128MB v2.5 Very Strange Problem

Discussion of official MiSTer Project addons. (https://github.com/MiSTer-devel/Hardware_MiSTer)


damopinn
Posts: 10
Joined: Wed Oct 07, 2020 7:50 am
Has thanked: 3 times
Been thanked: 5 times

128MB v2.5 Very Strange Problem

Unread post by damopinn »

I had some v2.5 pcbs left over from when 64mb with 2x32 was a thing. When I saw some 64mb chips on aliexpress for $5 I decided to try to put them on the pcbs. The weirdest thing happens one chip tests ok at 140 and the other only at 80 so its pretty useless, I resoldered the chips in the other positions and I get the same results even though the chips are in the opposite posotions. I had a single chip board left over so I put the chips on these and both are bad all the way down to 70. The boards I soldered with 32mb are all fine at 140. So I don't think my soldering is bad. I was thinking maybe this was the reason for the update and the Ldo, The chips look genuine and the markings dont rub off with acetone or ipa, Does anyone have any idea what might be happening? I suppose I could get some v3.0 pcbs from jlcpcb but I'd like to know if anyone thinks it's worth it. I was never very motivated to pay top price as I have no nostalgia for neo geo and some how saturn works fine with my 2x32 abomination.

Xbytez
Site Admin
Posts: 479
Joined: Wed May 20, 2020 3:36 pm
Has thanked: 227 times
Been thanked: 805 times

Re: 128MB v2.5 Very Strange Problem

Unread post by Xbytez »

The fact you have individually soldered both 64MB chips onto the single chip memory board does seem to show they are faulty, as at one point sellers were producing 64MB memory modules on the single chip memory pcb boards, so I would have expected in your tests for at least one of the chips to work and not error if they were working correctly and genuine.

damopinn wrote: Fri Sep 29, 2023 5:25 pm

When I saw some 64mb chips on aliexpress for $5 I decided to try to put them on the pcbs.

Considering the retail price from legitimate sources such as Digkey or Mouser for the SDRAM chips, I would conclude that at $5 these from Aliexpress are either fake or faulty at this price point

damopinn wrote: Fri Sep 29, 2023 5:25 pm

The chips look genuine and the markings dont rub off with acetone or ipa.

Laser remarking of chips will not rub off with acetone or ipa.

damopinn
Posts: 10
Joined: Wed Oct 07, 2020 7:50 am
Has thanked: 3 times
Been thanked: 5 times

Re: 128MB v2.5 Very Strange Problem

Unread post by damopinn »

The only part that really makes no sense to me is that one of the chips apparently is ok on the dual board its always the one in the bottom position on the memtest, I've no idea which chip that is on the board. If you press 'c' you can choose which chip to test. I suppose there are some good chips in China as you can buy working sdram boards from some sellers. One day a new core will motivate me to get some genuine mouser chips. I hoped maybe it was a problem caused by the boards,

parawizard
Posts: 18
Joined: Tue Jan 02, 2024 8:31 am
Been thanked: 1 time

Re: 128MB v2.5 Very Strange Problem

Unread post by parawizard »

Could be with some heat something changed. Definitely irritating though

AngelicLiver
Top Contributor
Posts: 433
Joined: Fri May 29, 2020 8:50 am
Has thanked: 86 times
Been thanked: 120 times

Re: 128MB v2.5 Very Strange Problem

Unread post by AngelicLiver »

Very unlikely the act of soldering caused issues. You can safely assume that all chips from AliExpress are fake or binned defective, it is rampant and widely documented.

Post Reply