Page 1 of 2 12 LastLast
Results 1 to 10 of 20

Thread: hybrid dual sproggys....

  1. #1
    Low Bitrate doig's Avatar
    Join Date
    Apr 2003
    Location
    Melbourne, Australia
    Posts
    64

    hybrid dual sproggys....

    Hey all, total newbie here!

    I've been following these posts for a while, and keep noticing that the sproggy in its current form wont work with a via epia m. While its 'with in limits' from what i read the start up draw is too large for a sproggy to handle.

    Is it then possible, and without a major redesign of the sproggy, to split the load between two of them? I know that there are dangers to running two in paralell, so why not split the two major voltages between them? Have one sproggy take care of the 3.3v (set it all to 3.3v, like the old sproggys were all 5v) and say the harddrive, dvd, ect 12+ and the other take care of the 5v and everything else. My electrical knowledge is limited to basically nothing, and so i couldnt even begin to design such a thing...

    On something slightly different, has anyone had any sucess with sproggys and the via epia v? Apart from ddr vs sd ram and the m looking better (imo anyway), the only real difference i can see is about AUD$100 in favour of the v series, which is a hell of an incentive for me to go that way.

    Hope this has been of help...

  2. #2
    Constant Bitrate
    Join Date
    Sep 2002
    Location
    Redmond, WA
    Posts
    171
    The startup current theory was mostly speculation.

    From what I could find, it is signal timing or voltage ramp up timing issue.

    I have a dual 787 and dual 788 sproggy and that does not solve the problem alone. I was able to power the m9000 with my unmodified sproggy 2.7 (dual 787 and single 788) on one occasion. I did it using a BASIC Stamp IC. I used the stamp to control the various signals (Power Good, PS_ON, actual sproggy power on, etc).

    I still haven't figured out the exact timing needed. The time I got it to power up, was actually due to a bug in my stamp code, where it was randomly power cycling the stamp chip restarting the startup sequence.

    My suggestion is to not waste your time with the sproggy if you want to power an M9000 or M10000. Just buy an Opus, or Magic Power (or wait until may and pickup a cheap 70w from mini-box.com).

    The sproggy is still useful for powering other motherboards, harddrives, assessories, etc.

  3. #3
    Low Bitrate doig's Avatar
    Join Date
    Apr 2003
    Location
    Melbourne, Australia
    Posts
    64
    Thanks for that. Well that firmly blows designing a dc-dc to well out of my leauge! Looks like i'm stuck with either an inverter or prebuilt one.

    Does this happen with the epia v series though?

  4. #4
    Newbie
    Join Date
    Nov 2002
    Location
    Berkshire, England
    Posts
    27
    Why not just strap a cheap microcontroller to the sproggy. What do you mean by you can't find the right timing? Surely you get just create the pulses without any highly specific timing, just as long as it's within the stated values in the ATX spec

  5. #5
    Constant Bitrate
    Join Date
    Sep 2002
    Location
    Redmond, WA
    Posts
    171
    In theory it sounds like it should be easy.

    My sproggy powers a P3-933 motherboard perfectly, but try it with the M9000, and it fails to POST.

    I don't have an oscilloscope so I can't easily measure what is happening with the sproggy outputs and timing. It's probably something simple, but I wasn't able to figure it out. I just got tired wasting time on it. Wasting a few hours easily pays for an Opus.

  6. #6
    Low Bitrate doig's Avatar
    Join Date
    Apr 2003
    Location
    Melbourne, Australia
    Posts
    64
    so its a timing issue, not a power one....

    doesnt that mean that the via wont work with any homebuilt device?

  7. #7
    Newbie
    Join Date
    Nov 2002
    Location
    Berkshire, England
    Posts
    27
    Please correct me if I'm wrong but I believe sproggy simulates the power good signal with a simple r-c circuit and the waveform which is produced is violating atx specifcations.
    Power good signal must occur in under 1 second from turn on and over 0.1 seconds. This is simulated correctly by the r-c circuit (I hope). However atx specs also state that the power good signal must rise from 0v to 4.75v in 0.01 seconds which it definately does not.
    I imagine most boards don't care about the risetime but for some reason epia boards obviously do. A simple solution is to stick a not gate in swap around the resistor and capacitor, and adjust the values according to suit the change in voltage threshold on the gates input.

    Sorry if this info has been overdue but I do not know if this is the reason (since I have no sproggy or similar) but it's seems the most logical to me.

  8. #8
    Variable Bitrate stimps's Avatar
    Join Date
    Nov 2001
    Location
    brisbane
    Posts
    326
    well i sure wish VIA would help, I emailed them about it and they replied the usual "we dont care" attitude, ... "if your dc to dc supply is within the atx specs, it should work"
    No attempt to actually help with the problem and tell US the customer, what is actually happening, therefore saving our time and either fixing it, or just using a good off the shelf item. I have a EPIA M6000 and a sproggy and I simply can not get it to POST.
    Ford XR6T blueprint 4.0 L twin cam turbo.
    Xenarc 7" touchscreen M10000 Audigy 2 NX
    ITX 120W PS 80gig HD, 256MB Ram, USB wireless internet/LAN
    Griffin Powermate Panasonic DVD-CDW
    Pioneer 5x7 Four way, 2x12 pioneer subz
    600watt 6chan USA audio AMP

  9. #9
    Variable Bitrate stimps's Avatar
    Join Date
    Nov 2001
    Location
    brisbane
    Posts
    326

    ITX SPROGGY power problems, FAILING TO BOOT

    Oh, and I have just heard a whisper of another thing. It appears that the EPIA M mainboards like 3.0volts, not 3.3v. Can anyone else confirm this?
    Could this be why Mine wont boot?
    Please help me dudes:P I dont want to fork out for an Opus. Especially since it looks like it wont work with that either!!

  10. #10
    Raw Wave Rob Withey's Avatar
    Join Date
    Apr 2000
    Location
    Bedfordshire, UK
    Posts
    2,139
    Originally posted by stimps
    well i sure wish VIA would help, I emailed them about it and they replied the usual "we dont care" attitude, ... "if your dc to dc supply is within the atx specs, it should work"
    No attempt to actually help with the problem and tell US the customer, what is actually happening, therefore saving our time and either fixing it, or just using a good off the shelf item. I have a EPIA M6000 and a sproggy and I simply can not get it to POST.
    No disrespect intended, but it is your power supply that is at fault, not their motherboard. Why should they help debug your power supply for you?


    Rob
    Old Systems retired due to new car
    New system at design/prototype stage on BeagleBoard.

Page 1 of 2 12 LastLast

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
  •