Current limiting MOSFETs....

A closer look at the latter 555 circuit reveals a complimentary emitter follower output. Its function is to presumably provide plenty of power?

A 1988 patent application is more elaborate:

The interesting thing about the patent is that it doesn't distinctly identify U7 as a 555. Instead it uses "monostable" nomenclature, perhaps for legal reasons?

For what it's worth, here's Bob Pease's topical take:

A 1988 patent application is more elaborate:

The interesting thing about the patent is that it doesn't distinctly identify U7 as a 555. Instead it uses "monostable" nomenclature, perhaps for legal reasons?

For what it's worth, here's Bob Pease's topical take:

Q4 is a switch that grounds the solenoid when a control signal is applied. D4 provides resistance to drain the coil's reverse EMF faster after the control signal is removed. D3 protects V2 from the V1 spike. While D2 protects Q4 from reverse EMF after the coil is de-energized. All of the dozen or so Bally equivalent coils in my possession come prepackaged with a free wheeling diode already soldiered across the coil's terminals. Long story short, D4 can be dropped for the purpose of this thread. When the monostable output turns Q2 ON it in turn turns ON Q3 to supply the spike voltage to the coil. It seems there ought to be a way to simplify this subcircuit. More thought is needed after other tasks are completed.

Here's my nomination for the perfect 555 monostable MOSFET pick and hold solenoid driver:

But Bally's been grabbing coin ever since Moloney pitched punchboards. Dropping coin to modernize capacitive pick and hold circuits goes against the grain. If it ain't broke, don't fix it, grab coin instead.

...

I rather Like Bob Pease's solution and I think I will go with it. If the CPU locks on a coil his design will prevent it from melting.

These are 24VDC coils, and are supposed to be momentary - pop bumpers, kickers, etc. None are supposed to remain on.

For every electronics problem there is probably a Bob Pease solution! I'll take another read through of "A Tribute to Bob Pease - Troubleshooting Analog Circuits" there may be some more gems in there for me!

Thanks!

John :-#)#

With all due respect, Pease's solution seems the simplest way to achieve "Capacitive Pick & Hold #1" shown at:

Bally just drives the 16 solenoids directly by 74L154 to a resistor to a CA3081 transistor array, then the SE9302 transistor through a blocking diode. The odd pull-up resistor and a small cap. No real protection at all. Coils and transistors died for various reasons from poor ground connections etc.

John :-#)#

Thank you for taking the time to jog my memory. Your previous posts pertinent to this precise problem now ring a bell, so to speak. Bally's counter-intuitive design takes me back to when a then older (in his late twenties) engineer confided to me how engineers assume that the other guy knows what they themselves know. Given my complete lack of understanding of the challenges faced by Bally at the time, it seems best to refrain from second guessing Bally's original, counter-intuitive design.

Some threads stimulate me into creating a web page. This is one such thread. It motivated me to assemble an apparatus from the solenoid coils, springs, steel rods, and wood in my bone pile. Voilà, a Do It Yourself (DIY) knocker in its de-energized state:

and here it is energized:

When energized, my DIY knocker makes a Bally-worthy knock.

My next step is to incorporate various solenoid drivers into the apparatus. Protection may be packed into the (previously unknown to me) NXP33886 used in the LATCHING VALVE CIRCUIT shown at the bottom of this link:

Cute, the gel battery will provide enough current I'm sure.

I use Williams knockers for simulating gun bangs on my carnival rifle range. The knocker is underneath of the counter that secures the rifles. I'll post some photos if anyone is interested...

Yikes! That is way more protection than is realistic for my project. Pease's solution is far simpler and fewer parts count.

Thanks,

John :-#)#

Yes, please post photos of your Williams knockers.

My hometown hosted its annual fair last week. A Las Vegas comedian/magician/hypnotist made an appearance at a midway venue. During his act, people desperate to be the center of attention - drama queens - allow themselves to become hypnotised. It doesn't seem to matter if they make total fools of themselves, as long as they're center stage and are rewarded with receiving a round of applause afterward.

You may have confused the complex 555 circuit with the relatively simple circuit at beneath it. That circuit contains only a NXP33886 IC. The chip inputs MCU signals to directly drive a solenoid without further ado.

The ultimate way to do this is to measure the inductance in real time and PWM the drive current. A seated relay or solenoid has a lot of inductance.

A uP could do it all, including burnout protection. Sounds like a product. Maybe someone has done it.

Some BLDC motor drivers sense inductance to commutate the drive.

No doubt these are all clever circuits, but they're all for solenoids rated for continuous duty. Solenoids rated for momentary duty will burn up at continuous currents required for "hold in." If they were AC types, it would be an entirely different story because many times the hold in is around 10%. But that's not the case for DC types. Power is not an issue for momentary types because the activation is so brief, and the application usually limits the duty substantially.

I'll get photos next week and post them.

I played with hypnosis when I was young, a friend who was a good sport asked to be told he was in Madison Square Gardens and giving a performance. He really seemed to be there as far as I could tell.

Hypnosis is weird, from my reference books back then they claimed that

Too many MPU signals required with the NPX33886.

Bob's version just needs the single "ON" signal, and it automatically reduces current after a predetermined time... I will have a watchdog circuit that disables the solenoid drive signals if the CPU locks up, Bob's circuit is for the odd time when the watchdog doesn't work for some unknown reason.

John :-#)#

Flipper control circuit for pinball machine

Abstract

A flipper control circuit is provided for a pinball machine having a flippper, a flipper switch for activating the flipper, means for holding the flipper in an actuated position until the flipper switch is deactivated, and a solenoid coil for controlling the movement of the flipper in response to the voltage applied to the solenoid coil. A first voltage is applied to the solenoid coil when the flipper switch is activated. A second holding voltage is applied to the solenoid coil when the flipper is in the actuated position, to hold the flipper in the actuated position until the flipper switch is deactivated. A switch electrically disconnects the first voltage from the solenoid coil when the flipper is in the actuated position.

...

When the solenoid plunger controlled by solenoid coil 24 nears the end of its stroke, an attached mechanical lever opens the end of stroke switch 22, thereby disconnecting the 50 VDC to coil 24. With the plunger resting against a plunger stop and the 50 VDC removed, a holding current is supplied to coil 24 via diodes 50 and 46. The solenoid coil 24 requires much less current to sustain the solenoid plunger against the plunger stop than to move the plunger, and thus the voltage supplied to the solenoid coil through diodes 50 and 46 can be far less than 50 volts.

--

This board is not for technicians so they can upgrade the rest of the game, it is for folks who want to repair a game they own with a minimum of fuss (because they are NOT technicians) and to make it as reliable as is reasonable.

John :-#)#

calculate the energy dissipated as heat (P = V² / R):\(P=\frac{28\text{\ V}^{2 }}{14.1\ \Omega }=\frac{784}{14.1}\approx 55.6\text{\ Watts}\)Because a pinball coil is tight, dense, and wrapped in an insulating plastic bobbin (with no heat sink or ventilation), it cannot shed heat. Nearly 100% of those 55.6 watts are t rapped, causing an immediate, exponential spike in temperature.

Class 130 (B) or Class 155 (F) polyurethane/polyamide magnet wire insulation. Th is thin enamel film is rated to withstand a maximum temperature of 130°C to 155 °C (266°F to 311°F).At a continuous 55.6W, the internal core of the coil reac hes this breakdown temperature within 3 to 5 seconds. The Result: The enamel beg ins to soften and degrade. While the coil might not be visibly smoking yet, its lifespan is permanently compromised because the insulation has broken down struc turally.

the physical geometry of the coil causes a runaway chain reaction: The Short-Ci rcuit Cascade: As the enamel melts, adjacent bare copper wires touch. This bypas ses whole sections of the coil, dropping the resistance significantly below 14.1 ohms. Current Spike: When resistance drops, current increases (I = V/R). If the resistance drops to 5 ohms, the current spikes from 2 Amps to 5.6 Amps, shoving the wattage up to 156 Watts.The Meltdown: At this point, the temperature instan tly rockets past 200°C (392°F). This is the melting point of the nylon/plastic coil sleeve inside the core. The sleeve warps, melts, and fuses to the metal pl unger, completely seizing the mechanism within 10 to 15 seconds. Concurrently, t he outer paper wrapper chars and smokes.

rchives like the Pinrepair Guides and community forums such as Pinside. Technici ans frequently document that if a driver transistor shorts out upon powering a m achine up, a failure to cut power within roughly 10 seconds will result in a "cr ispy," ruined coil that must be cut out and replaced.

hich, according to datasheet, trips in 1 second timeframe, well away from a dama ging duration. The part does have a 1 ohm initial resistance, but because the cu rrent is near 95% design value, there should be no problem there. And since your driver pulsewidths are 25-30 millisecond range, an activation repetition freque ncy of up to 5Hz ( times per second) , keeps the average DC current below the MF

-R030 Hold Current rating of 0.3A, which is 0% chance of nuisance trip.

t two and two together.

In my role as ham radio club secretary, it falls upon me to interpret cryptic tech talk. After about a half a dozen aborted attempts, the above character string displayed by my newsreader finally makes sense.

Someone offers the excellent idea to protect solenoid coils with a Positive Temperature Coefficient (PTC) thermister. Elsewhere in-thread someone seems reluctant to apply this fix to dual wound flipper solenoids. In My Opinion this fix ought to also work for dual wound flipper solenoids, an A-17875, for instance.

To protect coils, all you need do John, is to insert-soldier a PTC thermister between one coil wire terminal and its original connecting wire.

The transistor failures can be alleviated with Pease's circuit.

# # #

OK you guys, all of the ingredients for my new solenoid webpage finally gelled. The 555 pick-and-hold circuit presented elsewhere in- thread will drive an A-17875 from a grab bag bequeathed to me by the widow of a late relative. This relative ran a regional Wyoming coin-op business - back in the halcyon days when malls were viable and arcades blossomed within. If only my relative was still alive, he could coach me on how to effectively talk shop with John.

Cheap is a relative term. Anything that you do "per solenoid" has to be repeated for all solenoids. And, has to be present ON the solenoid as you want the driver to be identical circuits.

Then you have different flavors of solenoids. And, have to ensure the device is *in* the solenoid -- not just dangling off one of the terminals. (otherwise, it risks physical damage or "removal" when someone eventually discovers that the protection device has failed and will just bypass it to avoid having to purchase a replacement!)

That's why this is done in software. You can effectively code the duty cycle based on the SOA of each particular solenoid (load) in each specific application -- coil XYZ in one use may see different operating conditions than in some other use.

There have been numerous different approaches to this, in the industry, over the years. All have consequences (cost/price, reliability, serviceability, performance, etc.)

Consider the premise: the CPU has "lost its mind" AND the circuitry that is intended to guard against that has failed. What if the bridge that produces the unregulated supply fails? (hey, we're EXPECTING failures so why not imagine and protect against ALL of them?)

The problems with things like PTCs are:

- they can silently be activated countless times in a short period

- they have a finite lifetime

- they can fail

- they represent an additional manufacturing step & cost etc. This assuming there are no "bugs" in the system that the protection device is countering!

And, most importantly, they do nothing to address (or even indicate) the underlying problem!

One advantage of nonreseting devices is that it brings the failure to the attention of someone who can *possibly* take action to address it.

I use a *lot* of hammer drivers in my current design:

- to open/close skylights

- to open/close HVAC vents

- to control irrigation solenoids

- to control domestic water sources

- to isolate the system from the municipal water supply

- to backflush the water softener

- to unlock doors

- to control high current contactors etc. I.e., to perform actions that would typically require mechanical action from a human user.

I use "pulse on" and "pulse off" interfaces as they avoid static (stuck at) failures in the controls. They are also considerably easier to handle, in software, as you can "address" a single device at a time (instead of mapping N of them -- 8 -- into a single access and then having to wrap all such access in atomic operations -- which is what inevitably happens when hardware-types design interfaces without understanding software).

If the processor "goes south", then *it* should be reset to bring it to a known/safe state. The field should also be powered down as you have no idea what state it may be in.

AND, SOMETHING MUST BE MADE AWARE OF THIS *FAILURE*!

In my case, the PSE is signalled when the reset occurs and *it* decides when (if!) to release the node from the reset state.

If it notices a pattern in such events, it can opt not to provide power to the PD to ensure it is rendered inoerable (just like a human can unplug a device that is repeatedly "blowing a circuit breaker")

And, having done these things, can alert a user as to the reasons for its actions.

I am pretty sure John doesn't have an upstream "agent" that can perform these tasks automatically. *BUT*, he can involve the user to ensure a potential problem doesn't persist and morph into a more serious condition (can something catch fire?)

Are you, perhaps, going too far? Belts and braces?

What other "likely" failures might you expect (e.g., if this is a peripheral to a "CPU board", what if the CPU is not connected -- floating bus)? What is immediately upstream from the drivers (e.g., we used PIAs at one point and then realized that they can fail to initialize and all pins default to inputs -- floating high!)? Is there something that ensures the field is not activated "in transition" until you know the system is capable of imposing control?

Have you addressed the inevitable tinkerers who THINK they know enough to "see what's wrong"? What happens when supply is accidentally shorted to the low side of a coil (and thus the driver)?

First old-timer that I worked with (at an arcade in the Combat Zone) used a single length of wire as his sole troubleshooting tool; he'd connect two carefully selected (?) points in the circuit together and observe the result(s): "THIS coil is energized so let me connect this OTHER coil to its low side and see if it, too, pulls in -- before deciding the coil might be toast..."

(of course, now the driver for the first coil sees a doubled load...)

D-K suggests this sort of half-baked understanding to be more common than we'd like (I don't let anyone inside my boxes as I am sure even "professionals" would be hard-pressed to service them without documentation)

No way am I going to be asking my customers to modify their games, that would not be possible in most cases. The protection must be on the single MPU board which has all the electronics.

The solenoids are powered from the control transistors which source the

The only continuous duty coils are the flippers and they are always active and have two windings and End Of Stroke switches so rarely present a problem. A non-issue here.

I am not redesigning the original circuit, rather I'm just trying to make it more resistant to failure. Also all of the controlled coils are only fired momentarily, none of them are supposed to be left on at all.

Hence the Watchdog Reset circuit.

Risk of fire is extremely low, as long as the solenoid circuit breaker works as it should! There may be smoke however from the coil as it fails...

John :-#)#