Status 11/18/2013

Last couple of weeks have been very busy.  Taking an old EM playfield and converting it to modern drivers has been going well.  Lots of things have been  learned in the process.

1.  Don’t use the old wiring.  While it looks very convenient and seems to have many of the connections that are needed, it is from the 1960s and the insulation is made of cloth, not plastic.  It is a pain to strip the wire which makes it difficult to get a good crimp.

2.  The wire doesn’t necessarily run where it needs to run.  There is a common ground for almost everything so even more wire could be removed from the bottom of the playfield.  If the playfield would have been stripped, and started from scratch, only the needed wires would have been added back.

3.  The cheap crimper (KF2510), works incredibly well.  It’s been used for 100 mil spaced connectors, standard pinball molex connectors, and spade terminals.  It has three different sizes for different connectors.  While it isn’t as nice as the $300 molex crimper that only supports one terminal type, it works great for general use.  The trick is to start the crimp so that the ratchet holds the connector in its teeth.  Slip the wire in to the terminal, and then finish crushing the crimp.  The crimps are nice and tight and hold the wires very securely.

After rewiring, this is what the bottom of the playfield looks like so far:


The first picture shows the driver board at the top center of the picture, with an input card on at the bottom of the picture.  Second picture shows the entire playfield.

The solenoid driver at the top handles eight solenoids.  It includes eight inputs so all the real time aspects of driving the solenoids is local to the little card.  The four wire cables on the edge of the is a transmit/receive/power and ground connections.  To each solenoid that is controlled by a switch, there is a single wire to ground at the bottom of the solenoid (NFET configuration), and a single wire to indicate the switch is closed.  The hot side of the solenoids are all bussed together with the high voltage.  The common side of the switches are bussed together with a ground wire.  (Each switch connection has a 2 pin header with one of the pins being ground.  This gives the flexibility of not needing to run a ground bus.).

The input driver board is very similar except there are 16 inputs per card.  Each switch input is a two pin connector which can provide a ground if a common ground wire bus isn’t available on the playfield.

Last big thing that looks out of place is the cheapo power supply.  It is a Meanwell LPP-150-48.  It only provides 3.25A at 48V.  Bought two of them and planned to split the flippers between the two power supplies, but during testing over the last couple of days, it seems like a single power supply might be able to run the flippers.  Of course, the second supply is already bought, so it might as well be used.  Each of the supplies is $9 on ebay.

Nice to hear the bam, bam of solenoids being driven at 48V.  I’m probably going to take a little time to build a second solenoid driver card.  (Right now there aren’t enough drivers on a single card to support two of the pop bumpers and one of the kickout holes.  Parts are already in house for another two driver cards).  Each type of solenoid has been wired up to make sure the driver card configuration is working properly.  The flippers have a 48ms initial kick and then a 25% on the PWM for hold.  The kickout hole has a 48ms initial kick, no hold, and a dead time configured in case the kick doesn’t clear the ball from the kickout hole.  There is a post that goes up and down, which has no initial kick, and only 25% PWM.  Last type is the pop bumper which has an initial kick, no PWM, and a small dead time.

The input card is a little more dull at this point.  It is currently wired up and shows the status on a computer sitting next to the playfield.  Boring.  No scoring, no rules, hey it is just a test bench.  Enough rules might be written so a game can be played and scoring can be added but no real plans at this point.

About these ads

5 responses to “Status 11/18/2013

  1. I’m happy to see that a switchmode power supply is working out for you. After reading this thread on pinside about how only linear transformers will work I was skeptical:

    I have a background working with battery chargers. We would design both linear transformers and switchmode transformers and both seemed to get equal output (switchmode was much lighter and ran more efficiently since there wasn’t as much heat build up), but switchmode usually took longer to design because linear is nothing but windings, iron core size and steel quality.

    That transformer you found on ebay looks like typical chinese design but hey if it works for $21 shipped who cares? You talk about using PWM to reduce the voltage for holding power. This is obviously what most people would do who already have a circuit to do so, but is there any standard off-the-shelf part that would do something similar? For instance if I wanted to simply wire 48V directly to a solenoid? I know a relay wouldn’t react quick enough, wonder if a 555 timer could do it.

  2. That topic was thrown out there because somebody at pinball expo talked about a much lower cost power supply. After re-listened to the talk, it was Skill Shot Pinball. Went to their website and found out that no produces are currently listed. Seems they are just in their infancy. The price stated was $30 for 32V. That may be interesting depending on the specs.

    The two SMPS were only bought to test the driver cards at higher voltages. I really thought the power supply was going to burn up in the first five minutes of use. (Another reason that I bought two). They have held up surprisingly well so far. They certainly aren’t going to be in the final design of Disaster. (I could run a multi-day stress test to see what happens, but that would probably drive the family bonkers).

    So the Antek power supply that everybody is using is $125 and only produces 398W of power. (70V * 5A + 24V * 2A). (Here comes the part when I get people angry at me.) Dinosaur! There, I said it…dinosaur. If the same technology was used for laptop computers, the wall wort would need to be 5lbs. Power supplies in pinball machines are going to move from transformer based to SMPS, it just hasn’t happened yet. The only argument against computer power supplies is they will not last 30 or 40 years. The issue seems to be the quality of the caps. The Antek supply will also fail, but replace the one bulk cap and the one bridge rectifier and it is a new power supply. It is more difficult to find the correct cap to fix in a SMPS.

    Now recognize the following comments come from an embedded guy so take them with a grain of salt. You can run a solenoid off a 555 counter/timer…but why? So in the current setup, there are three different styles of PWM configurations (flipper, pop bumper/kickout hole, and gate). Each of them have different timings based on what makes the most sense. Using a 555 counter timer, each chip is going to need a different RC circuit to get the timing correct. More circuitry would probably be needed to control a flipper where an initial kick is needed and then switched down to a lower duty cycle for the hold. It is just so much cheaper/easier to use a microcontroller. So a 555 is $.90 in qty 10. It can control one solenoid. An MC9S08SH8 is $1.88 in qty 10 and it can easily control 8 solenoids. It can also report status back for scoring. If starting from scratch, I’d be looking at the STM8S003 which costs $.703 in qty 10.

    Before you go down that route, send me an email.

  3. heh. meanwell is a very well respected PSU company. they might be chinese but they really do make a good product. my homebrew pin is using the 10A 48v psu and it works just fine.

    • I’ve had good luck with Meanwells and used them in a bunch of laboratory equipment products. I never had a problem with them.

      Are you running 220V or 110V? There are many good options for 10A, 48V if 220V, but if you change down to 110V the choices get much more expensive.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s