I-Appliance BBS
The Official Source for Internet Appliance Upgrades and Mods

Click Here!
BBS Main List | Sign In | Sign Up | Search | Help | Linux-Hacker.netReply to Thread | Printer |

Home / I-Opener Areas / I-Opener Technical Stuff
V1/V2 mod for 2.2v Vcore
Simple mod to V1/V2 to support AMD K6-III

New MessageV1/V2 mod for 2.2v Vcore (modified 0 times) Linuxguru
Profile | Email
For V1/V2 units without the divider consisting of R342/R343, it is possible to make a simple modification to drop the reference voltage below 2.5v. As has been mentioned in other threads, the Vin at pin 3 of U16 is obtained from the junction of a pin on Q28 (possibly collector/drain) and R266 (100K). The other end of R266 has 2.5v from the voltage reference. Q28 is probably part of the voltage shutdown circuitry. It operates at cut-off in normal operation and does not enter this calculation.

The mod consists of a resistor soldered from pin 3 of U16 to ground, to drop Vin from 2.5v. Pin 4 of U16 is ground, so it is possible to solder a 0603 or 0402 resistor between pin 3 and pin 4 of U16 (10358). Since R266 is 100k, a 100k resistor between pin 3 and 4 drops Vin to 1.25v. By a curious coincidence, R202 is 10K and R203 is 7.5k on my V1, giving a Vcore of almost 2.2v (rated Vcore of K6-III) with a Vin of 1.25v on pin 3 of U16. To get other values for Vcore, R202/R203 may be adjusted suitably.

To summarize: Enable split voltages by moving R60/R61 to R184/R185. Solder a 100k resistor between pin 3 and 4 of U16 on a V1, to drop Vcore to 2.2v for the AMD K6-III. Solder a resistor in parallel to R302 to get a higher current limit if necessary, for clocks higher than 3 x 66 (default).

09-04-2000 13:16:32

New MessageRE:V1/V2 mod for 2.2v Vcore (modified 0 times) ravton
Profile | Email
Since nobody else has said it, Thanks ALOT for clarifying this linuxguru. It's hard to pull the meat out from 5 different threads, all of which are 40+ messages.

Time to warm up the iron...

Thanks!

ravton@yahoo.com

09-06-2000 19:11:16

New MessageRE:V1/V2 mod for 2.2v Vcore (modified 0 times) Kudzu
Profile
Just to double check, the pin order on U16 is like this?

5 6 7 8
-----
| |
-----
1 2 3 4
(-)

(-) Bridge with 100k SMT resistor here.


I tried this with an 805 style resistor and it dropped mine down to around 2.2vDC. The only problem now is getting it to boot, it black screens on me. I hope that is just the bios pushed in too far.

Has anyone noticed that the k63 gets real hot quickly using these mods?

Kudzu

09-11-2000 22:53:11

New MessageRE:V1/V2 mod for 2.2v Vcore (modified 0 times) Kudzu
Profile
Crap. This board hates ASCII diagrams.

(-) should be at 3-4.

Kudzu

09-11-2000 22:55:49

New MessageRE:V1/V2 mod for 2.2v Vcore (modified 0 times) Linuxguru
Profile | Email
Kudzu, you did the split-voltage mod also, right? Without that, the CPU is almost certain to get hosed. Also, at 2.2 v, it would be advisable to stick to 200 MHz to avoid tripping the current limit.
09-12-2000 01:48:31

New MessageRE:V1/V2 mod for 2.2v Vcore (modified 0 times) vailr
Profile
What about 1.8 volts for the K6-2+ and K6-3+ CPU's? And can anyone "re-engineer" the power supply so that NO extra heat is produced on Q16, etc.
09-12-2000 15:41:28

New MessageRE:V1/V2 mod for 2.2v Vcore (modified 0 times) Kudzu
Profile
Yes, I moved r60/1 to 184/5, and right now it's still set to 200.

It's probably just the bios. I'll have to resolder the iopener to the original configuration, fix it and try again.

Vailr, that's a very good question. I've seen some mention of a secondary power supply somewhere in the k63 discussion threads, but no mention of how to put it together?

It would be great if it were possible, I've got a 1A 1w-35w project kit power supply ($15) sitting here on my desk just itching to be used.

09-12-2000 16:58:18

New MessageRE:V1/V2 mod for 2.2v Vcore (modified 0 times) vailr
Profile
Not talking about adding in a 2nd PS. Would be better to re-engineer the existing PS to provide 1.8 volts, but without the extra heat produced by simply adding resistors. The question remains as to whether the AMPS required by the 500 Mhz. K6-2+ or K63+ could be supplied with the existing PS unit (after the 1.8 CPU voltage and "no heat" circuitry mods).
09-12-2000 19:34:28

New MessageRE:V1/V2 mod for 2.2v Vcore (modified 0 times) Turbo3
Profile
Yes, you can run a K6-2+ at 500MHz in the iopener using the built in supply. You must lower the value of R302 to allow the maximum current from the 3.5 volt supply and adjust the vcore down to 1.8 volts. This is 0.2 volts lower than the value marked on the chip (ie 2.0 vcore) but is necessary to keep the amps low and keep the temp of Q16 down. At this speed it takes 7.18 Amps from the 3.5volt supply so a 1Amp external is way too small. You would need to add the Maxim MAX1711 evaluation kit (about $60) which is good for 12Amps. From all the testing I have done, the maximum current you can get from the 3.5 volt supply is a little over 7 amps then the voltage drops and it auto shuts down.

As for modifying the built in supply to generate 1.8 volts directly, that is not possible since the Maxim chip only supports two switching voltage regulators. One for 5 volts and the other for 3.5 volts. The processor and other chips still need 3.5 volts so the only thing you can do is drop the voltage through Q16 to get down to 1.8 for the core.

09-12-2000 20:39:18

New MessageRE:V1/V2 mod for 2.2v Vcore (modified 0 times) Kudzu
Profile
So how would I go about installing the MAX1711? And what does it replace on the board, the MAX1631EAI?
09-13-2000 00:37:36

New MessageRE:V1/V2 mod for 2.2v Vcore (modified 0 times) vailr
Profile
One other thing to keep in mind is the "Power Now!" feature of the K6-2+, which will allow running on only 1.4 volts when the CPU is at 200 Mhz.
Could probably run with the stock PS at 1.6 or 1.7 volts at 400 Mhz. The limiting factor is the Q16's heat. Anandtech has an article about this CPU here: http://www.anandtech.com/showdoc.html?i=1318
09-13-2000 10:50:59

New MessageRE:V1/V2 mod for 2.2v Vcore (modified 0 times) Tackhead
Profile
Turbo3: How's the 3.5V generated on the Maxim chip?

Just to play devil's advocate, I wonder how much of the system would run stably if we were to drop this to 3.3V? Or 3.2V?

I realize any attempt to lower the 3.5V will require lots of stability testing. Instead of having only *one* chip to worry about (a CPU running 1.9V instead of 2.2V spec), we've got dozens (most everything on the motherboard running at 3.3V vs. 3.5V spec) of chips to worry about, each from different manufacturers, and each with different tolerances.

(This is turning into a limbo contest: "How low can we go"?

09-15-2000 13:27:32

Reply to Thread | Printer |
All times are PSTPowered by UltraBoard v1.62



Copyright © 2000, Netmake Inc. All Rights Reserved.
See Terms and Conditions for more information.




i-opener opener laptop notebook computer help drivers dll free windows dos repair fix linux mac macintosh 2000 95 98 nt pc configure hardware software sound video netscape explorer network networking lan wan software cmos fat bios printer card mouse modem ide scsi cd rom controllers scanner tape hard drive cgi scripts source code mp3