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
A stable 600 mhz is possible
At least on one machine

New MessageA stable 600 mhz is possible (modified 0 times) Ragnar
Profile
Gang,
With the right combination of parameters, a stable 600 mhz machine is possible. Start with am AMD K6-II+ or an AMD K6-III+. My particular processor is reported by Sandra to be an AMD K6-III with model info as K6-2+/III+ (mobile) 450-600 mhz 2.0V. When I bought it from SoftwareandStuff it was listed as a K6-II+ 450mhz. Hardwire your machine to boot at 300mhz using a multiplier of 3 and an fsb set at 100mhz. Use k6clk.exe in autoexec.bat to go higher. It is an absolute MUST to use Jacks dropping resistor mod to deal with Q16's heat problem. I am using a large heatsink/fan combo to cool both the processor, the dropping resistor and Q16. I had to cut a hole in the back and the radio frequency shield for it to fit. I installed the dropping resistor so that it touches the heatsink and is cooled by the airflow generated by the fan and directed by the heatsink. A second 5v fan is blowing air onto Q16 from the ram door. I cut off half of the door to make an intake. This air blowing in also over pressurizes the rest of the inside of the interior, forcing the hot air out of the case. I cut slots in the radio frequency shield above the processor and the hard drive to let the heat escape there. Installing the current controlling resistor at R302 correctly is also an absolute MUST. I also use a power brick that provides twice the current that the original power brick does. I didn't have to drop the Vcore at all.

Please don't think that I am taking any credit for getting to 600mhz first, or just reporting that I did first. Everyone of these techniques were developed by someone else on this board. (Thanks to all of you who not only developed these techniques, but freely shared them with the rest of us.) I just may have been the first to put them all together and have had a little (?) luck on my side. What I would like to see is that someone else with these paramenters repeat what I have done. In the scientific community, if an experiment is not consistently repeatable, then you don't have a valid theory or a "patentable" process.

Also a question for Jack. What model info does Sandra report for the AMD K6's that you sell? I would like the see if the non-mobile K6's would also be able to go to 600mhz.

07-25-2001 10:49:13

New MessageRE:A stable 600 mhz is possible (modified 0 times) Ragnar
Profile
After thought,
When I first bought my I-opener, it was for the sole purpose of being able to provide a cheap way to be able to receive e-mail at home. And if Netpliance had not gone out of business, I never would have bothered to hack this machine at all. :-> Oo la, where life will sometimes lead you!!!
07-25-2001 11:01:53

New MessageRE:A stable 600 mhz is possible (modified 0 times) vailr
Profile
Don't forget the (possibly free) mini DC-to-DC voltage converter, which should produce less heat & wasted power than the resistor & Q16 method.
07-27-2001 06:58:59

New MessageRE:A stable 600 mhz is possible (modified 0 times) GWIZAH
Profile | Email
Ragnar,
I wonder how loud exactly is this monster you got chugging away at 600Mhz? I remember people on this board were worried about adding HD's and Lasagna fans due to the increased noise coming from the box! So you have to fans running in there, and a HD. Is it loud at all? Have you experimented with the DC to DC converters valir mentioned?
07-27-2001 07:36:18

New MessageRE:A stable 600 mhz is possible (modified 0 times) Wild_Pencil
Profile | Email
I'd highly recommend installing the OEM case, if fan noise and space become an issue.

Two of my IOpeners have full-sized heatsinks and 12volt fans installed (admittedly, running at 5 volts to reduce fan speed and noise) It's also nice to be able to add a laptop CDRom without much grief about where to mount it.

About the only bummer about these cases (for me, anyways) is my having to notch out one of the side-panels to add an external floppy-drive header for the Floppy Drive hack.

-WP

07-27-2001 08:10:06

New MessageRE:A stable 600 mhz is possible (modified 0 times) Ragnar
Profile
Gwizah,
Granted, it is not as quite as a stock heat sink no fan 200 mhz machine, but decibel wise it is quieter than my Dad's desk top. Actually the 5V fan is louder than the 12V fan which is being powered only by 9Vs. What little whine there is comes from the 5V fan. It seems to be turning at the higher rpm. If I can find a small 12V fan to replace it, I will. Running two 12V fans under voltage will be quieter than one 5V fan running at 5Vs. I would like to try the DC-DC route, but don't feel too comfortable about misleading someone to get it. I do feel like the larger power supply and the 5V fan that pushes the hot air out of the case ( and it really overpressurises the case, the air coming out of the top is barely warm ) does the "trick", at least in this case.

I definitely am thinking about getting a larger case, I seem to be running out of room for even one more small mod. I am using a usb cdrom, but I have an ide zip drive attached using the header on the mod 3 bracket. Though you can't cold boot to it, it does serve as my "floppy". I also would like to inboard the voltage converter that I "Bodged" from Radio Shark parts, (Boy, am I ever a Junkyard Wars junkie!!). It provides 12V, 5V, and a variable V that I use at 9Vs to drive the 12V fan, thereby saving me having to solder on really small or weak parts.

07-27-2001 08:41:24

New MessageRE:A stable 600 mhz is possible (modified 0 times) Ragnar
Profile
By the way,
Just how much current does a K6 processor need? The litte Radio shack variable voltage regulator is only a couple of dollars. And it can be varied from 1.2V to ~37V. In my bodged voltage converter for my zip drive and fan, the little regulators don't get all that hot, relatively. Just a thought, lets see if someone can shoot this one down. I used Las_Vegas' and TackHead' method to mod my V1. Now I realize I could have as easily used that variable voltage regulator in place of the voltage divider. And could have taken it from the 3.3V or the 5V easily found on the motherboard, thereby not wasting any extra energy provide by the power supply. Oo la, hind sight.
07-27-2001 09:05:45

New MessageRE:A stable 600 mhz is possible (modified 0 times) BadFlash
Profile | Email
Ragnar asked "Also a question for Jack. What model info does Sandra report for the AMD K6's that you sell? I would like the see if the non-mobile K6's would also be able to go to 600mhz."

The K62P IS a mobile processor. I don't have SANDRA and have not run benchmarks. It is just the same as the K62+ but no PowerNow. Electrical specs are the same as for the K62+. I measured around 6.5 amps at 450Mhz. This is +/-15% due to the specs of the cooling resistor. Stability (IMHO) or lack there of, is due to the power supply.

07-27-2001 09:20:19

New MessageRE:A stable 600 mhz is possible (modified 0 times) BadFlash
Profile | Email
Ragnar-
No way can that RatShack power supply put out 6.5-7amps at 2V. There are switching power supplies that can be used and are covered elsewhere here.
07-27-2001 12:46:49

New MessageRE:A stable 600 mhz is possible (modified 0 times) Ragnar
Profile
I reckon that it will be as easy to get those processors up to 600 without any extra effort than the power now model. In fact, it would be less of a headache because it took me a couple of week to find faith in k6clk. The only real world test that I have used to know that I am at a higher speed is with playing solitaire and using a stop watch to time the 52 card pickup display when you win. But, darn it, sometimes it took more than an hour between tests.

Badflash,
I didn't mean to imply that one Radio Shacks little regulators would do it. But ganging up several in series might. I've just got to figure out what that dc-dc converter does and see if I can't replicate it using easily found parts. Sure, it won't be as streamlined as what the others are using, but it will be cheaper to produce, and easier on my conscience.

07-27-2001 13:04:22

New MessageRE:A stable 600 mhz is possible (modified 0 times) mp3boombox
Profile | Email
ragnar what that DC-DC adapter does is farly complex and there realy isnt a cheeper way of replaceing it. Short of getting a sample for FREE heheheh.

as for the 600mhz thing i realy dont think any one is going to see that speed on the iopeners. different parts of the MB cut out at that speed. Why?? i dont know its from a number of differeent things?

Though I dont know and wont state an un true fact my idear is more then likly low power/to much heat or just plain old OVER load on the conponents. to be honist i do believe that the MAIN board controler chip should be cooled some how if possable. any thing above 500mhz and you got a HOT air from your processor, a hot hard drive and a hot little chip being sandwitched between all of that. Chances are thats where a lot of the failing is comming from.

as to getting to any thing above 450/500mhz WITH OUT a TRUE DC-DC your just asking for problems. The way i had had my IO setup befor with my q16 i could clock to 525 mhz but DANG wasn't that thing HOT. Now with my DC-DC and modified cooling system i clock at 550mhz just fine. thats a k6-3 450 over clocked 100MHZ WITH OUT A PROBLEM. And there isnt much heat produced eather. The majority of my HEAT comes from my hard drive and controler chip.

I my self am still waiting for someone to rewrite the drivers to get the 3D hardware enabled on the graphics chip

09-16-2001 16:43:52

New MessageRE:A stable 600 mhz is possible (modified 0 times) Ragnar
Profile
mp3boombox,
The compay that "gives" you that DC-DC step down converter expects that you are potentially able to send them some business as a result of giving you a "free sample".
This free sample in the market place is actually worth ~$60. Because I have no way to send them any further business and feel that it is morally wrong imply that I do I will not ask them for one. I didn't mislead or lie to Netpliance to get my I-openers so don't tell me that I am being hypocritical. I want to be able to live with my conscience according to my beliefs. I hope that you can live with yours'. By the way, it takes talent and intellegience to spell "like a drunken Elizabethian type setter", so why do you go to the lengths to make people believe you are otherwise? And don't give me any genetic nonsense about it because I am lefthanded and dyslexic myself but have made the effort to overcome it. Thank goodness for spell checking programs.
09-16-2001 18:57:58

New MessageRE:A stable 600 mhz is possible (modified 0 times) mp3boombox
Profile | Email
wooo buddie, I don't know what side of the bed you got up on but, read my missspelling a little closer. Cause what I talked about was a little bit differnt then what you stated.
weather I'm a bad speller, dont have time or DONT ''' care is a totaly different ordeal.
as to the DC-DC company thing yes I recognize that there a company, and samples are a way of getting out your product. Did you know I'v purchased 7 of them already! AND installed them on other peoples computers along with my own. So don't try to go GUIDING people around over a fantacy that may be fact or fiction. Granted most people will only sample ONE but some LIKE me have purchsed a few.

Also befor about subing the dc-dc, if you could build a DC-DC modual by hand and build the PCB that small and work that well I'd like to see it done. And as I was stating befor getting a HIGH clock speed is not in a high rate of success with out a proper DC-DC. Chances are more of a long or short term road to disaster. Heat heat and brown outs are the leading cause to IO malfunction.

I dont want to come off harsh or what ever but what I have just said is what I said in the former posting. Next time instead of going off on company rights and proper or inproper intention's (i.e. samples ver. buying). Please stick to the topic thats how things get acheaved. If any one had any thing to say about sampling vers buying they would have said it a LONG LONG LONG LONG TIME ago.

09-17-2001 18:36:17

New MessageRE:A stable 600 mhz is possible (modified 0 times) Ragnar
Profile
mp3boombox,
I don't recall in any of your post that you ever said that you purchased these converters. If I missed it, I apologize. For me, this whole i-opener experience has been simply an intellectual exercise in problem solving. The cheaper it cost to solve a problem the more pleasure I get out of it. With the down turn in the computer industry why waste money on something that can be quite irritating at the same time and for a little extra money buy a 15 inch flat screen for ~$300 and for another $150 put together a 1 gigahertz machine. This thing in reality is nothing but a fun toy. I realize that others may see their machines differently, so to each his/her own. :-P Happy hacking to all!!!
09-17-2001 19:22:21

New MessageRE:A stable 600 mhz is possible (modified 0 times) Ragnar
Profile
P.S
Running at 600mhz stable is not a fantasy. It may not be achievable with all machines, but it is with at least two that I know. I will admit, thought, I did kill a V1 running at 500 mhz. The power mosfets at the lower right hand side were the culprit. Oo la.
09-17-2001 19:33:43

New MessageRE:A stable 600 mhz is possible (modified 0 times) *SF*
Profile
mp3boombox, Would You share Contact Infomation & Price with Us interested in purchasing the PT6702
09-17-2001 20:58:57

New MessageRE:A stable 600 mhz is possible (modified 0 times) mp3boombox
Profile | Email
yes give me a day or so, so i can find the information. This summer i had a few weather experiences. And the warrentie on my hard drive, modem, and a few other parts came in handy after a lighting storm. hence i have not posted any thing in a while HEHE.

ALSO as ragnar stated in oneof his last post's My main use for my hacked iopener is to be my LAPTOP. it takes a few to set it up and get it running but. 128megs ram 550mhz and 12gigs HD space. it does what i need it to do. Also its what the doctor orderd when it comes to FILE transfering. Wish i had the will power to hack my IO appart and make a TRUE IDE connector. file trasfering with a high speed cdrom would be much faster then a USB laplink or network adapter. but nothing is perfict!

09-20-2001 18:49:31

New MessageRE:A stable 600 mhz is possible (modified 0 times) StrangeDog
Profile
The I-mod 3 has a "true" IDE header just waiting to use. No soldering required. I have not actually hooked it up yet, but I have a drive and cable ready to go (I won an IBM small form factor drive in an auction). My quandry is whether to further overtax the PS on the IO or to find a place for a 5v power supply. If you have the IMOD3 you could probally just hack a hole in the case and hook to any optical drive in an external case.
09-20-2001 20:17:13

New MessageRE:A stable 600 mhz is possible (modified 0 times) Ragnar
Profile
mp3boombox,
Have you found the info where to find and how much the pt6702 is, yet? I have just snagged a pristine V5 off eBay for a very reasonable price. I powered it up to test and found out that it hasn't even phoned home. Ran through the set-up with not problems. If it true that you can create a quiet high megahertz machine using the original heatsink and no fan I am open to try it. I would appreciate if anybody would describe their setup(s) using this method. Thanks.
09-22-2001 14:14:53

New MessageRE:A stable 600 mhz is possible (modified 0 times) YouBecha
Profile
Hey you guys with the speed,

can you give my your memory specs (brand or something) I think part of my problem is memory.


http://www.geocities.com/mr_bubba_zanetti/
09-22-2001 16:35:16

New MessageRE:A stable 600 mhz is possible (modified 0 times) Ragnar
Profile
128 mhz memory chip is from http://www.memorytogo.com. It is for the i-book. Don't remember the part number.
09-22-2001 18:09:11

New MessageRE:A stable 600 mhz is possible (modified 0 times) Kudzu
Profile
PT6702
$38.85 USD
Min. 10
Available at http://www.digikey.com
1-800-344-4539

There will be a waiting period, as Digikey currently has none immediately on hand.

Kudzu

09-23-2001 05:32:48

New MessageRE:A stable 600 mhz is possible (modified 0 times) Ragnar
Profile
The Datel dc-dc 2.5V 8amp convertor is available on their website for $52 + $8 s/h. I just might have to try one. I would love to get a 550-600 mhz silent machine. In the dc-dc converter thread it seems that rrunes has done it. Does anyone have a step-by-step procedure how to wire up this puppy to work? The afore mentioned thread has a write-up for the pt6702 but not the Datel. Thanks.
09-26-2001 09:25:48

New MessageRE:A stable 600 mhz is possible (modified 0 times) Ragnar
Profile
I just received the Datel dc-dc converter. I ordered the 5V-2.5V because I am putting it in a V5. I remembered in a thread that Badflash said V4's and V5's processors won't post below ~1.9V or so. But this thing is a small, well-built little device. It looks more and more like it's going to be a ridiculously easy hack to do. There is 6 pins on the device but only three will be used for the simplest hack. I will post a step-by-step write up when I get a chance in the next few days.
10-01-2001 14:27:31

New MessageRE:A stable 600 mhz is possible (modified 0 times) mp3boombox
Profile | Email
Um hehe, YOUR mixing your hacks. the wont boot thing was to do with the q16 chip setup not the DC-DC setup. THE dc-dc has its own voltage settings which you set via 4 wires? 3 wires if orget. YOUR just going to be over powering your processor!!!! by a half volt. MEANING LOTS OF HEAT!!!! I run my io at 550mhz at 1.6v just fine. V4 with the dipswitch thing. OVER powering your processer is going to keep you from hitting the higher cpu mhz. I'd be suprised if you where able to do 550 with the processor UNDER 130F IF at all.
mine runs at 105F over clocked 100mhz moble k6-3 450mhz@550mhz.
10-01-2001 16:48:50

New MessageRE:A stable 600 mhz is possible (modified 0 times) Ragnar
Profile
mp3boombox,
Are you trying to start a flame war? The 2.5V is easily varied.
10-01-2001 19:29:48

New MessageRE:A stable 600 mhz is possible (modified 0 times) Ragnar1
Profile
OK, gang,
I have figured out the datel. My particular model has six pins on it. rnunes model has six tabs in the same location. On one end there are two pins. On the other there are four pins. If you lay the model on it label so you are looking at the bottom orient it so that the side with two pins is at the top and the side with four pins ia at the bottom. Ignore the top two pins. These are for a "logic enabled shutdown. Since I have the 5V to 2.5V model I don't need this. The four pins at the bottom are what matters. To hook it up in the I-opener:

From left to right:

Pin 1 is Voltage in. Rnunes takes his from the Q16 pad. For the 5V model I take power from the left hand pin of Q30. Between Pin1 and Q30 I have a 5V relay coil installed for a couple of bucks from Radio Shack and controlled from the top two pins on CN12. Powers on and off with the I-opener on/off buttom.

Pin2 is the negative connection for the Datel. I attach it directly to the screw pad to the left of C154.

Pin4 is the Datel's power out pin. It will connect to the tops of C154 and C155. Place your favorite votage divider between this pin and C154,C155 to get voltage lower that 2.5.

Pin3 is the output return for the Datel. The third leg of your voltage divider goes to this pin.

Don't forget to take Q16 out of the loop by removing it, lifting, or clipping the two left hand tabs.

I chose the 5V to 2.5V for a couple of reasons. The first, because it was available and the 3.3V to 1.6V was not. And for true overclocking, not just maxing out the top end of a mobile processor, a higher voltage is necessary. And while I can drop the 2.5V lower easily, I ain't figured out a easy way to increase the top voltage that the Datel provides. More info and pics ( if anyone is interested ) later this week.

12-02-2001 00:10:32

New MessageRE:A stable 600 mhz is possible (modified 0 times) Ragnar1
Profile
Ok,
A small change, connecting Pin1 to Q30 causes a slight problem. Connecting it instead to CN12 the datel works well.
Ragnar
12-02-2001 23:00:33

New MessageRE:A stable 600 mhz is possible (modified 0 times) Ragnar1
Profile
1Gang,
It looks like Datel has the 10amp DC-DC 3.3V to 1.8V back in stock. Go to http://www.datel.com.
12-27-2001 12:37:23

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