Boot routine/ bank selection?

Discussion of hardware and software development for Super NES and Super Famicom.

Moderator: Moderators

Forum rules
  • For making cartridges of your Super NES games, see Reproduction.
VIP Quality Post
Posts: 27
Joined: Thu Jan 24, 2019 7:53 am

Re: Boot routine/ bank selection?

Post by VIP Quality Post » Tue Mar 03, 2020 7:24 pm

Here is my drawing.
Attachments
output.pdf
(105.38 KiB) Downloaded 74 times

lidnariq
Posts: 9492
Joined: Sun Apr 13, 2008 11:12 am
Location: Seattle

Re: Boot routine/ bank selection?

Post by lidnariq » Tue Mar 03, 2020 7:50 pm

Nothing immediately obviously wrong there. I mean, pedantically you should have an extra OR gate combining /CART and /RD to the 74'257s, but that's semantics instead of a red flag.

You said you measured the voltages on the ROM (not just the card edge) when it was in isolation? Have you tested whether low voltages are actually driven instead of floating (by e.g. comparing to the +5V supply instead of comparing to ground) ? You also said you tried tying the ROM's /G and /E inputs to ground and deferred bus control to the '257s?

VIP Quality Post
Posts: 27
Joined: Thu Jan 24, 2019 7:53 am

Re: Boot routine/ bank selection?

Post by VIP Quality Post » Tue Mar 03, 2020 8:16 pm

lidnariq wrote:
Tue Mar 03, 2020 7:50 pm
You said you measured the voltages on the ROM (not just the card edge) when it was in isolation? Have you tested whether low voltages are actually driven instead of floating (by e.g. comparing to the +5V supply instead of comparing to ground) ? You also said you tried tying the ROM's /G and /E inputs to ground and deferred bus control to the '257s?
Yup, I measured Q0-Q15 off of the ROM, but it was referenced to GND. I'll check with reference to +5V later.
Yes to driving the /G and /E gnd and only leaving the '257 for bus control.

I think I will set up a PCB that only has the ROM, '257s, and CIC setup without setting up SRAM. I am using some older 6264 chips that I had laying around from previous projects. As far as I am aware, they work (removed from functional electronics), but it could be that they are bad and I'm glossing over them. I'll write back later with whatever the results are.

lidnariq
Posts: 9492
Joined: Sun Apr 13, 2008 11:12 am
Location: Seattle

Re: Boot routine/ bank selection?

Post by lidnariq » Tue Mar 03, 2020 8:21 pm

... And the ROM drives the data pins low / leaves them floating, even though you know it's erased (i.e. all $FFFF) and /G and /E are tied low? There's something Very Suspicious about that—that just doesn't make sense.

I can't imagine you'd've made this mistake, but you haven't got the ROM in backwards, right?

VIP Quality Post
Posts: 27
Joined: Thu Jan 24, 2019 7:53 am

Re: Boot routine/ bank selection?

Post by VIP Quality Post » Tue Mar 03, 2020 8:58 pm

lidnariq wrote:
Tue Mar 03, 2020 8:21 pm
... And the ROM drives the data pins low / leaves them floating, even though you know it's erased (i.e. all $FFFF) and /G and /E are tied low? There's something Very Suspicious about that—that just doesn't make sense.

I can't imagine you'd've made this mistake, but you haven't got the ROM in backwards, right?
No, I checked it's in correctly and it's getting Vcc and Vss on the proper pins when the cart is powered, so that should also suggest it is in right. This is why I'm beginning to suspect the RAM as some culprit.

lidnariq
Posts: 9492
Joined: Sun Apr 13, 2008 11:12 am
Location: Seattle

Re: Boot routine/ bank selection?

Post by lidnariq » Tue Mar 03, 2020 9:04 pm

Just to make sure I'm understanding correctly, you've verified that the voltages on the ROM's data pins are high, and the value becomes low on the other side of the '257s?

VIP Quality Post
Posts: 27
Joined: Thu Jan 24, 2019 7:53 am

Re: Boot routine/ bank selection?

Post by VIP Quality Post » Tue Mar 03, 2020 10:49 pm

Wow, just keep finding weird stuff. Maybe I bought some bad chips or something. I removed all the parts from the boards. I checked the EPROM again (still has correct data in all banks). I stuck it on a breadboard with the two '257s. With the 257 not hooked up, I can read the memory fine. With the '257 hooked up, it causes some really weird oscillation on the ROM to mux bus. It does not show up on the data bus. I was just checking Q0-Q15 with my multimeter before, so that is probably why I wasn't catching the waveform. I attached a pic of it here. I will just get some more and see if the next handful of chips I get will replicate the same behavior. It seems too sharp and regular to be an issue with my probe connection/setup, plus I stepped away from the desk and let it run with nothing touching it continued to have this periodic behavior.
20200303_213543.jpg
To test the SMD ICs I mounted them on some breakouts I made a few years ago, that are known working (what can go wrong, right ? :oops: )
20200303_213552.jpg
I have them both hooked up on the same power supply, it's 5V 8A so I know there is more than enough current for the circuit to draw. I was getting 11mA with the ROM and 40mA with both of the '257 connected. That is below what the spec for the 257 (70mA) but that could be peak. It's difficult to tell if this is with all of the muxes I have (10pcs) since they need to be mounted to the breakout for convenient testing.

VIP Quality Post
Posts: 27
Joined: Thu Jan 24, 2019 7:53 am

Re: Boot routine/ bank selection?

Post by VIP Quality Post » Tue Mar 03, 2020 10:53 pm

As an addendum, I see the oscillation on all channels of the EPROM that are connected to the 257s. The ones that are floating seem OK, or at least I haven't been able to find it on the scope (yes, I am using triggers)

lidnariq
Posts: 9492
Joined: Sun Apr 13, 2008 11:12 am
Location: Seattle

Re: Boot routine/ bank selection?

Post by lidnariq » Tue Mar 03, 2020 11:04 pm

That looks like 60Hz that went through some kind of digital TTL buffer (thus the asymmetric 75% duty cycle). Check your grounding.

Oziphantom
Posts: 856
Joined: Tue Feb 07, 2017 2:03 am

Re: Boot routine/ bank selection?

Post by Oziphantom » Wed Mar 04, 2020 1:25 am

some decoupling caps might also help. I've seen some odd things happen when something browns out.

VIP Quality Post
Posts: 27
Joined: Thu Jan 24, 2019 7:53 am

Re: Boot routine/ bank selection?

Post by VIP Quality Post » Thu Mar 05, 2020 9:51 pm

Sorry I haven't replied in a few days, been busy with work.
lidnariq wrote:
Tue Mar 03, 2020 11:04 pm
That looks like 60Hz that went through some kind of digital TTL buffer (thus the asymmetric 75% duty cycle). Check your grounding.
That's interesting, I've never seen that before. I will check again this weekend.
Oziphantom wrote:
Wed Mar 04, 2020 1:25 am
some decoupling caps might also help. I've seen some odd things happen when something browns out.
I have a few on the PCB, I think i used .1uF right off the IC power supplies and a 22uF for a general filter between my 5V and GND planes. I'll toss some on the breadboard and see if that improves whatever effect I'm seeing here.

VIP Quality Post
Posts: 27
Joined: Thu Jan 24, 2019 7:53 am

Re: Boot routine/ bank selection?

Post by VIP Quality Post » Mon Apr 13, 2020 12:26 pm

Wow! Between work and the plague, it took a while to get back to this. I found the issues. While I had the due diligence to check if my programmer was working fine, and my board was fabbed fine, I never looked at my EPROM adapter. I just got one off eBay a while back and assumed it worked. I found some swapped traces on the board, which would explain why it could verify but not work right on my board.

Thanks for all the help, lidnariq and Oziphantom.

Post Reply