[NTLK] eMate memory upgrade

Tim eft37 at earthlink.net
Mon Sep 19 12:51:39 EDT 2011


Hi Frank-
I wonder if you ask PCBman/David a little more, as I remember getting his eMate expansion board and him offering a discount towards a new bigger board. I wonder if his design had "growth" potential also.
-Tim

>
>Message: 3
>Date: Sun, 18 Sep 2011 23:37:14 +0200
>From: "Frank Gruendel" <newtontalk at pda-soft.de>
>Subject: Re: [NTLK] eMate memory upgrade (was: High resolution images
>	of	eMate mainboard)
>To: <newtontalk at newtontalk.net>
>Message-ID: <06AC36A85E7E4FC1A4A74D6E28444C32 at xpdesktop1>
>Content-Type: text/plain;	charset="us-ascii"
>
>>> I've added images of an eMate mainboard that might come in handy for 
>>> all who like to tinker with or repair eMates.
>
>> Nice. There are two unpopulated RAM chips. Do you have an organ donor
>> that you could use to add DRAM and NVRAM to a working board and see what
>happen?
>
>If you haven't a clue what I'm talking about, this image might help...
>
>	<http://preview.tinyurl.com/3vsm6kd>
>
>Also, you might want to have a look at the Apple eMate memory upgrade
>discussion: 
>
>	<http://support.apple.com/kb/TA38458?viewlocale=en_US>
>
>Actually, there's but one unpopulated installation space. RC1 looks like
>one, but isn't.
>
>The unpopulated installation space is most likely not for DRAM. The DRAM is
>in U14 and U15.
>
>I assume that when the eMate was developed, the developers weren't sure what
>size of Flash RAM chips would be available at a reasonable price when the
>eMate would be ready to be shipped. This happens quite often, and normally
>you'd layout your board in a way that will allow for different chips to be
>installed.
>
>So they layouted the board in a way that at least three different sizes
>could be installed: U16 (smallest), U17 (medium) and either U19 or U9
>(largest). As we can see, they settled for a LH28F016 Flash RAM chip in U19,
>which provides the 2MB x 8 that the eMate comes with by default.
>
>If the Apple developers DID have a Flash RAM extension in mind, it can be
>safely assumed that they would have chosen two identical chips. So an
>additional chip would have to be installed in U9, since U9 is the only
>unused installation space that has 56 pins like the LH28F016 in U19. But
>maybe they simply put U9 there for a chip with yet another pinout.
>
>This being so, and yours truly being a cautious person, I recently asked
>David aka PCBMan for an opinion on the chances of success in this kind of
>surgery. He told me that U9 is not for an additional chip, but for a
>DIFFERENT one. Since he's usually well informed, I refrained from simply
>soldering a donor chip there. Instead, I checked if this installation space
>could be for a LH28F016 at all. This would require at least that the supply
>voltage and ground pins are the same. Alas, even that isn't the case. So,
>unfortunately, David is right, and such an attempt would fail miserably.
>
>However, if you look at one of the eMate upgrade boards, for example the
>MicroTech one,
>
>	<http://preview.tinyurl.com/3moejva>
>
>you'll see that a Flash RAM upgrade with just one additional LH28F016 Flash
>RAM chip is definitely possible. But I do not intend to design a new eMate
>memory upgrade board.
>
>But maybe there would be another way...
>
>The LH28F016 has three unused pins. So it would be possible to put up to
>eight times (2 x 2 x 2) the Flash RAM into such a case without the need of
>additional pins.
>
>Given the fact that the original chip's name is LH28F016, chances are good
>that chip manufacturers would have named these chips LH28F032, LH28F064 and
>LH28F128. Believe it or not, but there's a LH28F032 out there that has the
>same pinout as the LH28F016 except for pin 3. In the LH28F016 this pin is
>not connected, in the LH28F032 it selects between the lower and the upper
>half of the memory.
>
>Just to prove how helpful it is to have an unpopulated board, I had a look
>at pin 3 of U19. If Apple had never wanted a larger memory chip, this pin
>wouldn't be connected to anything. But it is. It is connected through a via
>to U18 on the opposite side. U18 is a five pin IhaveNoIdeaWhat, and it is
>(not surprisingly) not installed.
>
>Depending on the foresight of the Apple developers, it might have been
>within the realms of possibility that 64MB and 128MB chips would work, too.
>Unfortunately, though, the other two unconnected pins of the LH28F016 aren't
>connected to anything on the board, so this isn't possible. Apart from that,
>there doesn't seem to be a LH28F064, and the only LH28F128 I've found
>doesn't come in the required package.
>
>The next task would be to find out what U18 would have been, had it been
>installed. Once we know this, it shouldn't be much of a problem to figure
>out what other components might need to be added around U18 to make the
>larger size chip work. Replacing the Flash RAM chip wouldn't be easy, but it
>can be done. Given the fact that the board was designed for a larger chip, I
>have no doubt that the eMate's OS would support it. Especially since in the
>2x00 it'll work quite happily with twice the amount of Flash RAM that the
>eMate comes with.
>
>Frank
>
>-- Newton software and hardware at http://www.pda-soft.de
>
>
>
>
>------------------------------




More information about the NewtonTalk mailing list