PDA

View Full Version : P4S533 vdimm mod.



CrawlingEye
10-01-2002, 10:27 AM
I've heard that connecting the pins for 2.7v along with the pins for 2.9v (all 3) will give 3.1vdimm, but when I was told, I was also told that it's never been tested on the P4S533, but has worked on other Asus boards. Has anyone tried this on their P4S533?

I'd be REALLY happy if I could go to 3-3.2vdimm (anywhere in there would be fine), but don't want to risk going to some insane voltage and killing my mobo/RAM.

Thanks. :)

DDTUNG
10-01-2002, 02:13 PM
Welcome to Xtreme!

By pins I assume you mean jumper pins.

What you describe amounts to connecting two resistances in parallel in the feedback circuit of the Vmem controller chip. To make certain that it will work, you need to measure the resistances between the pins and ground individually. After you do that please post the values here and I'll work out the math for you.

DDTUNG:cool:

CrawlingEye
10-01-2002, 02:47 PM
Not sure, just the jumpers (vmem adjustments are done with jumpers on this board).

How would I measure the resistance of the jumpers themselves?

Thanks for the help so far. :)
I really appreciate it.

DDTUNG
10-01-2002, 03:08 PM
Take a multimeter, set it to read ohms. Connect the black lead to any ground point on the mobo(mounting stud, for example), and touch the red lead to the jumper pins one at a time and note the readings.

DDTUNG

CrawlingEye
10-01-2002, 03:20 PM
Thanks. I'll do it in a bit. I'm on the phone now. I really appreciate your help. If I can get 3v-3.2v I'll love this mobo all the more. :)

CrawlingEye
10-01-2002, 03:33 PM
Hey, I have one final Q. I do this while the mobo's on or off?

DDTUNG
10-01-2002, 03:35 PM
Off.

DDTUNG

CrawlingEye
10-01-2002, 04:07 PM
Got it.
(not sure if this'll help, but I'll add anyway)

There's 4 pins for it.
The way they mobo manual has it is this:
Connect pin 1 and 2 and you get 2.5v (default), connect pin 2 and 3 and you get 2.7v, connect pin 3 and 4 and you get 2.9v.

I couldn't get a read on pin 1, so here's the resistance I got on the other 3:

Pin 2: .9 Ohms
Pin 3: .1 Ohms
Pin 4: .5 Ohms

Thanks again. :)

Edit: Just put the black (negative) probe against the case chasis, that should've grounded it fine. :)

DDTUNG
10-01-2002, 04:17 PM
OK, by no reading on pin 1 I assume you mean reading of 0.

Are you sure the multimeter was set to read ohms and not kohm?

Also, can you measure the resistance of:

1&2 jumpered
2&3 jumpered
3&4 jumpered

DDTUNG:cool:

CrawlingEye
10-01-2002, 04:22 PM
Was set at "RX 1k Ohm"

I think that's the only setting my multimeter has for Ohms. I'll check the booklet in a sec. :)

CrawlingEye
10-01-2002, 04:31 PM
Just read it, then took a good look at my multimeter. You're right, it's measuring kohms. I'd measure in ohms if my multimeter could, but it has no option for it. :(

I hope you can still work with what I have... Thanks. :)

DDTUNG
10-01-2002, 04:48 PM
Originally posted by DDTUNG
OK, by no reading on pin 1 I assume you mean reading of 0.

Are you sure the multimeter was set to read ohms and not kohm?

Also, can you measure the resistance of:

1&2 jumpered
2&3 jumpered
3&4 jumpered

DDTUNG:cool:

I'll need these other values as well.

DDTUNG

CrawlingEye
10-01-2002, 04:51 PM
Whoops, sorry. Missed your post. :)

I'll get the combo readings in a couple minutes. :)
...and yeah, it didn't move at all for pin 1. =\

Edit: how should I measure it with the jumper on? Do you just want me to measure them both at the same time, or actually use the jumper itself?

:)

DDTUNG
10-01-2002, 04:55 PM
Stick the red lead on the jumper itself.

DDTUNG:cool:

CrawlingEye
10-01-2002, 06:22 PM
Done. Kind of odd and not very sensible results though... I even double tested each to check if I had done it improperly...

Here's what I got:

Pins 1 - 2: 1.0kohm
Pins 2 - 3: .1kohm
Pins 3 - 4: .1kohm

Is that normal for it to be the same ohms while supposedly getting an increase in voltage? :confused:

Thanks, again. :)

DDTUNG
10-01-2002, 06:28 PM
It could be your multimeter only give 1 place after the decimal point, so it is rounding up.

Give me some time. I have to try and figure out what the circuit is.

DDTUNG:cool:

CrawlingEye
10-01-2002, 06:43 PM
Take as much time as you need, you're doing me a huge favor here. :)

Thanks a lot, again. :banana: :banana:

DDTUNG
10-01-2002, 06:51 PM
NP.:)

Perhaps you can fold for Team Xtreme while you wait. Team number is 3365.

DDTUNG:cool:

CrawlingEye
10-02-2002, 02:27 AM
?? You mean the 3dmark team? :)

DDTUNG
10-02-2002, 02:31 AM
Sorry I've been out all afternoon.

I meant the Folding@Home team. We are in a fierce battle with Hotmail and can use all the help we can get. Go to the Folding forum and read about it.:)

DDTUNG:cool:

CrawlingEye
10-02-2002, 02:42 AM
Ok. :)

CrawlingEye
10-02-2002, 04:22 AM
Should I get a multimeter that reads just ohms?
(paraphrase: will it help you out?)

DDTUNG
10-02-2002, 04:28 AM
If you plan on doing a lot of Vmodding in the future you should get a digital multimeter that has ranges of values for voltage, resistance, and current.

For this particular case you have given me the values I need, more or less. It's just more difficult to figure it out without the board in front of me. I'll work on it.

DDTUNG:cool:

CrawlingEye
10-02-2002, 04:40 AM
Thanks. I'll definitely consider picking a new one up.

I just got this for $10 at radioshack. I was trying to stay cheap on with it when I got it. :)

If I do continue to vmod in the future, I'll be certain to get one. :)

jgh
10-19-2002, 08:38 PM
Crawling eye.
Was there ever an outcome to this? I was pretty interested in the answer. Thanks JGH

CrawlingEye
10-25-2002, 12:56 AM
I'll be finding out in 5 minutes, as I'm finally implementing it right now. I've just found out where to detect the vdimm voltage on my board (which mosfet). It's the mosfet directly to the right of your 3rd slot RAM. It's around a bunch of caps, the bottom connection is the one you want.

I'm going to double check to make sure this is the trace by moving my jumper and seeing if the voltage reading decreases accordingly.

I'll reply with my results. :)

CrawlingEye
10-25-2002, 01:01 AM
Confirmed, this is the mosfet and connector to use to check vdimm voltage. The connector on the same mosfet (on the opposite side) will show you your 3.3v rail, all though this isn't really useful... I'm going to shut down and vdimm mod now. I'm only going to go for a minor vdimm mod, as I don't have ramsinks on my RAM ATM. :)

CrawlingEye
10-25-2002, 01:27 AM
So far I've lowered it to 17K Ohms with no notable change in voltage. It's staying at 2.90v. I'm going to try something like 16K Ohm in a few minutes.

CrawlingEye
10-25-2002, 01:42 AM
15K Ohm and I'm still getting 2.89v vdimm. I don't quite get what's up with it?

I'm using my case side as a temporary ground (which should be fine) and I'm using the far leg of the VR for the ground. With the center to the jumper.

CrawlingEye
10-25-2002, 01:55 AM
14.1K Ohm attempt, got 2.9v exactly, but this could likely be accredited to the difference in timing when I checked the vdimm.

I'm going to try 13K Ohm, if I don't see a difference in voltage at 10K Ohm, I'm going to bag this attempt in. Thanks for the help, DDTUNG.

CrawlingEye
10-25-2002, 02:26 AM
Still a no-go. To test that my connections were on properly to the VR, I put them on the test probes of my multimeter and read the resistence through them. It was the exact same as I was getting before.

If you have any more suggestions, DDTUNG, I'd be more than glad to hear them. :)

Edit: I hope you haven't grown tired of me or anything, DDTUNG. =\

CrawlingEye
10-25-2002, 07:51 PM
If it helps at all, the mosfet I'm using to read the vdimm is the "APL158AE3L1."

CrawlingEye
10-25-2002, 08:11 PM
I can't seem to find any DDR regulators that match up to anything on the board, as this seems to be the technique used (someone correct me if I'm wrong). It has a lot of chips on it made by ICS, so I went to www.icst.com and tried doing a search for a couple, with no results. So I bagged that in and went on their site and looked at their DDR voltage regulator, which seems to be "ICS9S857" however, I can't find that on my board.

ICSUT180032014193705AF is what I have found, which is right to the right of my DDR slots but it doesn't even show up on their site. =\

CrawlingEye
10-25-2002, 08:17 PM
I did find http://www.icst.com/products/summary/sis-frame.html I'm not sure if that'll help.

MrIcee
10-26-2002, 04:42 AM
CrawlingEye:)


What you have just found is one of the clock generators, this definitely is not the right chip. The voltage regulator is usually placed at the bottom of the memory slots or to the right of them.It also could be that the IC you have found does not yet have a tech sheet associated with it yet on the manufactiurers site. Also one more observation, generally on a VR, the center pin is the ground and 1 of the 2 outer pins are connected to the jumper.

Randi:D

CrawlingEye
10-26-2002, 10:22 AM
K, thanks. I must've gotten it backward then. Thanks for the clarification, MrIcee. I'll retry this mod soon. :)

CrawlingEye
10-26-2002, 11:32 AM
I got it in taking Jarmo's advice. Connecting jumper pins 2-3-4. It's giving me 3.11v :)

Thanks to everyone for helping.

CrawlingEye
10-27-2002, 03:05 AM
Here's some of my results so far. I think I need to get a new NB cooler. Probably order the Titan tomorrow, as I've heard this is the smallest and one of the best.

http://files.joako.com/SiSMBB465.jpg