I-Appliance BBS
The Official Source for Internet Appliance Upgrades and Mods		 Amazon Honor System Click Here to Pay Learn More
BBS Main List | Sign In | Sign Up | Search | Help | Linux-Hacker.netReply to Thread | Printer |

Home / I-Opener Areas / I-Opener Technical Stuff
Need opinions on Q16 possible replacement (Linuxguru, Turbo3 or anyone).

New MessageNeed opinions on Q16 possible replacement (Linuxguru, Turbo3 or anyone). (modified 0 times) Technosat
Profile
Hi guys. After succeding in getting 2 i-openers and having sent them all the way here to Italy I have been studing some old posts to proceed with my hack.
I have a V3 that I was thinking of upgrading with a k6-III-333 and am worried about all this horror stories I keep reading on the very high temperature that the Q16 mosfet could reach.
Some old posts by Linuxguru and Turbo3 suggested the use of a higher rated N-mosfet. After some research I have found a possible cheap alternative in the RFP70N06 by Intersil (here is the datasheet).
It is a 70A, 60V, 0.014 Ohm, N-Channel Power MOSFET with a 175' operating temperature in a TO220 package (which should also be easier to cool) and, most important, is actually available around here for around 3$.
Would this work with no problems (provided of course that the correct source/drain/gate connections are made) or does the original mosfet have some additional characteristics I should be aware of?
On the datasheet I see a "gate threshold voltage" which is 1.2-2v for the normal Q16 and 2-4v for the RFP70N06 but unfortunatly I have no idea of what that is or if it is of any importance.

Thanks in advance to anyone willing to show me the light :)

12-17-2000 01:21:15

New MessageRE:Need opinions on Q16 possible replacement (Linuxguru, Turbo3 or anyone). (modified 0 times) Technosat
Profile
Small addition to my previous message. I seem to understand that the gate threshold voltage is the minimum voltage required to "turn on" the mosfet. Is this correct? What is the current value in a modded V3 i-opener?
12-17-2000 01:38:54

New MessageRE:Need opinions on Q16 possible replacement (Linuxguru, Turbo3 or anyone). (modified 0 times) Linuxguru
Profile | Email
The broad conclusion from the earlier discussion was that the exact FET part doesn't really matter; both parts discussed earlier as well as the part mentioned above are adequately rated. However, the nearby electrolytic cap as well as the glue that is used in the PCB laminate will not hold up at sustained exposures to high temperatures (85 c for the cap, over 125 c for the PCB).

To summarize: A heat sink is needed for Q16, regardless of the MOSFET part used, if you want to sustainably overclock the AMD K6-III at or above 266 MHz. The III+ will go to 300 MHz before a heat sink is required.

12-17-2000 16:44:14

New MessageRE:Need opinions on Q16 possible replacement (Linuxguru, Turbo3 or anyone). (modified 0 times) Technosat
Profile
Having the RFP70N06 in a TO220 package will probably have the added benefit of moving the heat away from the motherboard and giving a better choice of heat sinks. I will probably end up having a very large heat sink which will distribute the heat over a wide area and 10mm above the motherboard. I will post as soon as I begin working on that in case anyone is interested in the results.

One last question : does the gate threshold voltage I mentioned above matter in any way?

Thanks for the feedback.

12-17-2000 17:25:21

New MessageRE:Need opinions on Q16 possible replacement (Linuxguru, Turbo3 or anyone). (modified 0 times) Programmer
Profile
Check the specs and watch where and how fast the temp rises. I had considered putting in an 8030L that has some super nice thermal specs.. just never got around to it.. (home time is expensive... :))
12-17-2000 20:47:09

New MessageRE:Need opinions on Q16 possible replacement (Linuxguru, Turbo3 or anyone). (modified 0 times) Linuxguru
Profile | Email
Gate threshold voltage: My guess is that 2-4 volts is too high. In this configuration, Q16 is used as a source follower, so the source is already at a voltage of +2.0 v (or whatever value of Vcore has been chosen). Add 4 to this, and the gate voltage has to be in the neighbourhood of 6.0 v. Depending on where U16 is derives its supply voltage from, this may be too high. There has to be sufficient margin above this to allow op-amp U16 to control the current Ids through Q16 (higher currents demand larger values of Vgs).

To summarize: stick with the existing FET(s), 6030 and 28N03, or find one with similar transfer characteristics.

12-17-2000 21:52:57

New MessageRE:Need opinions on Q16 possible replacement (Linuxguru, Turbo3 or anyone). (modified 0 times) Technosat
Profile
Thanks for the tip.
I have found a FET from Philips Semiconductors (BUK9518-55) which seems to be an almost perfect match for the usual q16.
I was also taking a look at the schematics that Turbo3 posted on his yahoo folder. I see that the 3.5V coming out from Q27 are used to generate Vcore and an additional 2.5V.
Are they used for anything else? I assume the 3.5 volt line is used for the I/O vcc of the CPU and only to drive Q16 and Q17 for the Vcore + 2.5v generation. If this is the case, couldn't it be possible to play with the MAX1631 to lower the output to 3.3v, lowering the voltage drop (and the heat) on Q16?

Finally, any idea on what are the 2.5V used for?

12-17-2000 22:55:54

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