VOGONS


First post, by H.W.Necromancer

User metadata
Rank Member
Rank
Member

Hallo guys,
I have pulled out this bad boy from my last scrap haul. To make it easier there is a link - I have exactly this version of the board.
https://theretroweb.com/motherboards/s/qdi-p5 … -explorer-ii-sp

I have made it to run, fortunately there were just a bit damaged ram slots - no scratches, no missing parts. I have updated the bios to the latest available on the TRW.
It runs with my testing CPU (Intel Pentium 90Mhz)

However if I install ANY dual voltage CPU (Intel MMX, K6, some Cyrix..anything that expect a different IO and Vcore voltages) it does not post. I have realized the CPU is even cold and simply not working.
There is -- -- on my post analyzer. With the single voltage CPU the board works just fine.

I have tested the MMX cpus in another board to confirm they are all good and not burned by the board. These all are posting and seems to be fully working.
(No need to say that in case the mmx would be getting too much volts it gonna be hot as fuc...)
And as you know Intel MMX is kind of durabe and for some time it can work in a sigle voltage system (with cooling). So I do expect there is no vcore at all.
--
This board is fully jumperless. It needs to identify the chip and set the voltage accordingly.
My assumptions are something in the VRM is dead .One local guy told me he had the same board many years ago and as he remembered he had the same problem - alegedly he repleced one of the power transistors - but it was many years ago and he obviosly do not rememeber all the details).

I have found two posts about this board not running MMX but running fine with other CPUs here on Vogons, but these are old and not resulting in any conclusion.

I have checked the 2 darlington transistors in circuit. What is weird is I have 3.3V on the first one (closer to the edge) - which is fine!, bud I have 5V on the middle pin (C) of the second one (it is ok) but 4,7V on the both side pins (E, B I think) which seems to be a nonsence. Just to add those measurements are still the same - does not matter if there is a CPU or not, or which CPU it is.

My first step gonna be taking out that transistor and put it on the LC component tester. But any suggestion will be appreciated as I am not sure how this auto-detect VRM works...
-

Thank you

Reply 1 of 2, by dominusprog

User metadata
Rank Member
Rank
Member

The Pentium 90MHz needs only a 3.3V but MMX variant also need a 2.8V for the core. So the second regulator you mentioned has probably gone bad, but first check all the traces and solder joints.

Duke_2600.png
A-Trend ATC-1020 V1.1 ❇ Cyrix 6x86 150+ @ 120MHz ❇ 32MiB EDO RAM (8MiBx4) ❇ A-Trend S3 Trio64V2 2MiB
Aztech Pro16 II-3D PnP ❇ 8.4GiB Quantum Fireball ❇ Win95 OSR2 Plus!

Reply 2 of 2, by H.W.Necromancer

User metadata
Rank Member
Rank
Member

EDIT: the board has been fixed. I have used a shotgun approach, desoldered all the transistors, mosfets and voltage references in the power stage. All transistors tested as good on my LC tester. But I can ´ test voltage ref. So nothing to lose. I have salavged the same references from a more damaged but very simillar board - just the donnor is not jumperless and missing power transistors... And here we go! Dual voltage CPUs are working again. 99% one of these vRef died. Pretty sneaky fault! The seconf board is maybe fixable as well. Next time...👍
-/
I hope this could help to other guys - most likely it is not an uncommon failure of this board.

dominusprog wrote on 2023-11-15, 12:31:

The Pentium 90MHz needs only a 3.3V but MMX variant also need a 2.8V for the core. So the second regulator you mentioned has probably gone bad, but first check all the traces and solder joints.

Attachments

  • IMG_1171.jpeg
    Filename
    IMG_1171.jpeg
    File size
    170.16 KiB
    Views
    150 views
    File comment
    One of these the vRefs most likely died
    File license
    CC-BY-4.0