Might your problem not just be that the drive (regardless of fact it's an SSD) is just too big for the system?

Looking on the K6Plus site this board has a FIC beta BIOS, so not one patched by Jan. That means it's not automatically been patched for large IDE drives. Do you have positive confirmation that a 120GB drive should work with this BIOS?

For what it's worth, I've used Team Group and Silicon Power budget 120 gigabyte SSDs with a Soyo 5EMA+, an Epox MVP3G-M, and an Asus TXP4 all with K6 plus CPUs, and none had any issues. I've used (or tried using) SiL3114 PCI SATA cards with all three, but they would not work with the MVP3 motherboards (but I'm using one in the Asus TX board with no issues). I've had very good luck using Promise Ultra 100 or Ultra 133 cards with MVP3 motherboards, but even then they can be tricky. I've had them where everything seemed to be working fine, only to have them hang while booting up and the only solution was to switch to a different PCI slot. I've come to expect to have to do that after installing Windows 98, then installing the Via 4 in 1 drivers.

Be sure to disable any unused ports in the CMOS settings to free up IRQs. The consensus is that Via 4 in 1 4.17 drivers are the best for MVP3 boards with Win98.

Return 0; wrote on 2023-09-25, 21:37:

AMD K6-3+ (modified 2+) 100x6.0

Very similar to a broken CPU cache. Or general system instability.
When the speed of i/o is low, data arrives slowly and the CPU rests.

With UDMA, the bottleneck is eliminated, and the system crashes under load.
First, I would check the CPU cooling and increase its voltage by 5%

Or try another non-modified CPU to eliminate any issues with that.

I have a Socket 7 system that only supports up to UDMA-33 and in combination with a HDD that is UDMA-100 capable it always crashed at Windows 98 installation.
The reason was the board expected the HDD to run at UDMA-33 while the HDD expected the board to support much higher rates. So both did talk differently to each other eventually leading to crashs.
Solution was to limit the HDD to UDMA-33. No problems ever since.
Can't tell if that could be part of your problems since you also tried with different adapters but maybe it's a hint you can check.

I can recommend a promise tx4 300 pci sata controller. It has windows 98 drivers. Works like a charme.

Very similar to a broken CPU cache. Or general system instability.

The problem does not occur with the HDD, everything works as it should.

First, I would check the CPU cooling and increase its voltage by 5%

The processor has an Athlon XP cooler, the voltage is raised to 2.2V

Or try another non-modified CPU to eliminate any issues with that.

If the problem is the internal cache, then disabling it in the BIOS should solve the problem? I'll try.

I can recommend a promise tx4 300 pci sata controller. It has windows 98 drivers. Works like a charme.

I found this ad, will it work the same way?

Solution was to limit the HDD to UDMA-33

How to do this with a SATA SSD?

Looking on the K6Plus site this board has a FIC beta BIOS, so not one patched by Jan. That means it's not automatically been patched for large IDE drives. Do you have positive confirmation that a 120GB drive should work with this BIOS?

The BIOS detects it correctly as 128GB (120GiB) but it may still be as you write, the detection itself does not mean anything in the BETA BIOS.

ChrisK wrote on 2023-09-26, 06:00:

Or try another non-modified CPU to eliminate any issues with that. […]
Show full quote

Or try another non-modified CPU to eliminate any issues with that.

I have a Socket 7 system that only supports up to UDMA-33 and in combination with a HDD that is UDMA-100 capable it always crashed at Windows 98 installation.
The reason was the board expected the HDD to run at UDMA-33 while the HDD expected the board to support much higher rates. So both did talk differently to each other eventually leading to crashs.
Solution was to limit the HDD to UDMA-33. No problems ever since.
Can't tell if that could be part of your problems since you also tried with different adapters but maybe it's a hint you can check.

I can confirm this (having FIC VA503+ and several other VIA chipset motherboards). Most of your problems boil down to "something" setting a DMA mode, whilst motherboard doesn't expect anything faster than UDMA-33. Even Via 4 in 1 does this, "boosting" your HDD performance by turning on the DMA, dooming your Windows installation.

The solution is either limit UDMA on the side of hard drive, or fix the BIOS. There are some utilities for the first solution, but their functionality is limited only to certain contemporary hard drives.
For the second solution, there is also an automatic BIOS patcher fixing this issue ("UDMA66/100/133 on UDMA33_only_MB patch"), you can find it at http://rom.by/ I can confirm its functionality. I used BIOS Patcher ver. 4.51.

"UDMA66/100/133 on UDMA33_only_MB patch"

From report: UDMA66/100/133 on UDMA33_only_MB patch: not found.
BIOS JE4333
Do I need anything extra for it to find?

elcrys wrote on 2023-09-27, 20:07:

I can confirm this (having FIC VA503+ and several other VIA chipset motherboards). Most of your problems boil down to "something" setting a DMA mode, whilst motherboard doesn't expect anything faster than UDMA-33. Even Via 4 in 1 does this, "boosting" your HDD performance by turning on the DMA, dooming your Windows installation.

Reminds me of an nForce2 based laptop. Could not negotiate with a SATA SSD reliably. I asked at the SSD manufacturer's helpdesk. They told me the chipset was on their list of known incompatible chipsets. The chipset was designed in such a way it just gave random failures at the speeds involved.

Return 0; wrote on 2023-09-27, 19:13:

Solution was to limit the HDD to UDMA-33

How to do this with a SATA SSD?

That may not be possible at all. Just wanted to point into the direction that it may be DMA related.
And as you wrote that disabling UDMA completely in BIOS fixes your problems it really looks to me to be like that.
If a PCI SATA controller can't fix this you should try to get the BIOS patcher working as elcrys suggested (didn't know it to have such feature though).

While trying to remember how I found out that UDMA was the problem, I think in the second boot screen the harddrive was shown as UDMA-66 or even more.
Knowing the chipset only supported UDMA-33 I must have combined what was going wrong and playing around with the settings in the BIOS (aka disabling UDMA temporarily) the system eventually worked.
Second step was to use the right utility to limit the harddrive's UDMA mode and enabling UDMA again in the BIOS..
This was one of my very first lessons learned while starting with that retro thing a few years back.

Return 0; wrote on 2023-09-27, 21:19:

From report: UDMA66/100/133 on UDMA33_only_MB patch: not found. BIOS JE4333 Do I need anything extra for it to find? […]
Show full quote

"UDMA66/100/133 on UDMA33_only_MB patch"

From report: UDMA66/100/133 on UDMA33_only_MB patch: not found.
BIOS JE4333
Do I need anything extra for it to find?

I still have all the files so I ran it on JE4333 with the same result as you. Looks like you are out of luck for an easy fix. For PC Partner VIB804DS motherboard (VIA VPX), I got this (and it fixed the issue):

1C:\award>bp-4_51.exe vpxas619.bin /s
2BIOS Patcher ver. 4.51.
3Attention! Advanced qualification is required!
4
5Found 1Mbit BIOS!
6
71. New CPU Support           : -> fixed.
82. P3-detect error           : is not needed to be fixed.
93. New Koeffs Support        : -> fixed.
104. 32Gb-problem              : -> fixed.
115. Some HDD detect-problem   : -> fixed.
126. "MB"/"GB" string search   : -> fixed.
137. 65Gb-problem (1-st step)  : -> fixed.
148. 65Gb-problem (2-nd step)  : -> fixed.
159. Error display Freq>999MHz : is not needed to be fixed.
1610.Error display Koefs>9.5x  : not found.
1711.New Stepping Support      : is not needed to be fixed.
1812.Tualatin L2-init error    : not found.
1913.New Freq in Setup open    : not found.
20
2114.Set "Y" as default on exit: -> fixed.
22
23
24Write Allocate addinng: -> fixed.
25UDMA66/100/133 on UDMA33_only_MB patch: -> fixed.
26
27Tweak options addinng:
28
29CBROM V2.07 (C)Award Software 2000 All Rights Reserved.
30Adding modul.tmp  32.6%
31CBROM V2.07 (C)Award Software 2000 All Rights Reserved.
32Adding tweak.tmp  69.5%
33CBROM V2.07 (C)Award Software 2000 All Rights Reserved.
34Adding start.tmp  87.5%
35
36if you can`t see all messages - choose 80x50 mode or run with ">report.txt".
37(c)2002-2003 apple_rom, www.ROM.by
38C:\award>

Most HDD manufacturers (IBM, Maxtor, Quantum, Seagate, WD) were aware of this issue and they made utilities to limit UDMA modes on their HDDs.

Maxtor says this about it:

Overview: Some older Ultra DMA 33 motherboards with older BIOS' (revision date 10/28/98 and older) have exhibited compatibility […]
Show full quote

Overview: Some older Ultra DMA 33 motherboards with older BIOS' (revision date
10/28/98 and older) have exhibited compatibility issues with Ultra DMA/66/100
drives. The symptom manifests itself as a system hangs at boot, and performance
problems. On some motherboards the system does not properly check the UDMA
setting returned by the drive. The end result is that the motherboard attempts
to set itself up in an unsupported mode. These issues arise because of a
bug in the system's BIOS, NOT a problem with the Maxtor hard drive. Some
System's will also have performance problems due to the controller chipset.

To eliminate this problem, Maxtor has developed the UDMAUPDT.EXE The
UDMAUPDT.EXE changes the Ultra DMA Mode for the following hard drive
Families (and newer) ONLY:

· DiamondMax 6800
· DiamondMax Plus 6800
· DiamondMax 36
· DiamondMax VL 17
· DiamondMax 40
· DiamondMax VL 20
· DiamondMax Plus 40
· DiamondMax 60
· DiamondMax VL 30
· DiamondMax Plus 40 UDMA/100
· DiamondMax 60 UDMA/100
· DiamondMax VL 30 UDMA/100