Bally E series Power Supply Woes

[ramegoom]

I'm in the process of restoring several E2000 slots, and have concluded that the power supplies are unreliable junk. The bridge rectifiers somehow burn themselves off of the board, the heatsinks on them are bordeline inadequate - assuming the mounting screws don't loosen - which they all seem to do, and the big capacitors die a miserable death, probably due to years of heat cycling, and finally, the pin headers lose their solder connections to cold joints. Again, heat cycling and vibration a major factor.

I can repair them, and watch them fail again. Single sided PC board design is not a good thing here. Kind of reminds of the planned obsolescence the automakers used back in the 70's.

So, now I'm going to clone these boards, if for no other reason, to hedge the reliability issue. I could use 105degC components, larger bridge rectifier that produces less heat, and double-sided PC board to strengthen the solder connections.

But before I do, does anyone know if this has been done before?

[Amachanic]

As far as I know, no one has tried to build a better power supply? I my self just rebuild and repair the old ones. I'd be interested in seeing a new version when done.

Gary

[ramegoom]

Yeah, that's what I have been doing on the original boards, but it seems it's been done several times on some of them. Pads under the bridge rec are mostly just swimming over a layer of burned epoxy laminate. And those pin headers, many of them have shed their tin plating on the pins, exposing the oxide-ridden layer of contact. I can pretty much trace many of the odd problems the machines are showing, to the connections there.

I have several Williams-Gottlieb-Bally pinball machines in my game room, and they all have had issues with those Molex pin connectors. So it's a given, that's one of the fixes in store for the slots.

[dhellis]

I'm in the process of restoring several E2000 slots, and have concluded that the power supplies are unreliable junk. The bridge rectifiers somehow burn themselves off of the board, the heatsinks on them are bordeline inadequate - assuming the mounting screws don't loosen - which they all seem to do, and the big capacitors die a miserable death, probably due to years of heat cycling, and finally, the pin headers lose their solder connections to cold joints. Again, heat cycling and vibration a major factor.

I can repair them, and watch them fail again. Single sided PC board design is not a good thing here. Kind of reminds of the planned obsolescence the automakers used back in the 70's.

So, now I'm going to clone these boards, if for no other reason, to hedge the reliability issue. I could use 105degC components, larger bridge rectifier that produces less heat, and double-sided PC board to strengthen the solder connections.

But before I do, does anyone know if this has been done before?

The single sided board probably would be OK as long as the heat was kept to a minimum. I don't know of anyone that has
attempted a redesign but you need to be able to maintain the ZC voltage as well as the 10 volt UR voltage and  5 volts.
I know that switching power supplies can deliver more current and they generate less heat, my concern there  would the
frequency of the switcher may cause the MPU to reset. On the connector for the power on the MPU board is one pin (11) that
is designated as an 'Antenna' the antenna runs around the board so if RF is detected it will send the MPU into a reset.

Really though the biggest problem is caused by heat so I wonder if just adding a cooling fan would be a better option.

[ramegoom]

These machines all have that nasty cooling fan on the upper right side (the one that accumulates an indescribable clot of casino dust). Not sure how effective it is though, seems to suck air from the outside and blow it out at the coin tray.

I'm thinking I wouldn't try to re-invent the wheel, but rather, just clone what's on there. Except for the 8 amp bridge. I'd use a 25 amp bridge instead (larger package) and same 12,000 mfd. cap except use a higher temperature rating (probably the biggest expense on the board). The rest can be the standard 7805 regulator on a small heatsink, and 4004 diodes for the rest. Couple of standard capacitors and resistors, and a single RF choke. Seems pretty basic.

As for the antenna, are you saying the trace on the bottom of the board is the actual RF pickup? What kind of RF is it looking for? What effect does RF have on the CPU? I'm assuming low frequency RF, like AC hum. That's very interesting.

[dhellis]

I don't think it is looking for AC hum, according to the Slot Simulator Operating Test/Trouble Shooting Guide (FO 650-25, for K661 tester) the antenna is looking
for the presence of high level RF interference. It is tied to the 555 time and will send a high to pin 2 which eventually sends a reset to the CPU. As long
as the RF interference is present the CPU will remain in reset.

[ramegoom]

Hmm....must have been some sort of cheat device back then. I can't imagine anything otherwise. Does the manual go into the reason for that RF? I tend to believe the watchdog portion of the CPU can be taken out of the picture for these old "home use" machines.

Curious why Bally went to that RF detection.

[dhellis]

Hmm....must have been some sort of cheat device back then. I can't imagine anything otherwise. Does the manual go into the reason for that RF? I tend to believe the watchdog portion of the CPU can be taken out of the picture for these old "home use" machines.

Curious why Bally went to that RF detection.

Like you, I was thinking of something along anti-cheating. No the manual doesn't go into the reasons at all.

[ramegoom]

Aha, that has to be it. So, I can disable that feature pretty easily on the new supply board - looks like I'll have to prototype it first.

Now you really got me wondering how an RF signal can "fool" the ECU :-)

[dhellis]

Aha, that has to be it. So, I can disable that feature pretty easily on the new supply board - looks like I'll have to prototype it first.

Now you really got me wondering how an RF signal can "fool" the ECU :-)

OK so just to be sure that I am not confusing you or others, the antenna is on the MPU board, not on the power supply.
I mentioned the antenna in reference to using a switch mode power supply and because the frequency might have a
negative impact. Most switch mode supplies will use frequencies between 100 Khz to 1 Mhz. The lower frequency may
not have any impact but a 1 Mhz signal is 50% of the CPU frequency and that may lead to stability issues.

[ramegoom]

OK, got it. Thanks for the clarification and your responses. There are three different voltages that are produced by that power supply: 5V regulated, 7.5v unregulated, and 10v unregulated but heavily filtered. As I understand it, the 5v is for perimeter switches, and the 10 volt is for the heavy loads. 7.5 is unfiltered and can drift substantially depending on load (which it doesn't seem to have) and it also has a small resistive load.

While on the E2000 subject and have the attention of a Bally expert, I find it interesting how the sound board gets its power, and could never understand it. The few machines i have with the actual sound system have an AC transformer dedicated to it, but oddly, the primary winding is 50 volts and the secondary is 10 volts. Why 50 volts, not 120 volts? It gets its 50 volts from the main transformer, not the AC line.

The only reason that makes sense to me is that the main transformer is set up to run EITHER on 110 vac or 220 vac, so using a sound system transformer made for 50 volts would work on either scenario, but if it were straight to the line, you'd need to have a dual primary transformer, or, a separate part for 220 vs 110. Could I be on track?

[dhellis]

I don't have a real answer for this, other than possibly was designed that way to prevent a large popping sound. The 50 volts
is not immediately available, it goes through the delay relay. I really have no idea because I have never seen an E series with a
sound card.

Your explanation sounds  reasonable, you wouldn't have to change anything if you wished to run on 220v

[ramegoom]

Well, give me a few weeks, and dhellis, I may be calling on you to critique my creation. I would feel much better knowing that the power supply is reliable and not subject to crapping out indiscriminately.

[dhellis]

I would like to see it and am sure there are others that will be interested as well.