brianw@microsoft.UUCP (Brian WILLOUGHBY) (12/22/89)
sb@pro-generic.cts.com (Stephen Brown) writes: >You favour getting the 65816's 6 volts from a resistive divider? This assumes >that the current it draws is the same for all operations, and I don't know if >this is safe to say. As the current drawn by the '816 changes, so will its >supply voltage, unless the resistive divider is very robust (in terms of >current) indeed. Whoever still clings to the resistor divider is certainly no Electrical Engineer, and I'm glad I don't own any products that are designed so poorly. In any event, I want to resurface an old question that (I think) was asked: someone mentioned the need to modify the motherboard when adding the 6 volt supply, yet someone else countered this. I think the point is that you need to avoid sending 6 volts to every chip in the Apple. Since the ribbon cable from the TransWarp GS sits in the 65C816 socket, I would assume that AE connected the 5 volt signal from the motherboard CPU socket to the 5 volt line on the W65C816 that resides on the accelerator. If you do not disconnect this contact at some point (on the motherboard socket, in the ribbon cable, or on the accelerator socket itself), then any 6 volt supply you generate will be *shorted* to the Apple 5V supply. Yikes! >Since a 13 megahertz 65816 is going to be a pretty penny (does anyone know how >much, in small quantity?), I'd probably spring for the two-buck regulator. Call Western Design Center, 602/962-4545, they have no distributors, and they only sell direct. Back in July, I ordered a 10 MHz W65C802 for $24.84, 4 MHz parts were only $17.80. Minimum order is $50, though, so you'd better have some friends who are also interested. Or you could just buy an extra chip for backup. Me? I purchased two 4 MHz and one 10 MHz chips, and only paid $5 for COD delivery. Then I sold a 4 MHz chip to a friend with a TransWarp. Brian Willoughby UUCP: ...!{tikal, sun, uunet, elwood}!microsoft!brianw InterNet: microsoft!brianw@uunet.UU.NET or: microsoft!brianw@Sun.COM Bitnet brianw@microsoft.UUCP
ericmcg@pro-generic.cts.com (Eric Mcgillicuddy) (12/26/89)
In-Reply-To: message from brianw@microsoft.UUCP In my original post, I mentioned that the 65816 have the traces to the 5V supply bypassed to prevent diffiulties with the other chips. I have thoroughly discarded the resistor divider idea. It was quick and dirty, but not the best solution. Several solutions are possible a 6..2V Zener is one, a 6V regulator is another. Choose what you can work with. I assume that the people using this idea are not electrical engineers so I attempted to keep it simple. Oh well. The power for the entire card comes from the slot, it is poor practice to draw power for a card from the motherboard. I would not buy a product that did and I'm sure the AE people would not design it that way. As for hacking the hardware to work, I would suggest that you limit yourself to 10MHz. The designer probably built the board for 10.5MHz and de-rated it to 7MHz. This is the rule of thumb for most designs, by reputable companies at least. The other components can thus withstand the increased speed and power dissipation. I would strongly reccommend not trying to push it to 13MHz, the professionals who did took precautions. You might get away with it, but I don't recall any statements as to the failure rate of these accelerators, might be as low as a few minutes. With the TWII released, does anyone know it a TWGSII is coming soon. It might be useful to wait for the real thing.