Saturday, June 27, 2015

Ball sensor

 

In most pinballs, the ball is detected by switches. In older systems these are leaf switches. In more modern ones they are micro switches. In some systems they are contactless.

Leaf switch clip_image001
Micro lane switch
 
Leaf lane switch image lane switches between the lane guides  

For contactless ball detection are 3 systems present:

  • Optical detection with a infrared diode and sensor. the ball breaks the beam.
  • Electronic circuit that detects the ball when it gets close to a coil.
  • Magnetic detection
optos
image
 
optical detection coil detection magnetic detection

 

The sensor

I don’t like mechanics switches and I prefer a contactless approach. Less things to get damaged or break. Also for the switches the playfield need a long and narrow slit, what is difficult to make when making the playfield by hand. 

There are induction sensors available on ebay or aliexpress, that are very cheap. Less as 2 usd shipped! I don’t understand how they make them for this amount.

image

http://www.aliexpress.com/item/Free-Shipping-10pcs-lot-Original-Gaode-Inductive-Proximity-Sensor-LJ12A3-4-Z-BX-NPN-3-wire/2038864848.html

The biggest drawback of this specific type is the detection distance. For the 12mm version this is 4mm. Therefore the sensor needs to very close to the pinball.

There are other types with larger diameters that have much longer detection distances, but also more expensive. This one has a diameter of 30mm and a detection distance of 25mm. It can therefore be mounted below the playfield and sense through it. Still need to order one in for some testing.

image

http://www.aliexpress.com/store/product/SE-3025A-Genuine-inductive-proximity-switch-sensor-switch-NPN-3-wire-NO-DC6-36V/808493_1980850614.htm

The other alternative is to design my own induction sensor. Then you can adjust the coil size to what is needed. I did some experiments with a LC oscillator, and this looks feasible. I will design a small PCB with and 8pins AVR processor to do some more test.

I have not finalised what approach I will take. For that need to do some more experiments.

12mm sensor

To test the 12mm sensor I designed a holder for it.

sensor holder sensorholdertop  
top bottom  

This is screwed in the playfield

sensor_insert sensor_insertass  
sensor_playfield sensor_insertxx  

To bring the sensor close to the ball, the sensor has to stick through the playfield. A small insert is designed with a thin front to bring the sensor as close as possible to the ball. The problem is that the this front is very thin and thus less strong, and question is if it will be strong enough.

tn_IMAG1773

The printed insert feels strong enough.

Now I need to find some time to glue the insert in a piece of plywood and do some testing if it sensitive enough.

I also designed a shooterlane with the sensor. Here the sensor works good, because the ball is located and centered above the the sensor.  tn_IMAG1879

 

tn_IMAG1880

More about this part in a later blog.

 

 

.

Thursday, June 25, 2015

LIGHTS !

 

Yes, a pinball without its lights is no pinball.  Also music is needed , but that is problem for later.

LEDS

The lights I going to use on my fabulous pinball are addressable LEDs. These LEDs are all connected in series to form a long string. Each LED can be individually controlled in intensity and colour.  Per string you only need a single bit to control them. The playfield is likely make out of multiple strings to make routing easier. 

The LEDs I found are already pre-soldered with wires in between them, ready to go. Standard they have 50 LEDS per string, but that number can be reduced or increased as needed. 

tn_IMAG1585

This is one LED. They are WS2812  RGB LEDSs.

tn_IMAG1586

The PCB is tiny: only 10mm in diameter and most of it is used by components or wire connections. This does not leave much area to fix and hold the PCB, so they have to be held by their wires.

Insert lighting

For the insert lighting I needed some bracket to hold the LED in place.  This is the design I came up with after a couple of tries. The LED wires are clamped between the fingers of the holder.

The line on top is to align the LED with the center of the insert. For the different size of insert the holder is different lengths.

insert_ledholder

Blue is LED model. Yellow the LED holder.

Here is one LED holder mounted on the rear of a test piece.

tn_IMAG1884

 

With a LED inserted into it.

tn_IMAG1885

The LED just hangs in the air, and is only hold in place by the wires.

 

tn_IMAG1887

Another view.  I have not yet programmed an Arduino, so was not able to control them yet.

 

Playfield lighting

The playfield also needs some lighting. I am not certain if these LEDs are bright enough for this purpose, otherwise just need to place some more.

The LED holder has 2 parts:

  • A stem that hold the wires and PCB in place. This sticks through the playfield.
  • A optional hood that diffuses the LED light.

Both parts snap together .

 

ledholder_cutview

This shows the cut through of the led holder. Blue is led model. Gray is a piece of the playfield.

 

top_led1

Bottom view of the light. The hole in the bottom is for an optional melt-nut. This can be used with a washer or small bracket with screw to keep the LED in place. Normally the LED will just be kept in place by the wires compressed in the hole. This holds the LED quite good in place.

 

top_led2

Top view of the diffusor dome. The real one is much rounder than this very course rendered one

 

. tn_IMAG1882

Here the actual dome and holder. The wires are folded around the stem and both are pushed together into the playfield from the top.

Monday, June 22, 2015

Inserts

 

Overview

Inserts , like pins, are available in tens of colours , shapes and patterns.

Some shapes and patterns from a old Williams book:

image

image

Some of the inserts for sale at the moment:

image Image result for playfield insert
Image result for playfield insert Image result for playfield insert Image result for playfield insert

 

Design

Because I use RGB LEDs, I don’t need to have coloured  inserts, white opaque is enough. The colour of the led will set the insert colour.  The insert only need to be transparent need to diffuse the LED light, so the insert is evenly bright lighting up.

I did some experiments to try to replicate patterns of inserts in 3d printed inserts, but this is difficult. The ‘transparent’ ABS is not transparent enough after 3d printing, to see the patterns on the bottom side of the insert.

The only pattern that works good enough is a circular pattern. Circular patterns on the bottom do not work, but using a low infill (35% ) circular pattern works quite well. That is because it is close to the surface of the plastic, so it is still visible from the top.

The other insert that works well is the plain and pattern less one.

The diameters of inserts are some imperial diameter. Because I only have metric drills, I changed the diameters to metric. I used the following diameters: for circular inserts 19mm, 25mm, 30mm and 40mm for the arrow.

Here some examples

insert_plain1 insert_circle insert_star
insert_arrow inser_circle insert_worst

 

The result

Some inserts just printed on the Mendel90 3d printer. The heated print bed is a bit uneven due to the many layers of glue-stick on it. Occasionally I will smooth the glue with some water when it gets too rough.

For the real prints of the insert that go in the playfield, I will print it on glass coated with ‘slurry’. Slurry is ABS dissolved in acetone. Things printed on slurry get a very smooth bottom. This will make the top surface of the printed insert as smooth as glass.

tn_IMAG1745

You can see the patterns on the bottom of the inserts with various patterns.

tn_IMAG1735

I glued some inserts in a bit of plywood with normal PVA glue.  This works very well. I could not push the insert out of the wood. I am surprised that the PVA glue does adhere to the ABS plastic so well.  Later on will try epoxy glue too, see how that works..

Here the test board from the bottom. The holes are 25mm the whole depth of the hole, without a ridge like real playfields have. Possible the ridge will hold the insert even better in place. Something to try out too.

tn_IMAG1884

The plastic part screwed on the bottom is a LED holder. I will write a blog about that later.

 

I will make some pictures with the inserts lighted up.

Ball gates

 

The list of part that a pinball playfield need is endless.  I am already designing for months now, and I am still not finished with all the mechanical parts..

Overview

There are several type of gates. In the picture below is a wire gate.

image

They are used in places where the ball is not moving to fast. Faster balls could bend the wire.

There are also stronger gates that use a piece of spring steel to reflect the ball back. They are in places where the ball speed is higher when it hit the gate. For example on the end of the shooter lane.

image image

 

The wire gate

I like to make everything out of plastic for easy reproduction . So no metal bracket but a plastic one. The gate itself will be made using a metal wire, because this more easy absorbs and returns the ball energy.  And it is easy to fabricate too.

Here the 3d model. The plastic shape  strength is increased with the triangles behind the vertical parts.

 

gate2

The gate wire used is a 1.2mm thick stainless spring steel wire. very easy bend with a plier.

tn_IMAG1788

This are some test pieces made with soft iron wire to size the gate.  Below is the un-bended steel wire waiting its turn.

There the resulting wire gate.

tn_IMAG1899

For the photo I just stacked the parts. When mounted on a real playfield the right pin is more to the left, closed to the wire.

The firepower playfield needs two different gate sizes.

tn_IMAG1792

 

Stronger gate

The stronger gate I designed in plastic with a small piece of spring steel to absorb the ball force.

springgate3 springgate2 springgate1

The printed product

tn_IMAG1896

The pinball lays on the side of the gate, otherwise the would fall before I could make a photo!

 

tn_IMAG1897

But the design is not a 100% success.  It works, but can be improved. After designing and testing one, you understand why the gate are designed in the current way.

I planned to insert the spring steel vertically in the moving part and bent it 90 degree. No such luck – the used spring steel is just hardened too much: it just snaps . next time only use straight pieces of spring steel.

The issues with the gate are:

  1. The ball have to lift the gate to much: the change the ball got stuck under it is quite real.
  2. The weight the ball has to lift is too much too.
  3. The spring steel is to short.

So in the next iteration the moving part will be more balanced, lighter and shorter.