between them.  Again this is a physical consideration, you could just twist the leads of R3/C4 together and solder them above the board.  This is sometimes called "aeroboard" construction and is a bit inelegant but sometimes required if you run out of space (which hopefully won't happen to you), not professionally perfect but still perfectly valid.

One of the first things you need to do is mark where the real links go (since in the limited programme I used I had to use blue links to map were the copper is).

Links

There are no horizontal links.  The horizontal "links" shown are only to illustrate which components are actually connected.

From left to right (using the coordinates on my layout) there are real links between;

1h-1i, 6h-6i, 7h-7i.  These connect the ground rail to some of the triplets.

6f-7e (diagonal).  Grounds pin 3 of the IC, input reference or "cold".

10g(or 10h)-13c.  Chip output to speaker.

11e-11j.  Battery +ve to the IC pin 6.

As long as connections get to where they are supposed to go (and not accidentally to somewhere else) it matters not how they get there.  (there are exceptions, but not in this case)

{There is a style of prototyping construction called "birdnesting" where you don't have a board at all, just tack solder all the bits together full length in what quickly becomes like a birds nest having a bad hair day.  This obviously isn't robust, but as long as all the connections are correct it will work just as well.}

Quote from: LaurenEtiquette question:  Should I wait to post a new question until someone addresses the old one?

NO!  Keep 'em coming.

Quote from: galaxiexThere are no dumb questions!

None of us were born knowing anything at all about this stuff.  There are no dumb questions - the only "dumb" question is the one you failed to ask.

Quote from: Laurendo I have to move things over in any way?

Yes (as above).  The letter rows must stay the same but the number columns need to be shifted over a little bit to get away from the edge of the board.  Assuming the short column at the end is "1" then my layout "1" will be "3" on your board, just add two to the column numbers.

{
Quote from: joecool85What a great thread.

Sorry, I should have asked before making the "executive decision" to roll two build threads into one, but I'm glad you agree it's a good idea.}


Lauren - I'm still a little worried about your soldering skills and how to brush them up before you actually start on the board.  Have you done any before, none, some, a lot?  Have you watched any soldering "how to" vids?

The classic method is to take some fine tinned copper wire (single strand, no insulation) and make up a "drink coaster" matrix of 10x10, then solder all the joins.  Most trainees are making good joints by the end (and even the most hopeless generally get it by the end of the second  ;) ).

I hesitate at launching you off to RS again for tinned copper wire, they'll probably only stock and try and sell you a 100 metre roll when you only need a metre or so (sorry, Aussie is metricated, but then so is much of electronics).


For your iron you need some sort of stand or rest to prevent it getting out of control and burning a hole in something (e.g. you!).  A large ashtray or similar makes a good improvised rest.

You will need a damp pad for wiping the iron tip on before you make a joint.  There are specific heat resistant sponges but I find a pad made out of some old linen material works well; just get it good and damp before you start (it should sizzle slightly when you wipe).

Solder (and dross) tends to accumulate on the iron tip and you need to shake/flick this off before you wipe, therefore I have my soldering iron in a cardboard box, opening frontwards, so it forms a "booth" around the iron and catches solder splashes.

You will quickly find that your sidecutters are your favorite tool.  They allow you to form leads using them like pliers, prior to clipping off the excess.  Now about them offcuts; when you clip a component lead the offcut normally takes off for parts unknown at the speed of light, so when you actually clip a lead off you should direct the offcut away from your face.  Firing it into your "soldering booth" is one way of dealing with them, or you can just hold the offcut so it doesn't fly.

Similarly; cover yourself up.  I got up one hot summer morning and decided to do something while I was still in my dressing gown, and being hot I didn't have it wrapped around me, so when I accidentally flicked a pea-sized glob of molten solder ... you don't have to ask where it landed.  Very memorable - learn from my pain.

Before I start I generally use a felt-tip pen to mark all the holes on the component side where a component will go through.  If you just poke the tip into the hole and rotate it a bit it leaves a nice dot around the hole.  Helps keep you oriented.

Next is galaxiex's excellent suggestion that you try poking all the components in their places before attempting any actual soldering, "trying for fit".  This will expose any size conflicts (which you will somehow have to wangle to compensate for) and give you an idea of how it should look when finished.  I've tried to allow ample spacing so hopefully that won't be needed.

Polarisation:  there are some components that care very much which way they are connected, the IC, the FET, the electrolytic capacitors, and of course the battery.  The IC has its D-mark, the FET is D-shaped, and the electrolytics have a black band marked "-" (minus) to show their negative leg.  On the drawing this is shown by a small darker "D" on the negative side.

There is no danger here of getting zapped, anything exploding, burning the house down, or electroluxing the cat.  The only danger is that you might lock yourself in the bedroom in frustration.  Protip; when it all seems too much, take a break, have a coffee, even sleep on it, and things will look different when you come back fresh.

Building stuff is a bit like mountain climbing, a bit of a slog with no end in sight, then suddenly you have the world at your feet, your creation is making beautiful music and you feel great, can't stop smiling - or as I have been known to yell after knocking over a particularly difficult repair "So who's the King of the castle then?  I'm the King of the castle!".  It's a real charge and I think why people find it so addictive.  "I built that!", and in a mass produced throw-away world that's one hell of a statement.

If you say theory and practice don't agree you haven't applied enough theory.

galaxiex

#46
(Pedantic mode ON)  :lmao:

Reading the schematic... it's easy! Really!  :)



The big triangle in the center is our chip, LM386.
The numbers around it are the pin numbers. 12345678

The vertical zig zag lines are resistors.

The vertical zig zag lines with an arrow pointing at the zig zag is a "pot" or potentiometer, or variable resistor.

The -|(- or -||- symbols are the capacitors,
The -|(- with a plus + sign indicates an electrolytic capacitor with positive indicated,
but you will notice your *actual* part has the negative lead marked!!??
Whats up with that? I dunno, it's just the way it's done, so I live with it.
Watch the polarity marking closely on parts that have them.
They are important!  ;)

The values of each resistor and capacitor are shown next to each part.

The speaker is at the RH side, big sort of triangle thing with plus and minus indicated.
+ and - Not so important here, but important when connecting multiple speakers.

The MPF 102 JFet transistor is the circle at top left with funny lines inside the circle.



And finally Power is indicated at top left as +9V. That is where our 9 volt battery positive + connects.

All across the bottom of the schematic are those funny triangle shaped horizontal lines.



That is the "ground" symbol and where our 9 volt battery minus - connects.

And... lest I forget, all the lines connecting all this stuff are wires AND/OR solder pads on a circuit board.

More pedantic stuff...
Normally the components in a schematic are numbered such as R1, R2, C1, C2 etc.
U1 for an IC (or U2, U3 etc if there are more of them...

This schematic does not have part numbers so we will go by the part values.
There are not so many parts so it should not be hard to figure out.

If I say "the 220uF cap" you can look at the schematic and see right away which part I'm talking about.

Edit; posted while Roly was posting too!
(high pitched maniacal laughter) Ha ha toooo much.... information overload!!!!  :lmao:
If it ain't broke I'll fix it until it is.

galaxiex

#47
Quote from: Roly on September 12, 2014, 01:13:28 AM

Wah?  I screwed up?  Oh noze... must correct.

Huh?  Just checked against your board pic and they look he same to me.  One side of the board has two rows of donuts, the other side three, my drawing shows the amp built on the 3-row side of the board, vertical strict, horizontal ad lib.

Arrrgh! I don't/didn't see it. the "stripboard" copper traces mess with my eyes....
Sorry I find it confusing...

Must be a bit dyslexic...  :duh

Oh! Wait! I see it now! It's (the chip) turned sideways and you are only showing the "top half" of the board etc...

( I gotta remember to look at stuff for awhile until I "get it" before posting about it...)  ::)

Ok, that layout is much more compact than what I planned on doing.
No matter.  :) It's all good.  :)

Sorry Lauren for my earlier failure to see that the boards are the same.  :-[

By all means use Roly's Lauren's Ruby layout. That makes it yours! Lauren's Ruby!  <3)

(not that you need my permission.. I'm just rambling/babbleing... it's late, I'm tired, I should go to bed, gotta work in the morn...)

See Ya!  :)

If it ain't broke I'll fix it until it is.

Roly

Quote from: galaxiex(Pedantic Explicit mode ON)

Fixed.


Just something to note depending on which exact FET Lauren is actually able to obtain, the MPF102 leads go Drain/Source/Gate (my layout) while the 2N3819 goes Drain/Gate/Source.


Quote from: galaxiex"stripboard" copper traces mess with my eyes

Wanna know something?  Me too!  I avoid using Vero/stripboard because as soon as I look at the copper side I get instantly disoriented.  I'd much rather use "doughnut" or "dab-board" than stripboard for those projects.


Quote from: RolyThe letter rows must stay the same but the number columns need to be shifted over a little bit to get away from the edge of the board.

Sorry, sorry; this is one of those things that are "obvious" after lebenty-leben years that I should have explicitly stated somewhere.

Building stuff is a continual process of finding work-arounds for things that haven't turned out as planned.  I don't think you should have any worries, but if, for example, you run out of room when you get to the right-hand end we can always busk it by placing components on the spare half of the board if needed.


Quote from: galaxiexthat layout is much more compact

...and here's me thinking how "open" it is.   :lmao:

This;



...is what I'd call a fairly tight layout.

{Used to work for a joint making hand-held radios, up to eight transistor stages on a 3 inch by 4 inch PCB, all the resistors standing on-end and everything packed right in there, tight as...}

If you say theory and practice don't agree you haven't applied enough theory.

Roly

If you say theory and practice don't agree you haven't applied enough theory.

galaxiex

So tonight when I get home from work...
I will do a mock-up/layout of the board following Lauren's Ruby picture.  :)

... and post pics of course...  :)
If it ain't broke I'll fix it until it is.

Roly

{  ...and I reccon I could get eight on there with a single preamp  :duh  }
If you say theory and practice don't agree you haven't applied enough theory.

galaxiex

#52
Show off...  ;)  :)

(... but I notice they are not cascaded to have 4 gain stages in one amp,
but rather, 4 separate amps on one board... with 4 separate speaker outputs... and 4 separate inputs... 4 volume controls...)

Hmmm, a mini powered  4 channel mixer beginnings? Nah, it'd never work...  :loco


Quote from: Roly on September 12, 2014, 01:17:16 PM
{  ...and I reccon I could get eight on there with a single preamp  :duh  }

Eight!!!  :o with one preamp!!! Now you got somethin' cowboy!!!  :lmao:

But seriously, next post will be of Lauren's Ruby test placing parts.  :)
If it ain't broke I'll fix it until it is.

galaxiex

#53
Ok following Roly's latest drawing of Lauren's Ruby here is how I started.

Note nothing is soldered yet.

Pic LR 001

I've got the board, and the picture printed.
Marked the board with the + and - to help remember which is which.  ;)
Notice there is also a "1" mark at the top to help me remember where Pin 1 is in relation to the socket.
If you don't have a socket you could just place the IC on the board.

Pic LR 002 to 006

Here I have placed all the capacitors to see how it looks.
Looks good!  :)

Notice in pic LR 006 you can see the - minus sign on the 2 electrolytic caps.

On to the next...
If it ain't broke I'll fix it until it is.

galaxiex

#54
Pic LR 007

I have a PF5102 JFET that has the same pin out as the MPF 102.
From the left with the flat side facing you and the leads pointing down. 1. Drain 2. Source 3. Gate
You see on the data sheet it says "N-CHANNEL SWITCH"
No big deal, this JFET can still function as a preamp, it doesn't have to be used as a switch.
The designers may have intended for it to be a switch, and optimised it for such, but it will work just fine as a preamp in this circuit.
I always like to get the data sheet for whatever device I'm working with, for the proper pinout.
It's important to get it installed correctly!

Pics LR 008 to 010
Different views of how/where the JFET is placed.
If it ain't broke I'll fix it until it is.

galaxiex

#55
Pics LR 011 to 013

Here I have placed the 3 resistors.

Uh oh, R3 is not in the exact same place as the drawing.
No matter, this is what Roly was talking about several posts up...
Things don't always go according to plan.
This won't matter, we can still make the connection when we get to soldering the parts in place.
You could bend the leads on the resistor to make it fit to match the drawing.
I deliberately left it like this to illustrate the point Roly made.

That's all for tonight.
More tomorrow.  We get to soldering!  :)
If it ain't broke I'll fix it until it is.

Roly

Quote from: galaxiexShow off...  ;)  :)

...um...well... okay, yeah.  I did my back in at a drafting board laying out foot-square double-sided PCBs full of 74-series TTL using wax pencils and sticky "Bishop Graphics" tapes and shapes, dreaming of the software that has come along since that would make picking up a whole layout and moving it one-tenth over a simple matter rather than having to re-do the entire layout one-tenth over - then discovering that it needed to be two holes over anyway.   :grr


Quadrophonic.  Never mind 5.1 - Octophonic (now just got to find a guitar with eight pickups  :o ).

:crazy2:
|
|->-
|->-
|->-
|->-       8)
|->-
|->-
|->-
'->-

We could connect them all in series for one watt of the most shredded shred imaginable.

:crazy2:->-->-->-->-->-->-->-->-   :o

We could bridge them in pairs for double the output.

They would be awesome with a stereo chorus pedal.

Eight speakers arrayed in a circle of omni-directional coverage.

More seriously you could leave the two on the R-H end and add a stereo preamp for one of the smallest stereos in the world.  {Using micro pots and switches I reccon I could get that in a remote control case - I've already done a full single channel preamp (Gain, Treble, Bass, Middle, Master Vol) in a belt pack, and without going to SMD's.}

Quote from: galaxiexa mini powered  4 channel mixer beginnings? Nah, it'd never work...  :loco

Could do that for you sure (the real difficulty being providing miniature controls that can still be operated by finger.  A cellphone App maybe?).

Just to take this idea seriously for minute; at a minimum a four channel mixer needs four input sockets, and four level controls.  Even using sub-mini pots if you piled this on top of the board it would hide it, these add a lot of volume (or to put it backwards, look how much we get inside tiny modern components).  If you aren't going to use XLRs but 6.5mm TRS you can't realistically go any smaller even 'tho you can certainly get micro plugs and sockets down to 2.5mm.  You're not going to fit this "buskers PA" into a minty tin.  {...ponder...}


The possibilities are endless (still within a mint tin).   :)


{PCB layout.  I was driving from Melbourne to Canberra one time over what we laughing call "the alps" (2,228 metres/7,310 ft) when I stopped a guy coming out of the unmade road we intended to take over the top.
"What's it like?"  I asked him.
He thought for a moment, "That" he said, "depends on what you are used to" and drove off.}


Something I do (which I don't think will work here with this DIYlayout) is to design my layout in Protel EasyTrax (free) as a PCB (but with an eye to it going onto stripboard), print it out actual size, then cut it out and glue it as an "overprint"* on the component side of the board.

I have found this to be very effective in culling errors at an early stage, particularly when working on Vero/strip board, because all the thinking is shifted to the front end planing/layout stage, and when you get to the actual component stuffing it's just brainless follow the dots/numbers, and you can concentrate on making a good job of the build itself without trying to work out the layout or where you are at the same time.  The result is that boards built like this have a significantly higher "go-first-time" score.

As it happens EasyTrax is flexible enough to do circuits and scale drawings in, and I've done all sorts of things from sheds to speaker boxes, in fact building my W-bin first in EasyTrax exposed a serious problem with the access hatch - before any wood was cut.

(* the white printing on the component side is called an "overprint" or "component overlay")


FET
Quote from: galaxiex1. Drain 2. Source 3. Gate

1. the +ve end of the channel,
2. the -ve end,
3. the channel current control (normally on, -ve to turn off)

In this device the "channel" between the "Source" and the "Drain" act like a voltage-variable resistor, its value determined by the (negative) voltage on the "Gate".

At rest with zero volts on the Gate the resistance is low, but as the Gate goes negative the resistance rises to ultimately effectively open circuit, a very wide range.
{n.b. for this FET family, and there are many others that differ}

This family of FETs are quite venerable and have wide application.  In linear mode an MPF102 will amplify literally from DC to 500MHz (and Hams use them as preamps at 432MHz); overdriven they make a fine signal switch.


Tiny matter of elegance; I try to get all my resistors oriented the same way, generally tolerance band down to to the right.  They aren't polarised, but their markings are, and it doesn't matter if you follow "up+left", putting them all the same way makes it much easier to "read" the board at a glance and have all the resistor values up the right way.


In both programming and electronic construction consistency and getting rid of exception cases is important.  For example;  DIYlayout thingie is all very flash, pretty, and glitzy, but it is also very limited as a drafting tool.  It has obviously been created by a keen electronics amateur who is a somewhat better programmer.  The older version I have is slow, clunky to use, very limited and inflexible, and bug-ridden.  It was intended for just this sort of minimalist thing, stomps I would guess, but realistically that is about all it is capable of.  Even my T-50 power amp is well beyond it.

It also has a very limited range of available wire colours.  Now I don't expect 16-billion ('tho why not?) but the basic resistor colour code set would have struck me an a good starting minimum, and if it could use the "base-trace" method of a base colour and a different spiral trace colour then you would have a rational set of 100 different and specific wire colours, but no, just a few, and some of those (black, green, red) already spoken for.   >:(
If you say theory and practice don't agree you haven't applied enough theory.

Lauren

Good morning!  Thanks, galaxiex, for posting step by step pics...Using then along side Roly's diagram is helping me get started.  Soooo, I just realized that your diagram is only using 1/2 the board, up until the solid line of pads (sorry...I tried to go through the thread to find the correct terminology, but there's too much info for me to find it.  LOL).  If you were sitting with me right now, you'd be laughing at how often I am saying, out loud, "Ohhhh!" as I reread what you've written and actually put things together.  It makes a lot more sense when it's actually necessary to understand for the build.  :)
I actually understand now how you labeled the board, too.  Sorry for saying they didn't match!  Even with all of your explanations as to why it does, it was not until this moment (actually working on it) that the light bulb went off. At least it did, right?  :dbtu:

I ordered a Helping Hand that should be here on Monday.  I think this is definitely a necessary tool, as [someone--sorry, again I can't find you in the thread!] mentioned earlier.  I am curious as to whether you trim the ends off of things like the capacitor, since the are so long.  Don't worry--I'm not cutting anything.  Just wondering. ;)

Taking pics as I go, too.  Will post when I have something that looks like galaxiex's.  :tu:


Lauren

How do I delete a post I started to draft and then changed my mind on?? It doesn't give a cancel option that I can see.

Lauren

Possible roadblock:


C2: 100uF Mine does not look like galaxiex's; it has leads (?) coming out of both ends.  Do I need to exchange it for something else?