Robotrontechnik-Forum

Registrieren || Einloggen || Hilfe/FAQ || Suche || Mitglieder || Home || Statistik || Kalender || Admins Willkommen Gast! RSS

Robotrontechnik-Forum » Technische Diskussionen » EC1834 replace EFS connectors to ISA-8 without success » Themenansicht

Autor Thread - Seiten: -1-
000
05.08.2018, 14:52 Uhr
Tronix



Hello,

Sorry, i don't speek Germany, so I will write this message in English.

I have EC1834 mainboard, Power Supply, MON videocard, FDC controller, MFM HDD controller and Cetronics controller. Thats all..

First of all, i solder adapter for MON videocard for connect it to MCE2VGA video scandoubler. I get stable monocrome video on modern VGA monitor.
Then I read on this forum (with help Google Translate) about modified firmware of MME U8821M controller for support AT (or PS/2) keyboard protocol. So, i burn 2716 EPROM with new firmware and get PS/2 keyboard worked very well.
Then i connect 360Kb floppy drive to Floppy Disk Controller and booted to MS-DOS v3.30. Computer worked OK.

And finaly, I replaced on the EC1834 motherboard one EFS slot on the ISA-8 slot. But, without success. I tried insert XT-IDE v2.0 into ISA-8 and boot up. XT-IDE addition BIOS start up, but HDD name is corrupted - every odd character is corrupted. So, XT-IDE reported "Boot sector not found".



Then i tried insert ISA-8 VGA videocard based on PVGA1A-JK chip. Characters displayed ok, but characters attributes is random.



The main question is: does anyone have a motherboard from the EC1834 with ISA-8 slots? If so, can I ask for a photo in high resolution? I am interested in the purpose of the jumpers WD1-WD5 on circuit.

Thanks.

Dieser Beitrag wurde am 05.08.2018 um 15:17 Uhr von Tronix editiert.
Seitenanfang Seitenende
Profil || Private Nachricht || Suche Zitatantwort || Editieren || Löschen
001
05.08.2018, 17:13 Uhr
srmeister



Hello Tronix,
i just played a little with my EC1834 in the last few days.

1. Regarding the modified keyboard ROM: I decided today, i do not longer intend to improve this version, as i found it too cumbersome to program U8821M in assembler. Just want to remind you, that some special keys wont work, like "print screen"(i think you might never need those). Also, sometimes my machine boots displaying "301 Keyboard error" but when i hit F1 it works. I did not find the bug after several days of looking so i decided to give up now.

2. I also soldered an ISA slot onto my mainboard and use a VGA adapter Tseng ET3000AX. It only works with newest BIOS ROM version 731/732.
Starting from the front of the board with #1, my switch is set like this:
#1 right, #2 right, #3 right, #4 left, #5 right, #6right, #7 left, #8 right. I think only #5 and #6 switches are relevant to graphics.
I dont know which jumpers you mean by WD1-WD5 but i suppose its those white DIP switches.

greetings,
Stefan

Dieser Beitrag wurde am 05.08.2018 um 17:32 Uhr von srmeister editiert.
Seitenanfang Seitenende
Profil || Private Nachricht || Suche Zitatantwort || Editieren || Löschen
002
05.08.2018, 18:06 Uhr
Tronix




Zitat:
srmeister schrieb
Just want to remind you, that some special keys wont work, like "print screen"(i think you might never need those). Also, sometimes my machine boots displaying "301 Keyboard error" but when i hit F1 it works. I did not find the bug after several days of looking so i decided to give up now.



Thank you for replay. And big thanks for your modified firmware for PS/2 support. Unfortunately, i don't have original Robotron or XT keyboard and your firmware very much helped. I never use "print screen" and boots displaying "301 Keyboard error" do not annoing.


Zitat:
srmeister schrieb
It only works with newest BIOS ROM version 731/732.
Starting from the front of the board with #1, my switch is set like this:
#1 right, #2 right, #3 right, #4 left, #5 right, #6right, #7 left, #8 right. I think only #5 and #6 switches are relevant to graphics.
I dont know which jumpers you mean by WD1-WD5 but i suppose its those white DIP switches.



Curently i have s723/s724 BIOS version. I will try to experiment with the version 731/732 soon. Thank you.

I did not mean DIP-switch. I mean jumpers around motherboard:



And also at circuit:









Seitenanfang Seitenende
Profil || Private Nachricht || Suche Zitatantwort || Editieren || Löschen
003
05.08.2018, 19:01 Uhr
srmeister



oh ok. i checked my board and none of the jumpers WD1-WD5 is set. like on your picture.
It is not necesary to change anything for ISA Bus to work, except change the ISA slot.

Edit: btw. those maybe just "vias" as this is a 6 layer board.

Dieser Beitrag wurde am 05.08.2018 um 19:04 Uhr von srmeister editiert.
Seitenanfang Seitenende
Profil || Private Nachricht || Suche Zitatantwort || Editieren || Löschen
004
06.08.2018, 17:59 Uhr
Tronix




Zitat:
srmeister schrieb
oh ok. i checked my board and none of the jumpers WD1-WD5 is set. like on your picture.
It is not necesary to change anything for ISA Bus to work, except change the ISA slot.

Edit: btw. those maybe just "vias" as this is a 6 layer board.



This is not my picture. I found this photo via google pictures. Here is the original link: http://files.pofo.de/EC1834/EC1834_Platine.jpg

On my motherboard, almost all the WD jumpers are shorted (set). This is a high resolution photo of my motherboard: https://hsto.org/webt/ip/cw/5n/ipcw5ng7d2o2nnhghnelb__esbu.jpeg
Also on my board there are some corrections in the form of black wires in the area of 8257 IC.

I do not think it's just a "vias". For example, if WD2 not set, D24 (74LS374) latch is always selected. Otherwise the state depends on the signals DEN, MEMR and CLK(?).
Seitenanfang Seitenende
Profil || Private Nachricht || Suche Zitatantwort || Editieren || Löschen
005
07.08.2018, 18:26 Uhr
srmeister



this may be related to the byte swap logic.

at least WD5 and WD3 are somehow related to the byteswap logic.
in my opinion most or all WD jumpers should be set.
i checked WD1 WD2 WD3 on my mainboard, those are set (0 ohm), although i can not see a jumper. So either it is jumpered on underside of the board or internally connected (vias).

I can understand you try to figure out the sense of these jumpers as your problem seems to be related to the byte swap logic. (for example, every odd char on XT-IDE is bad). But i guess these jumpers would not fail easily. Most often, one of the involved IC's goes bad. And this is hard to debug.
On the other hand, the byte swap logic must work partially, because otherwise the XT-IDE BIOS would not even load.

from the documentation it states that byte swap logic is used for following purposes:
- 16 bit acces of memory on 8 bit bus (this seems to work in your case. XT-IDE Bios resides in memory on 8 bit bus)
- 8 bit acces on odd I/O adress (would be interesting if you have other plug in cards available to test if they work on EFS slot - for example COM port card "ASK")
- DMA access on odd adress ( not related to your problem)

did you measure resistance from each pin in the ISA slot to the same pin in an EFS slot?
Seitenanfang Seitenende
Profil || Private Nachricht || Suche Zitatantwort || Editieren || Löschen
006
13.08.2018, 19:59 Uhr
Tronix



Thanks for the reply. You are right, I have cut the tracks on the underside of the board, so the jumpers on the upper side are soldered. I think I have one of the first EC1834 motherboard revisions. So, WD1-WD5 on my motherboard shorted, as well as on yours motherboard.


Zitat:
did you measure resistance from each pin in the ISA slot to the same pin in an EFS slot?



Yes, i measure resistance from each pin in ISA to the same pin in EFS and the resistance is 0..1 ohms. I also checked the neighboring pins for a short to each other. All OK.


Zitat:
would be interesting if you have other plug in cards available to test if they work on EFS slot - for example COM port card "ASK"



I don't have "ASK" card. I have FDC controller (worked in EFS slot), MON video card (worked in EFS slot), 384Kb (?) RAM card (worked in EFS, so total memory is 640Kb), HDC controller (not tested yet) and Cetronics (?) LPT card (not tested yet).

Also i have chinese ISA/PCI POST diagnostic card. I tried to insert it into the iSA slot, but it shows all the time the FF code.

Dieser Beitrag wurde am 13.08.2018 um 20:05 Uhr von Tronix editiert.
Seitenanfang Seitenende
Profil || Private Nachricht || Suche Zitatantwort || Editieren || Löschen
007
13.08.2018, 20:54 Uhr
T5000

Avatar von T5000

Hello Tronix,

the XTIDE Card likes not all CF-Cards in EC1834. I have about 15 different Cards. and only one 512MB Sandisk CF Card is working. The same XTIDE Card works with more CF-Cards on my 386er.

For VGA-Cards you need BIOS 731/732 and only Cards with ET3000 and ET4000 Chip is working. Also important ist the Card-Layout. You need old (big) VGA-Cards (no SMD) with an working 8-Bit Mode. These Cards are working, but not all Video Modes works correct.
--
Polyplay ESC2 Bauform4,K8924,EC1834,KC85/3-5,KC87,KC-Compact,Lada 21011,PC1715,PC1715W,A5105,CM1910,LC-80
Seitenanfang Seitenende
Profil || Private Nachricht || Suche Zitatantwort || Editieren || Löschen
008
06.10.2018, 18:38 Uhr
Tronix



Hello all,

I did something resembling a ISA logic analyzer to understand what was going on:





I did some experiments. I made on the ISA-card with CPLD two I/O ports - 0x3a0 and 0x3a1. Both ports are writable and readable. When writing, they save bytes of data, while reading they give back the previously written bytes. The seven-segment display shows the address of the I/O port where the recording occurred. I wrote these data to the port:


Quellcode:
cli
mov dx,3a0h ; port on ISA card with CPLD
mov ax,1234h
out dx,ax
hlt



On the XT computer on the seven-segment display i see 03A1 and the computer hangs. And on the robotron, 03A0 is displayed. Accordingly, the data in the port 0x3a1 is not recorded. Only the lower part is recorded in 0x3a0.

However, the main problem became clear: the Robotron is able to write/read to the I/O ports only bytes (out dx,al). Not words (out dx,ax). Robotron has no signal on the bus / IOCS16, therefore it is assumed that _all_ expansion cards work in 16-bit mode with I/O ports.

The user Vic3Dexe on one of the Russian forums offered a solution - the revision of the "bus steering" scheme:



It is also necessary, instead of the signal MEMR, to apply for D11.10 ~ (/ MEMR & / XIOR). Or MEMR | ~ / XIOR. (not shown in the diagram).

I do this modification



And now my VGA card based on ACUMOS AVGA1 worked perfectly. Also i tested XT-IDE in Hi-speed mode, EGA cards also worked.
Seitenanfang Seitenende
Profil || Private Nachricht || Suche Zitatantwort || Editieren || Löschen
009
01.12.2019, 03:31 Uhr
Robotron EC1834



Фотографии утеряны безвозвратно с этого поста?
Seitenanfang Seitenende
Profil || Private Nachricht || Suche Zitatantwort || Editieren || Löschen
Seiten: -1-     [ Technische Diskussionen ]  



Robotrontechnik-Forum

powered by ThWboard 3 Beta 2.84-php5
© by Paul Baecher & Felix Gonschorek