Page 1 of 1

Extra resistor or capacitor on some MBC5 cartridges

Posted: Fri Aug 12, 2016 1:33 am
by nitro2k01
This picture was posted in the Facebook group GameBoy Collectors. It shows two MBC5 carts, where one has an extra component attached. This could be a resistor, but also quite possibly a capacitor judging from the base color and shape. One of the component is connected to a convenient grounding point, the negative battery tab. To save you the trouble, the other side is connected to cart bus A14 which goes the A14 input on the MBC5 chip, and nowhere else. The ROM and SRAM chips' A14 lines come from the decoded outputs of the MBC5, so again, not directly connected to this A14 address line.

I'm assuming this component was added to fix some issue with an early revision of the MBC5 chip. Either a pulldown resistor or a noise suppression cap. Has anyone noticed this extra component before and investigated exactly what problem it's solving?

Re: Extra resistor or capacitor on some MBC5 cartridges

Posted: Fri Aug 12, 2016 6:02 am
by Jarhmander
It rather looks like an axial inductor from what I can see...
Image
As for what problem it solves, I guess it would be effective as a protection against really short electrical shorts (like a small glitch), but it's strange that it's connected to the negative side of the battery.

Well we should test the component. Can you get a closer picture of that?

Re: Extra resistor or capacitor on some MBC5 cartridges

Posted: Fri Aug 12, 2016 6:23 am
by nitro2k01
An inductor would make no sense. If A14 stayed high for any length of time (such as when executing code in the selectable bank region) A14 would be shunted to ground and pull A14 low hard, and potentially draw excessive current from the CPU's output transistor in the process. I have however seen capacitors in that kind of package, even if they are rare.

I've asked the guy to open the cart and note the colors of the bands, but he said that would have to wait a couple of days. But I figured maybe someone here would have a cart with this component on it.

Re: Extra resistor or capacitor on some MBC5 cartridges

Posted: Fri Aug 12, 2016 12:42 pm
by lidnariq
That light green color seems to be a kind of axial capacitor, IIRC.

I'd really love to find a listing of "body color to component type" but I haven't.

Re: Extra resistor or capacitor on some MBC5 cartridges

Posted: Fri Aug 12, 2016 7:23 pm
by Rahsennor
I have very little knowledge on the subject myself, but I showed the pic to a technician I know and he said it looks a pulldown resistor, either 15k or 15M. That's admittedly half guesswork though, since the last two stripes are barely visible.

Re: Extra resistor or capacitor on some MBC5 cartridges

Posted: Sat Aug 13, 2016 8:09 am
by Jarhmander
nitro2k01 wrote:An inductor would make no sense. If A14 stayed high for any length of time (such as when executing code in the selectable bank region) A14 would be shunted to ground and pull A14 low hard, and potentially draw excessive current from the CPU's output transistor in the process. I have however seen capacitors in that kind of package, even if they are rare.

I've asked the guy to open the cart and note the colors of the bands, but he said that would have to wait a couple of days. But I figured maybe someone here would have a cart with this component on it.
Hmm, missed the A14 bit somehow. That would make no sense indeed.

If this is a pull down resistor, then when does the address bus go floating?

Re: Extra resistor or capacitor on some MBC5 cartridges

Posted: Sat Aug 13, 2016 8:20 am
by nitro2k01
Jarhmander wrote:If this is a pull down resistor, then when does the address bus go floating?
I'm thinking, potentially during halt (power save) mode. If it's a capacitor, maybe there's a race condition that A14 needs to transition after A15 for correct function for some weird reason? I would have guessed the opposite if that were the case since A14 controls the state of the upper ROM address bits, where A14=0 sets them all to 0 to access bank 0, and A14=1 sets them to the selected bank. A15 is basically chip select for the ROM chip. It would then make sense if A14 had to lead and A15 lag, to ensure that the address bits are valid when the ROM chip is selected.