AC/DC Power Convert Questions

Good answer there from Dave. Re your other questions:

According to the sales page for that power supply, the output can be adjusted by ±10% which is a range of 25.2~30.8V. If you want a separate 5V rail you should use a second power supply fed from that one. This can be a non-isolated buck converter such as http://www.ebay.com/itm/DC-DC-5V-12...61?pt=LH_DefaultDomain_15&hash=item43d47b0d51.

Possibly. The polarity of the voltage you apply to the coil determines the direction of the magnetic field it creates. In a simple situation, the metal that the solenoid coil attracts is unmagnetised and will be attracted equally by a north pole and a south pole, but that might not always be the case. If the solenoid is designed to be energised with a specific polarity, there should be polarity markings near the connection points.

It's also possible that a solenoid coil includes a diode in series with the connection, in which case reverse polarity will not activate the solenoid at all, and/or a diode in parallel with the coil, to eliminate inductive kickback (look at the pages you find through https://www.google.com/search?q=diode+inductive+kickback&tbm=isch) in which case reverse polarity will cause very high current flow and will probably damage the diode and/or power supply. In those cases, polarity should also be marked.

If you're not sure, and the machine is currently in operating condition, measure the voltages across the solenoids when they're activated, to check the voltage and the polarity. Then match them with your redesign.

 

Agreed. I missed that in Dave's post. AFAIK ferrules or crimp connectors are preferred (around tightly twisted strands); otherwise, twist tightly, double the wire back on itself or loop it round the screw, and screw the washer down, without using solder.

Edit: I'm probably wrong about twisting wires to be crimped. See this thread: https://www.electronicspoint.com/threads/should-wire-be-twisted-before-crimping.272735

 

Since the OP is in the U.S. I would like to point out that the colors in a line cord will not be as posted by Dave.

In the U.S. the colors will be white for neutral, black for live, and green for ground.

Bob

 

If the solenoid is designed to be operated from the full 28V DC, then yes, I would say so. But I have no experience with the innards of pinball machines. I don't know why that wire would be labelled "Center Tap". Centre tap of what?

I don't understand the question. The power supply needs to supply all of the solenoids, so each solenoid circuit (the complete circuit, including the two-coil solenoid, EOS switch if applicable, and user pushbutton) needs to be connected across the power supply. So these circuits end up being connected in parallel with each other. Whether you run the power supply output wires (+28V and 0V) around the machine and tap into them as required, or run a separate pair from each solenoid circuit back to the power supply output, is up to you.

Have you tried to contact Al Garber, the creator of that picture?

 

A 1N4007 is rated at 1000V and 1A and will do what you want and will be suitable for many other applications. The diode only has to pass the solenoid current for a very short time so can be overloaded somewhat. Adding a diode will slow the solenoid release and a series resistor may be desirable to limit this.

You can use lower rated diodes but there is little point.

 

The diode needs to be rated for at least as much current as the solenoid draws. If the solenoid draws less than 1A, a 1N4001~4007 as Trevor suggests will be fine. As for whether to add on or not:

Without the diode, the inductive kickback will generate a high-voltage pulse which will cause arcing across the contacts of the switch when it tries to break the circuit (i.e. when the operator releases it). This will probably greatly reduce the lifespan of the switch.

With the diode, the switch will be protected, but the solenoid will not release as quickly as it would otherwise, because when the switch is released, the solenoid current will not fall to zero quickly; it will be maintained by the diode. The delay in the solenoid release will depend on the solenoid's inductance, mostly.

There are various compromises. As Trevor mentioned, you can put a resistor in series with the diode so the inductive kickback voltage is reduced but not to zero. You can use a zener diode instead of a resistor, which has the same effect.

I suggest you try just a diode, and if the delayed release is noticeable and has an effect on playability, we can consider our options. Another option is to drive the solenoid with a MOSFET (which is controlled by the swich) and use a diode with resistor or zener diode in series with it. This will protect the switch contacts while still giving a quick solenoid release.