Defender CMOS replacement

retroACTIVE

Active member
Joined
Aug 17, 2007
Messages
2,300
Reaction score
12
Location
KEEDYSVILLE, Maryland
Anyone have a replacement 5501 kit for Defender... I've seen them around but they are way to expensive. I'd like to know what would work, if there is a self contained battery backed solution or not...

If I can't find them I'd like to make a bunch if possible.

Thanks
 
Did you perhaps mean 5101 ram? If so, I can help you on those. Im only guessing because im not familiar with a 5501 ram. Is there some info on this mod you can post a link to?
 
Mark Spaeth used to make an adapter to use a Simtek 25c48 in place of the 5101, but I don't know if they're available anymore. Tom Callahan makes an adapter for pinball machines. I asked him quite a while back if he thought they'd work for Defender, and he wasn't sure, but it is a 5101 replacement. Tom's 5101 page is here: http://www.pin-logic.com/SIMTEK ADAPTER.htm
 
If you want to buy the 5101 in qty pm me and we can work out a deal on them. Send me your target price on 10+ of them and I will see what I can do for you.
 
Mark Spaeth used to make an adapter to use a Simtek 25c48 in place of the 5101, but I don't know if they're available anymore. Tom Callahan makes an adapter for pinball machines. I asked him quite a while back if he thought they'd work for Defender, and he wasn't sure, but it is a 5101 replacement. Tom's 5101 page is here: http://www.pin-logic.com/SIMTEK ADAPTER.htm

You must be careful using Tom's adapter. He only uses one of the two 5101 chip enables. If you have a machine that requires both chip enables (i.e. Gottlieb pinball machine) then his adapter will not work. Regarding vids... depends on the specific machine.
 
You must be careful using Tom's adapter. He only uses one of the two 5101 chip enables. If you have a machine that requires both chip enables (i.e. Gottlieb pinball machine) then his adapter will not work. Regarding vids... depends on the specific machine.
As near as I can tell, Spaeth's adapter only uses one CE too. It looks like pin 17 (ce2) of the 5101 isn't used at all, and it connects pin 19 of the 5101 (~ce) to pin 18 of the 25c48 (~ce). The Simtek chip only has the one CE pin. I can see where a chip would care about one or two CE signals, but what difference does it make to the machine? Why does a Gottlieb pin care about the second CE if the replacement chip only needs one? Just when I think I'm getting a handle on these circuits I get a curveball like this. :)
 
As near as I can tell, Spaeth's adapter only uses one CE too. It looks like pin 17 (ce2) of the 5101 isn't used at all, and it connects pin 19 of the 5101 (~ce) to pin 18 of the 25c48 (~ce). The Simtek chip only has the one CE pin. I can see where a chip would care about one or two CE signals, but what difference does it make to the machine? Why does a Gottlieb pin care about the second CE if the replacement chip only needs one? Just when I think I'm getting a handle on these circuits I get a curveball like this. :)

The Gottlieb circuitry requires two CE's.
CE1(active low) comes from address bits 11 and 12. CE2(active high) comes from address bit 13 It takes the right combination of all three address bits to turn on the 5101. To use only one CE as done by the 25C48 then you need circuitry to combine the two original CE's into one.
The old Ballys/Williams used one of the CE's for address decoding and the other to signify power was good. Since the 25c48 no longer cared if power was good, this made it easy to convert to a single CE device.

Ed
 
The Gottlieb circuitry requires two CE's.
CE1(active low) comes from address bits 11 and 12. CE2(active high) comes from address bit 13 It takes the right combination of all three address bits to turn on the 5101. To use only one CE as done by the 25C48 then you need circuitry to combine the two original CE's into one.
The old Ballys/Williams used one of the CE's for address decoding and the other to signify power was good. Since the 25c48 no longer cared if power was good, this made it easy to convert to a single CE device.

Ed
Thanks for the explanation Ed, makes sense now (you learn something new every Monday...). Since Spaeth's adapter only uses the one CE and is supposed to work in Defender, I'd guess Tom's adapter is good there. Having said that, Defender is notoriously finicky about it's 5101. I have a brand new Philips chip that can keep scores and settings, but can't pass the cmos test. I have an Intel 5101 that seems to work quite well, and one other (I forget the brand) that doesn't get along well at all with a switching power supply.
In that sense it's tough to say whether Tom's adapter will work. I've been told that a chip like the ds1220 (battery backed) is a bit more forgiving than the Simtek, but I haven't tried one yet.
 
Thanks for the explanation Ed, makes sense now (you learn something new every Monday...). Since Spaeth's adapter only uses the one CE and is supposed to work in Defender, I'd guess Tom's adapter is good there. Having said that, Defender is notoriously finicky about it's 5101. I have a brand new Philips chip that can keep scores and settings, but can't pass the cmos test. I have an Intel 5101 that seems to work quite well, and one other (I forget the brand) that doesn't get along well at all with a switching power supply.
In that sense it's tough to say whether Tom's adapter will work. I've been told that a chip like the ds1220 (battery backed) is a bit more forgiving than the Simtek, but I haven't tried one yet.

I know some games that use a 5101 are finicky about the (speed) of the 5101....might be the issue you're seeing.

Edward
 
I know some games that use a 5101 are finicky about the (speed) of the 5101....might be the issue you're seeing.

Edward
That's possible, but the Philips chip is quite a bit faster than the original (I don't remember the exact speed off the top of my head), so if it's a speed issue then it cares about a chip being too slow or too fast.
 
That's possible, but the Philips chip is quite a bit faster than the original (I don't remember the exact speed off the top of my head), so if it's a speed issue then it cares about a chip being too slow or too fast.

The Philips parts are nice parts -- 150nS. I have never seen a problem with them.

There are problem 5101's out there that run as slow as 600, 800 or even 1000nS. These are the ones that cause problems. Usually these are -real old- Intel, SGS, RCA and others.
Also must be careful with RCA parts -- some have CMOS-only interface pins that are not compatible with TTL logic.

Ed
 
The Philips parts are nice parts -- 150nS. I have never seen a problem with them.
I only have one Philips chip so it may be an issue with that individual part, but I've used it a few times and it never passes the cmos test, even though it does hold scores and settings. I read in rgvac that others had problems with Philips chips in Defender too. It seems to be a Defender-only thing.
 
Looks like DS1220 or equivalent would fit the bill. The 5101 has two chip enables one positive and the other negative. In the case of Defender the positive chip enable is connected to the reset circuitry which guards against enabling the chip when the +5 is not ready and the 1220 already has an internal power monitor circuitry to prevent this ... so the positive chip enable should not be needed for Defender. All other signals are compatible. And of course strap the unwanted address (A8-A10) and data (DQ4-DQ7) to ground.

I've created a pcb prototype for the 1220 part. I will give it a try when I have some time.
 
Last edited:
Back
Top Bottom