MMM
Page 1 of 3 123 LastLast
Results 1 to 25 of 54

Thread: Epox 8K9A2+ Vcore mod.

  1. #1
    Xtreme Christmas Lighter
    Join Date
    Jun 2002
    Location
    Upstate NY, USA
    Posts
    3,790

    Epox 8K9A2+ Vcore mod.

    Since this board only allows setting the Vcore to 2.0v(which only applies 1.95v on my board), I thought maybe people would be interested in coming up with this mod.

    The chip used on this board is the IRU3055(data sheet). Now I think the pin we need to work with is pin 3(Fb). This looks to be the same sort of setup as the 8KHA+ Vmod located over at OCInside.com.

    I think I'm on the right path... If I'm correct above, how do I know what value resistor to try? I see an 220k was used on the 8KHA+ but I don't have any idea if that will give me +.2v or +2v?

    Like I said, if someone has any ideas, please do share.
    Last edited by Jeff; 12-29-2002 at 04:00 PM.





  2. #2
    Hardass
    Guest
    Sounds like a job for DDTUNG!

  3. #3
    Xtreme Legend
    Join Date
    Jun 2002
    Location
    Hong Kong
    Posts
    3,289

    Smile

    Doesn't look good. A quick read of the data sheets indicates that the FB pin is connected directly to the output, which necessitates cuttingg or unsoldering the pin or cutting the trace, then soldering a VR.

    DDTUNG

  4. #4
    I am Xtreme
    Join Date
    Nov 2002
    Location
    South FL, USA
    Posts
    4,892
    Ok, I looked at it and if I had this board, I would pay attention to page 4. FB looks like a overvoltage protection, a feedback to the amplifier. I would put a 100K pot there to keep that +15% (top of page 10) protection in check.

    Now, slide down to explanations of D0, D1, D2, D3 & D4 all influence the output voltage (page 4). It says you can use a 10K resistor externally to raise the voltage from 3.3 to 5V. They might be talking about bypassing the internal source, not sure here.

    So, I would try the 1st paragraph to keep overvoltage protection in check. My 2nd para is tricky so I would start with a 47K pot on D0 pin/leg and decrease resistance to see if that influences the Vcore at all. You will probably need to be around 10K would be my guess but all this is guessing.

    If anything does go wrong, then most likely the video will blank out. In that case, just power down for 10 secs and then power back up.

    I would attempt this. I don't think you will hurt anything but it is guess work! I guessed at finding my Vdimm mod for my MSI Duallie board. I guessed correctly without any PDF file and it worked.
    Last edited by Ace-a-Rue; 11-23-2002 at 11:02 AM.

  5. #5
    Xtreme Member
    Join Date
    Oct 2002
    Location
    The Netherlands
    Posts
    435
    Good topic. I have an XP1800+ Tbred on 1,97V also with an Epox 8K9A2 (no raid). I need more voltage. Running @ 2056Mhz @ 30C It's begging me for more Can do 2070Mhz stable with 1,97 (thats 2 in the bios). 2100Mhz is what i need, more is also good ofcourse

    Edit: Ah..still gotta change my sig

    Changed
    Last edited by Fewture; 11-23-2002 at 10:51 AM.

  6. #6
    Xtreme Member
    Join Date
    Oct 2002
    Location
    The Netherlands
    Posts
    435
    I put 60K between pin 3 and ground. Nothing happens. I used 10K vr and 50K resistor. So 50K does nothing.
    Wil try lower and see what hapens..

    If anybody can say something usefull about this vmod please say it. There is nothing on the net about this and I'm an electronics n00b, but that won't stop me

    Edit:
    Doesn't seem to work. Had big trouble booting.. It's running now, bios seems to be screwed.. Runs stable though

    Edit:
    Ok, everything is ok now. Runs 100%, but no vmod! DDTUNG?? Can you look at this once more? I can measure some things if that helps.. I really want this to work, but I don't know what to do..
    Last edited by Fewture; 11-25-2002 at 12:45 PM.

  7. #7
    I am Xtreme
    Join Date
    Nov 2002
    Location
    South FL, USA
    Posts
    4,892
    After looking at this again, I have to revise my thinking. Like DDTung said leg #3 (FB) is the output but it also monitors it self to not let an over voltage occur above +15%, which is described on page 10 of the data sheet.

    I would try to influence leg #2 to see if that can do anything for you. I would try a very high pot like 200+ K.

    If I had the board, my discovery would be hit or miss. You have to be a willing explorer, ready for the unknown and willing to do anything.

    Good luck!

  8. #8
    Xtreme Member
    Join Date
    Oct 2002
    Location
    The Netherlands
    Posts
    435
    Originally posted by Ace-a-Rue
    After looking at this again, I have to revise my thinking. Like DDTung said leg #3 (FB) is the output but it also monitors it self to not let an over voltage occur above +15%, which is described on page 10 of the data sheet.

    I would try to influence leg #2 to see if that can do anything for you. I would try a very high pot like 200+ K.

    If I had the board, my discovery would be hit or miss. You have to be a willing explorer, ready for the unknown and willing to do anything.

    Good luck!
    Well, I think I have showed I'm willing! Almost ruined the board. So what do I do? Get a pot of 500K? Solder it to pin 2? And turn it down, right?
    Do nothing to pin3? Of should I solder somthing on there too? You are the brains here, I'm just crazy enough to do it. I won't hold you responsible, but I do want you to think..LOL

    Tnx a lot m8!

    Edit:
    So pin 3 lets me go up to 15% over 2,00V? 2,00 is max on the board but i actually gives 1,95V. That would give me 2,24 and thats good enough. 2,2 would do in fact.
    Am I correct?
    Last edited by Fewture; 11-25-2002 at 03:32 PM.

  9. #9
    Xtreme Member
    Join Date
    Oct 2002
    Location
    The Netherlands
    Posts
    435
    Well, someone is helping me. I think we're getting somewhere. I could still use some input so we can crack this mod. It looks like I'm gonna try pin2.. I'll let ya know what happens.

    (got a another pc here is case of death :P )

  10. #10
    I am Xtreme
    Join Date
    Nov 2002
    Location
    South FL, USA
    Posts
    4,892
    Have you ever used SMD grabbers?
    BIOSTAR TPOWER I45 UNOFFICIAL THREAD

    BIOSTAR TPOWER BOLT MOD FOR HEATPIPE AND HEATSINK

    BIOSTAR TPOWER I45 BIOS FLASHING PROCEDURE

    ABIT IP35 PRO HEATPIPE MOD

    ABIT IP35 PRO BIOS FLASHING PROCEDURE

    IP35 Pro: 9650@4000Mhz, par overclocker; Freezone Elite; 4Gb GSkill DDR-800@DDR-1068 (2 x 2gb); XFX 8800 GTS; Areca 8X PCIe in Raid 0 working at 4x speed; 4-250 Gb (single platter) 7200.10 drives; Giga 3DAurora case with side window.

  11. #11
    Xtreme Member
    Join Date
    Oct 2002
    Location
    The Netherlands
    Posts
    435
    Originally posted by Ace-a-Rue
    Have you ever used SMD grabbers?
    No, I know what they are but I can't find them here. But I know a little shop that might have them.
    But sms grabbers won't solve my problem. I can solder on the pins. They're small, but it's ok..

    Edit:

    I went as low as 15K on Pin 3. This almost fried the bios I think. I started hanging in the bios en a bit later i could not get into the bios. It locked during IDE drive detection.
    Now another guy is saying that it should be pin 3. I'm sure you'll understand I wont try that again!

    So this pin 2.. Any good at all? Should I go for it? Or do you guys have a better plan?

    DDTUNG:
    The cutting of pin 3. Would that work? Or are you not sure about that?

    When going over 1,85V the voltage increases by 0,05 sinstead of 0,025 with every step. The max voltage that can be programmed is 1,85. So where do they get the extra 0,15? Or am I missing something here?

    Edit:
    Ok, I got SMD grabbers!
    Last edited by Fewture; 11-26-2002 at 03:20 AM.

  12. #12
    I am Xtreme
    Join Date
    Nov 2002
    Location
    South FL, USA
    Posts
    4,892
    Put the SMD grabber on pin #2 and a 200+K pot and see what happens. If no changes, then you can start lowering the resistance very, very slowly. Move it a small amount and let the puter catch up with that change.

    The reason I suggest SMD grabber is to avoid the soldering screw-ups. Most of us are not real proficient at soldering very small jobs.
    BIOSTAR TPOWER I45 UNOFFICIAL THREAD

    BIOSTAR TPOWER BOLT MOD FOR HEATPIPE AND HEATSINK

    BIOSTAR TPOWER I45 BIOS FLASHING PROCEDURE

    ABIT IP35 PRO HEATPIPE MOD

    ABIT IP35 PRO BIOS FLASHING PROCEDURE

    IP35 Pro: 9650@4000Mhz, par overclocker; Freezone Elite; 4Gb GSkill DDR-800@DDR-1068 (2 x 2gb); XFX 8800 GTS; Areca 8X PCIe in Raid 0 working at 4x speed; 4-250 Gb (single platter) 7200.10 drives; Giga 3DAurora case with side window.

  13. #13
    Xtreme Member
    Join Date
    Oct 2002
    Location
    The Netherlands
    Posts
    435
    Originally posted by Ace-a-Rue
    Put the SMD grabber on pin #2 and a 200+K pot and see what happens. If no changes, then you can start lowering the resistance very, very slowly. Move it a small amount and let the puter catch up with that change.

    The reason I suggest SMD grabber is to avoid the soldering screw-ups. Most of us are not real proficient at soldering very small jobs.
    Ok, I will do it.. Small soldering is no problem btw. Don't know much about electronics, but soldering is a piece of cake Don't ask me why..

  14. #14
    Xtreme Member
    Join Date
    Oct 2002
    Location
    The Netherlands
    Posts
    435
    250K pot makes the vcore go down. Only a little bit of turning had effect. Went from 1,52V whitout pot to 1,48 with pot @ 250K. I turned it a bit more (tiny bit), nothing...bit more...1,47..bit more.. 1,46...1,45...1,reboot Guesnig low voltage reboot..
    Pot wasnt even turned half way. Wil measure the resistance later.

    All other voltages stay normal btw. Could not use the smd grabbers. Pins are too close together.

    What to do???

  15. #15
    I am Xtreme
    Join Date
    Nov 2002
    Location
    South FL, USA
    Posts
    4,892
    What to do???
    Question of the year! I wish I had an answer. Some boards are easier than others. It seems they are making it more difficult to allow us to o/c via mods. if that is the case, I am going to be very selective on board purchases. Actually, I stopped my impulse buying after the disasterous Asus P4T533 32 bit Rdram m/b.
    BIOSTAR TPOWER I45 UNOFFICIAL THREAD

    BIOSTAR TPOWER BOLT MOD FOR HEATPIPE AND HEATSINK

    BIOSTAR TPOWER I45 BIOS FLASHING PROCEDURE

    ABIT IP35 PRO HEATPIPE MOD

    ABIT IP35 PRO BIOS FLASHING PROCEDURE

    IP35 Pro: 9650@4000Mhz, par overclocker; Freezone Elite; 4Gb GSkill DDR-800@DDR-1068 (2 x 2gb); XFX 8800 GTS; Areca 8X PCIe in Raid 0 working at 4x speed; 4-250 Gb (single platter) 7200.10 drives; Giga 3DAurora case with side window.

  16. #16
    Xtreme Member
    Join Date
    Oct 2002
    Location
    The Netherlands
    Posts
    435
    Originally posted by Ace-a-Rue
    Question of the year! I wish I had an answer. Some boards are easier than others. It seems they are making it more difficult to allow us to o/c via mods. if that is the case, I am going to be very selective on board purchases. Actually, I stopped my impulse buying after the disasterous Asus P4T533 32 bit Rdram m/b.
    Well this mod is going to work!!

    I measured between FB and Ground (pin 3 and 17) and it reads the vcore. When i go above 1,85 (DAC Max?) it still reads 1,85. So the mobo is vmodding already.
    Any use this info??

    Comon guys..we can do this... DDTUNG? Some words you would like to add?

    I have 2 options.. a vmod or a §§§§ed board :p

  17. #17
    Xtreme Member
    Join Date
    Nov 2002
    Location
    Bangkok, Thailand
    Posts
    154
    I hope this might help. Please see.
    Theoretically, most voltage regulators have overvolt/undervolt protection circuit. They shutdown MOSFETs if the volt is under or over by a certain percentages. Furthermore, the max programmable volt is 1.85V.
    Therefore, if you reach the max overvolt limit (as some new m/bs have already allowed for i.e. 2.0V), you need one more resistor to connect Vsense to GND.
    Got to do it together FB-GND and Vsense-GND.
    Last edited by KS1; 11-27-2002 at 03:28 AM.
    1. A7V8X, 1800+ AIUGA0241 @ 1.66Ghz @ 1.5V air-cooled, 256MB Samsung PC2700
    2. 8RDA+, 1700+ JIUHB0308 @ 2.7GHz @ 2.1V, Vapochill PE R-22), 2x256MB OCZ PC3500

  18. #18
    Xtreme Member
    Join Date
    Oct 2002
    Location
    The Netherlands
    Posts
    435
    Look no further!! Got it!

    If it works on higher voltage is stil something to test. Had the vcore on 1,65 and turned the pot. Went up to 1,85V.

    Will see what happens on higher voltage. I'm very happy! Pin 3 and 17 are the pins. I used 250K and it starts working about halfway.

    Do NOT use a fan connector as ground. This was my first try and the board didn't like it. Lockups and stuff. Another guy said this was because the ground wasn't clean, Pin 17 is.

    Sorry about the F word mods

    So Jeff...happy??

  19. #19
    Xtreme Christmas Lighter
    Join Date
    Jun 2002
    Location
    Upstate NY, USA
    Posts
    3,790
    Yep. I just got home so I've missed your adventure. I'll probably get a 250k pot. this afternoon and try it this weekend.

    GREAT WORK!!!





  20. #20
    Xtreme Member
    Join Date
    Oct 2002
    Location
    The Netherlands
    Posts
    435
    Originally posted by Jeff
    Yep. I just got home so I've missed your adventure. I'll probably get a 250k pot. this afternoon and try it this weekend.

    GREAT WORK!!!
    Tnx m8

    Lemme know what you gain ok? I tried contacting you by PM btw..wanted to tell you my progress since you where looking for the mod. I'm guessing you kan do it with a 100K pot.

    I went from 2050 rock stable to 2130 stable. But hight voltage is needed so I'll take it easy. Seen 2 Tbred's die on 2,2V @ 40C. Both on water. 1 died completely and the other one could not clock high anymore.

    Be carefull turning the 250K pot!!! You have been warned! It flies up! Mine went from 1,5 to 1,85 with a slight twist. So a smaller pot or a resitor whit a pot would be better and more safe. I took 250K to be safe for trying..

    Good luck!

  21. #21
    I am Xtreme
    Join Date
    Nov 2002
    Location
    South FL, USA
    Posts
    4,892
    Originally posted by Fewture
    Look no further!! Got it!

    If it works on higher voltage is stil something to test. Had the vcore on 1,65 and turned the pot. Went up to 1,85V.

    Will see what happens on higher voltage. I'm very happy! Pin 3 and 17 are the pins. I used 250K and it starts working about halfway.

    Do NOT use a fan connector as ground. This was my first try and the board didn't like it. Lockups and stuff. Another guy said this was because the ground wasn't clean, Pin 17 is.

    Sorry about the F word mods

    So Jeff...happy??
    So let me get it straight in my mind.

    1st; can you or can you not use a SMD grabber with this set-up??

    2nd; you ran a 200+K (250K) pot or resistor between leg/pin3 and ground, and then between leg/pin17 and ground. Is that correct?

    3rd; how do you adjust the pots, simultaneously?? Or, do you do it from the bios, which I would think would be the best way once you have done the pre-setting on the pots?

  22. #22
    Xtreme Member
    Join Date
    Oct 2002
    Location
    The Netherlands
    Posts
    435
    Originally posted by Ace-a-Rue
    So let me get it straight in my mind.

    1st; can you or can you not use a SMD grabber with this set-up??
    Impossible. Pins are way too close. I soldered 2 pins together as well. Argh!! I wasa bit shakey this morning. Do NOT try this if you are not experienced..you will ruin the board just like I almost did.
    The SMD grabbers don't fit between the legs..


    2nd; you ran a 200+K (250K) pot or resistor between leg/pin3 and ground, and then between leg/pin17 and ground. Is that correct?
    No no no!!!! LOL! Hold it! I ran 250K between Pin 3 and 17. Thats it! Very simple! Just 1 pot.


    3rd; how do you adjust the pots, simultaneously?? Or, do you do it from the bios, which I would think would be the best way once you have done the pre-setting on the pots?
    See above. And yes I set it once and then leave it. Way too riscy to start fiddling with it. It's extremely sensitive.

    Tomorrow (it's 20:57 here..) I will take the pot off and measure it for your pleasure. Then you buy the correct value and it won't be so sensitive.
    I have a 100K and I'm hoping it wil go on. Hoping the resistance used is less than 100K that is. I think it is actually..

    When I putt a pot between pin 3 and ground from a fan connector the board went nuts for some reason. I think I pushed it into the OVP. It would boot perfectly but after a while it went completely crazy. Started hanging in the bios and later the bios was just blue stripes! EEK! I thought it had died! After half an hour without the mod I pushed the button and it booted..luckily..

  23. #23
    Xtreme Member
    Join Date
    Sep 2002
    Location
    NY
    Posts
    267

    Re: Epox 8K9A2+ Vcore mod.

    Originally posted by Jeff
    Since this board only allows setting the Vcore to 2.0v(which only applies 1.95v on my board), I thought maybe people would be interested in coming up with this mod.

    Shoot..... I've only Got 1.93v !!! Need I little more.... Is this working reliably?

    Damm, I only have a 50k pot, and it's a holiday... You know I'm almost @ 2.5GHz. On AIR...

    Royd
    Last edited by RoydRage; 11-27-2002 at 08:11 PM.

  24. #24
    Xtreme Member
    Join Date
    Oct 2002
    Location
    The Netherlands
    Posts
    435

    Re: Re: Epox 8K9A2+ Vcore mod.

    Originally posted by RoydRage
    Shoot..... I've only Got 1.93v !!! Need I little more.... Is this working reliably?

    Damm, I only have a 50k pot, and it's a holiday... You know I'm almost @ 2.5GHz. On AIR...

    Royd
    Vcore is just as stable with pot as without pot. Sometimes you see swings in voltage on certain mobo's. I was told this is caused by a bad ground. On my KR7A I had huge voltage swings. I used a fan connector as ground...bad idea. When you use a ground of the chip it should be as stable as before.

    On my KR7A it was real easy to solder. But this Epox board is quite hard. The pins are really small and very close together. It's doable though.

  25. #25
    Xtreme Member
    Join Date
    Sep 2002
    Location
    NY
    Posts
    267

    Re: Re: Re: Epox 8K9A2+ Vcore mod.

    Originally posted by Fewture
    Vcore is just as stable with pot as without pot. Sometimes you see swings in voltage on certain mobo's. I was told this is caused by a bad ground. On my KR7A I had huge voltage swings. I used a fan connector as ground...bad idea. When you use a ground of the chip it should be as stable as before.

    On my KR7A it was real easy to solder. But this Epox board is quite hard. The pins are really small and very close together. It's doable though.
    Hi,

    What I really meant was the board stable with the mod....

    Did you use a Magnifier light when you did the soldering... There are techniques that were developed dating way back to the Kt-7... For this soldering... I lot of us have gotten quite good at it.....

    I have a 50k pot on had.... Did you get a chance to measure?

    Thanks, Royd

Page 1 of 3 123 LastLast

Bookmarks

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •