Coil Winder - Rev 2

CJ

Well-known member
Joined
Jun 3, 2004
Messages
14,147
Location
California
built a winder a while back,

it worked, but was hard to control,

so i reworked the thing and like it ok

so to build a winder,

1) find a DC motor with a half inch shaft,  size it for what size coils you build,
2) Granger - A)half inch shaft-only 5 bucks
                  B)1/2 inch coupling
                  C)jacobs super chuck, this will be expensive
                  D)sealed bearings with blocks-not too expensive
                  E)mechanical counter - diy electronic counter no good,
                  F)two pulleys for 3/8 inch 1/5 th pitch belt 16 and 64 teeth
                  G)one  belt - 3/8 inch wide 140 teeth  1/5 th pitch

3)Buy a small variac, put it into a 2:1 transformer, rectify the output, feed the motor the dc

thats it, you can put in a reverse switch if you want,

for precision work you can put a decoder on the shaft and feed it to a high speed dual comparator which can be divided with 7400 series logic chips to wind fine pitch, get a small ghecko drive and a stepper motor and a precision screw of some sort

then make a wire tensioner out of some coroprene pad and a c clamp of some sort to vary the pressure,

we have a DIY reverb output being wound, plate Z is right between 6V6 and 6K6 so i can run both tubes with minimal mismatch, bi fi, 3 pri 2 sec  2700 to 99 turns

 

Attachments

  • diywinder.jpg
    diywinder.jpg
    224.7 KB · Views: 363

CJ

Well-known member
Joined
Jun 3, 2004
Messages
14,147
Location
California
winding machine update- installed a tension device, just some springs and a cabinet hinge,

coast about 3 bucks plus some wood scraps, and a cut up sock,

th spring is in the back, hooked up to the eye ring, it has a 1/4 20 screw for tighten and loosen adj.
 

Attachments

  • winder1.jpg
    winder1.jpg
    256.9 KB · Views: 198

CJ

Well-known member
Joined
Jun 3, 2004
Messages
14,147
Location
California
and you can buy a 5/16-18 threaded rod for #1.09 and make a hand winding device with a teflon guide  like this>

that is a vibro champ output being wound on a twice stack bobbin that was made from two bobbins cut in half and spliced together,

instead of pri-sec we are going pri sec pri sec pri sec pri

use colored tape for lead tagging,

black is start 1, brown is fin 1, red is start 2, org is fin 2, yellow is s3, green f3, blu is s4, violet is f4, gray is s5, wht is f5, i still need 4 more colors, wtf, over?  :eek:



 

Attachments

  • winder2.jpg
    winder2.jpg
    200.2 KB · Views: 196

maxwall

Well-known member
Joined
Nov 16, 2004
Messages
1,086
Good topic here, sorry to derail the winder bit , but is there info on how to build a toroid winder , those are painfully drawn out to do by hand, like as in Pultec toroids and other toroids.  Ytube hosts some good videos but there a tease to watch because they make it look so easy with the right $8-15K machine , where does DIY fit in  , has anyone made a attempt , without empty-ing a 401K account on a decent machine.

No issue with finding old cores and #34 wire though... this part is covered.
 

CJ

Well-known member
Joined
Jun 3, 2004
Messages
14,147
Location
California
you can get a Yuze toroid winder new for 2000 bucks, not the best but you can mod it to work better,

for 1 off pultec inductor, just get torwico core and use a Popsicle stick with a V in both ends,

doing multi strand today for a 100 watt  output,

put a wire rack on the ceiling with a bike hook,l



 

Attachments

  • wirerack.jpg
    wirerack.jpg
    168.4 KB · Views: 76

CJ

Well-known member
Joined
Jun 3, 2004
Messages
14,147
Location
California
added some hookup wire racks and a new tensioner for the big wire, plus multi wire teflon guides,

gonna get that gecko step driver going real soon,
 

Attachments

  • multiwinder.jpg
    multiwinder.jpg
    208.4 KB · Views: 130

Igor

Well-known member
Joined
Jul 17, 2006
Messages
2,193
Location
Israel
Woa, nice! I did something in this direction some years ago, inspired by your bloody trannie dissection posts.
TurboCNC and 2 steppers worked good for me.
 

Attachments

  • winder.JPG
    winder.JPG
    118.8 KB · Views: 198

CJ

Well-known member
Joined
Jun 3, 2004
Messages
14,147
Location
California
hot off the mandrel,

splice time,

winding goes pri-sec,    Not!

best output in the world for a pair of kt 88 or 6l6 quad

let me know if you want one of these transformers built up,

+/- 1 db from 10 to 100,000
 

Attachments

  • coil1.jpg
    coil1.jpg
    352.1 KB · Views: 157

joaquins

Well-known member
Joined
Feb 25, 2012
Messages
1,927
Location
Argentina
Please don't miss one splice! nice job... nice response too!

JS

PS: you are out of your mind!
 

CJ

Well-known member
Joined
Jun 3, 2004
Messages
14,147
Location
California
ok here is a pic of the new transformer next to the old output trans,

will have it wired up soon!

:D
 

Attachments

  • kt88t.jpg
    kt88t.jpg
    47.2 KB · Views: 136

bruce0

Well-known member
Joined
Jan 24, 2010
Messages
1,065
Location
Boston
Coil winder - rev "Ch3ap"  (Cheap)

I just wanted to count turns, but I only have two hands.  I tried a rolling microswitch wired to an arduino, outputting turn counts to my computer screen.  But at high speed it would skip.

Then I remembered these LDR's i had around from building compressors.  And foam core.  And some little felt feet.  And a variable speed drill.  And a neat little xacto knife.  And hot glue.

2 10K resistors, 2 LDR's and some wire.

It winds and unwinds, and counts turns in both directions.  It can go pretty darn fast.  I have some "faster" LDR's but I saved them for the next compressor.

It works using the ambient light in the room, although I do shine a floodlight on my workspace.
I am new to arduino, but it is only 30 lines of code or so (a counter with dbounce, working on two switches made of analog input LDR inputs)

two disks, black.  One for the mandrel (you can make more disks if you have more mandrels.

IMG_1806_zps2916e017.jpg


The felt rings help shield the light, and keep the lenses from getting scratched up.

IMG_1803_zps437357b8.jpg


It is precise to within 1 turn, but you can make it half turns or quarter turns if you want, but generally you have to stop on the same side of the bobbin anyway.

If it is not obvious, there are two LDR sensors placed radially on the other disk.  If the outer one sees light first, the drill is going backwards, and visa versa. 

Disk is hot glued to mandrel, in this case tapered wood shaft for pot core bobbins.

IMG_1802_zps9364f1b5.jpg


How do you like those precision cuts!

Sunk the LDR's into the foam-core, punched two holes for the leads, and hot glued them in.
Punch a hole in the felt and give the LDR an eyelid (not sure it is needed, but I did it).  The other felt pads are to prevent foamcore disks from touching.

IMG_1804_zps04e64bc2.jpg


That's the whole circuit.  5V off the arduino feeds two voltage dividers with 10K resistor connected to LDR and then ground.  And the analog pins on the arduino fed by the junction of 10K and LDR.  THis is an old piece of bell cable or something to keep it neat.

IMG_1804_zps04e64bc2.jpg


There is no attachment of the assembly to the drill (that is the great part of this design).  It just hangs there.  It worked fine with just the bell wire supporting it but I hot glued a wooden board on it for weight and to keep the two disks aligned better.


The whole assembly took me about 3 hours to put together and tweak and do two test windings (you have to tweek the "level" for the two LDR's which can be different).  The test was winding two completely repeatable Pultec RM8 cores, within 2% of my target Henries.  (wound 100 turns of reused magnet wire, measured core, adjusted the AL number in my spreadsheet  then just wound based on counts, and it was right on).  Second core was different AL, and they were next to each other on the packaging tape too.

I know this mount look loose, and that is what is so great about it.  It is loose, and works perfectly for me.  No tight tolerances, and no wiring to count backing up, it just does it based upon which LDR it sees first.


The arduino program is a hacked up version of one of the examples, it combines the edge detection example the analog read example and the debounce example... then I added a little bit.

I tried to make it calculate inductance on the fly but I didn't bother to download math libraries. 

I think this could have a little LED display output, but that is a project for another day (and would bring the cost up to the cost of a totalizer which are 60 bucks or so).

IF you make one, when you set up you want to put the sensors on the side of the winder AWAY from you, so your fingers don't block the light and screw up the counts.  But this is an easy, cheap way to count turns, and it gave me a change to learn a little about arduino.

I think you could hook this on most of the winders I see around on the site, and it would work.  In my case the sensor is held in place by the mandrel and the disk with the slots in it.  (there is no other attachement).

The code follows:
Tweaks are in variables and constancts, you might likely need to adjust them for your situation:

debounceDelay  Prevents the sensor from toggling when it is near the light threshold.  Don't want to count turns because a cloud or a kid rolls buy and the sensor is only part covered. (I use 5 milliseconds, if the LDR hits threashold for shorter than than then it doesn't count.  You can shorten it it works at 3 depending on light.  Since the slots cover about 10% of the arc of the circle, this limits counting to about 1000 RPM).  Longer slots or a lower value here would increase that.

Don't bother changing pins, unless you have a reason.  I left the digital pin version mostly intact (just commented out)  THe LDR's use the analog pins.

// const int  buttonPin = 12;    // the pin that the pushbutton is attached to
const int  sensorPin = A0;    // the pin that the ldr is attached to
const int  modeLeadPin = A1;    // the pin that the ldr is attached to
// const int  modePin = 8;      // the pin that says forward or backward



Same threshold for both sensor..., 50 milliseconds.  I tried higher but I think the LDR's have a "Dark Adaptation time" and mine was long, so you need to move it to the "operating range" for your LDR.  Do this by experimentation.  It is right when you spin the drill fast in both directions and it doesn't lose count.  Two high...you will get missed and false counts depending upon lighting.  Too low you will get missed counts.
const int sensorThreshold = 50;










/*
  State change detection (edge detection)

Often, you don't need to know the state of a digital input all the time,
but you just need to know when the input changes from one state to another.
For example, you want to know when a button goes from OFF to ON.  This is called
state change detection, or edge detection.

This example shows how to detect when a button or button changes from off to on
and on to off.

The circuit:
* pushbutton attached to pin 2 from +5V
* 10K resistor attached to pin 2 from ground
* LED attached from pin 13 to ground (or use the built-in LED on
  most Arduino boards)

created  27 Sep 2005
modified 30 Aug 2011
by Tom Igoe

This example code is in the public domain.

http://arduino.cc/en/Tutorial/ButtonStateChange

*/

// this constant won't change:
// const int  buttonPin = 12;    // the pin that the pushbutton is attached to
const int  sensorPin = A0;    // the pin that the ldr is attached to
const int  modeLeadPin = A1;    // the pin that the ldr is attached to
// const int  modePin = 8;      // the pin that says forward or backward
const int ledPin = 13;      // the pin that the LED is attached to
const long al = 3300;      // in nanohenries
const int sensorThreshold = 50;
// Variables will change:
int buttonPushCounter = 0;  // counter for the number of button presses
int buttonState = 0;        // current state of the button
int modeState = 0;        // current state of the forward/reverse mode
int lastModeState = 0;    // previous state of the button
int lastButtonState = 0;    // previous state of the button
int lastBounceState = 0;    // previous momentary state
int lastModeBounceState = 0;    // previous momentary state
int reading=0;
int mode=LOW;
// the following variables are long's because the time, measured in miliseconds,
// will quickly become a bigger number than can be stored in an int.
long lastDebounceTime = 0;  // the last time the output pin was toggled
long debounceDelay = 5;    // the debounce time; increase if the output flickers
long lastModeDebounceTime = 0;  // the last time the output pin was toggled
long henries=0;

void setup() {
  // initialize the button pin as a input:
  pinMode(buttonPin, INPUT);
  // initialize the button pin as a input:
  pinMode(modePin, INPUT_PULLUP);
  // initialize the LED as an output:
  pinMode(ledPin, OUTPUT);
  // initialize serial communication:
  Serial.begin(9600);
}


void loop() {
  // read the state of the switch into a local variable:
//  int reading = digitalRead(buttonPin);
// read the input on analog pin 0:
  int sensorValue = analogRead(A0); 
  int modeLeadValue = analogRead(modeLeadPin);
  //debug

  if (sensorValue > sensorThreshold ) {
    reading=HIGH;
  }
  else {
    reading=LOW;
  } 
if (modeLeadValue > sensorThreshold ) {
    mode=HIGH;
  }
  else {
    mode=LOW;
  } 
  // check to see if you just pressed the button
  // (i.e. the input went from LOW to HIGH),  and you've waited
  // long enough since the last press to ignore any noise: 

  // If the switch changed, due to noise or pressing:
  if (reading != lastBounceState) {
    // reset the debouncing timer
    lastDebounceTime = millis();
  }
 
  if ((millis() - lastDebounceTime) > debounceDelay) {
    // whatever the reading is at, it's been there for longer
    // than the debounce delay, so take it as the actual current state:
    buttonState = reading;
  }

  // If the mode changed, due to noise or pressing:
  if (mode != lastModeBounceState) {
    // reset the debouncing timer
    lastModeDebounceTime = millis();
  }
 
  if ((millis() - lastModeDebounceTime) > debounceDelay) {
    // whatever the reading is at, it's been there for longer
    // than the debounce delay, so take it as the actual current state:
    modeState = mode;
  }

  // compare the buttonState to its previous state
  if (buttonState != lastButtonState) {
    // reset the debouncing timer

    // if the state has changed, increment the counter
    if (buttonState == HIGH) {
      // if the current state is HIGH then the button
      // wend from off to on:
    //  if (digitalRead(modePin)==HIGH) {
      if (modeState==HIGH) {
        buttonPushCounter++;
      }
      else {
        buttonPushCounter--;
      }
    // Serial.println("on");
      henries=(buttonPushCounter*buttonPushCounter * al)/1000000;
      //Serial.print("mH:  ");
      //Serial.println(henries,5);
      //debug
      //Serial.print("  switch:  ");
      //Serial.print(sensorValue);
    // Serial.print("  mode:  ");
    // Serial.print(modeLeadValue);

      Serial.print("  Turns:  ");
      Serial.println(buttonPushCounter);
    }
    else {
      // if the current state is LOW then the button
      // wend from on to off:
      // Serial.println("off");
    }
  }
  // save the current state as the last state,
  //for next time through the loop
  lastButtonState = buttonState;
  lastModeState = buttonState;
  // save the reading.  Next time through the loop,
  // it'll be the lastBounceState
  lastBounceState = reading;
  lastModeBounceState = mode;
 
  // turns on the LED every four button pushes by
  // checking the modulo of the button push counter.
  // the modulo function gives you the remainder of
  // the division of two numbers:
  if (buttonPushCounter % 4 == 0) {
    digitalWrite(ledPin, HIGH);
  } else {
  digitalWrite(ledPin, LOW);
  }
 
}
 

CJ

Well-known member
Joined
Jun 3, 2004
Messages
14,147
Location
California
Bruce the Genius!  :D

and thanks for the link,

stock market is going up, transformer prices are coming down,

it's all good,
 

Attachments

  • wyatttrans.gif
    wyatttrans.gif
    1.7 KB · Views: 331

PRR

Well-known member
Joined
Jan 30, 2010
Messages
11,144
Location
Maine USA
> I have some "faster" LDR's but I saved them for the next compressor.

So use photo-transistors. Even with crude 1-resistor interfacing they are faster than anything you want spinning near your face. +5V to resistor, to collector, ground the emitter. Try 10K, 1K, 100K... you want the collector to go low when lit, go high when dark (whatever lit and dark are in your world). Dirt-cheap.
 

CJ

Well-known member
Joined
Jun 3, 2004
Messages
14,147
Location
California
they are fast, my cheap boss made a resistor counter to measure off large numbers from the big spools, actually he was not cheap, he was smart, we were doing a startup where DIY is very important til the bank is paid off,
 

bruce0

Well-known member
Joined
Jan 24, 2010
Messages
1,065
Location
Boston
PRR... you are right, maybe next time.

I didn't have any photo transistors.

I wanted to build it with stuff that was within 10 feet of me <grin>.  I am very happy with it.  I tweaked the software to have different constants for the thresholds on the two different LDR's and now I can run the drill full speed (the LDR's were very different, maybe I used two different kinds).  I mess up the winding when I go fast anyway.

PRR said:
> I have some "faster" LDR's but I saved them for the next compressor.

So use photo-transistors. Even with crude 1-resistor interfacing they are faster than anything you want spinning near your face. +5V to resistor, to collector, ground the emitter. Try 10K, 1K, 100K... you want the collector to go low when lit, go high when dark (whatever lit and dark are in your world). Dirt-cheap.
 

Latest posts

Top