• Please review our updated Terms and Rules here

Cheapo IDE interface problems

Time to acquire a TRS-80 John! To JonB don't take insult but make sure your IC's are oriented correctly. They face each other on the board.
 
None taken. :)

ICs are oriented correctly. It'd be difficult to get this wrong, because the screen print on the board is so clear.

James, I get zero activity when the TRS80 is trying to access the drive (under CP/M or LS-DOS). In fact I have never seen the activity light flash. I haven't cleaned flux off the board yet. Thought it wasn't conductive, how is it a potential problem?

When I get a chance it's out with the probe. Anything else I should be looking for? How can I test the TRS80 end? How robust are these ICs? Is it likely one of them is damaged by static? I mean, are any of them prone to this?
 
It's as if the TRS-80 wasn't talking to the board. I mean that is the way it seems to be acting. When you connect the 50 way connector onto the bottom of the Model 4 make sure pin 1 RED line is facing away from the power cord side. (if it is the desktop version) But if that was happening the warning light might light up on the cheapo. Just for the heck of it try reversing the 50 way and see if it helps at all. It won't hurt the Model 4. I have done it before when I was developing the prototype. Reversing the 50 way.

Oh and you are using this command: SYSTEM (DRIVE=X,DISABLE,DRIVER="XXXXXXX") XXXXXXX is the driver name just as an example. This is of course using LS-DOS 6.3.x Then trying to format drive X.

Just some suggestions.
 
Are you getting anything like this under LS-DOS? This was using my prototype. Acts the same with the final board.
 

Attachments

  • 100_1510.jpg
    100_1510.jpg
    95 KB · Views: 1
Err I mean the pins on the CF card male connector, the bit you actually slide the CF card into, not the IDE interface connector. It has 2 raised pins that indicate the +5v pins that are in the exact middle of the pins, and they both have power, ergo the CF card is powered. According to this page (http://pinouts.ru/Memory/CompactFlash_pinout.shtml) the two pins are 13 and 38. It's Pin 20 on the IDE connector, and it is connected to power and the screw terminal according to my voltmeter and continuity checker.

Re the picture, I don't get that. It just locks up, I never get back to the LS-DOS prompt. I used the same command as you are showing (actually, per the instructions for the LS-DOS driver).

Edge connector is correctly oriented. Wrong way round gives FAULT light on the board.

If the TRS80 isn't talking to it, how is it able to read the 80 value back from the status register? Is that a coincidence? What does the status register value actually mean? Do the bits have any significance?

This is a non GA Model 4 by the way - not sure that's pertinent...

How to clean the flux off? It's old school rosin cored solder. I brushed it with some methylated spirit (you might call it "wood alcohol", it is methanol) and I got a sticky residue on my hands. The flux seems to be gone, but the board lacquer is disturbed too. That said, I'm still getting 80 and no format when I run TEST.COM in CP/M.

I haven't had time to get the logic probe out thanks to a cancelled train... coming soon...

Thanks :)

JonB
 
I Thought you meant pins 13 & 38 on the IDE connector. My misunderstanding! Are you using the correct driver in LS-DOS named HDDVR6/DCT?. As far as CP/M I do not know what the driver name is. I do not use CP/M. Getting the 80 value is a good sign though that's what you should be getting. Means the M4 is talking to the interface. Sorry for obvious stuff just want to be sure.
 
Yep, that's the driver I used for LS-DOS. It was in the download. I don't think I have got any of the software steps wrong. They are pretty straightforward. Must be a hardware issue.
 
That's what I am thinking. Do you have any spare IC's you could try to substitute? My 1st suspect would be the 74xxx245 then the other chips. 1 at a time. See if it helps any. And then I used LS devices on the prototype and they worked just fine. Good luck keep us posted.
 
Jon, sorry you're having trouble getting this running. If you have an analyser or scope, that will be by far the best approach.

The board itself is quite simple:

  • U1 is an address decoder to generate the CompactFlash CS signal (/ADDRSEL in the schematic) at the appropriate port range, hence follows the address bus
  • U2 is a bus transceiver, gated by /ADDRSEL generated by U1 and direction controlled by the logic in U3
  • U3 generates the CompactFlash /IOR and /IOW signals from Z80 signals /IN, /OUT and /IORQ
  • U4 passes the address bus A0,A1,A2 from Z80 to CompactFlash, and also the /EXTIOSEL signal back to the TRS80. This signal is generated by U3 as /IOR OR /ADDRSEL; U4 is just serving as an amplifier to overcome the 150R pull-up in the TRS80

Re flux; personally I clean boards with isopropynol usually.

Hope that helps!
 
Spares are cheap to buy on eBay, or from Mouser.

And, unless you have some very expensive equipment, with a circuit this simple -- doing chip swaps IS how you find the "true cause." There are so few chips, it's not "random chance." Frank built the prototype, and gave you the most likely culprit. What would it hurt to try a new chip in that location?
 
Let me see what I can do with the logic probe first. One never knows....

My main problem is lack of time, but I am hoping to have a good look at it over the weekend.
 
Also something else is to look at is to pull each chip one at a time and verify that none of the pins bent under the IC and not making contact. If you have not already done that. Good luck! Hope you find the cause. Also maybe send some close up pictures of the top and bottom of board. May help us out. :D
 
pearce_jj said:
If you have an oscilloscope or logic analyser, I would suggest starting by looking at U2 pin 19 (buffer gate, generated by U1) and U2 pin 1 (buffer direction).
Click to expand...

U2 pin 19: Pulsing continuously
U2 pin 1: Pulsing (slower than pin 19) but only when the test program is running

I have checked all the IC pins and they are connected and showing signals. More than that I can't say because I don't have enough expertise. If you can guide me through a debugging process I think it can be sorted out.
 
pearce_jj said:
  • U1 is an address decoder to generate the CompactFlash CS signal (/ADDRSEL in the schematic) at the appropriate port range, hence follows the address bus
  • U2 is a bus transceiver, gated by /ADDRSEL generated by U1 and direction controlled by the logic in U3
  • U3 generates the CompactFlash /IOR and /IOW signals from Z80 signals /IN, /OUT and /IORQ
  • U4 passes the address bus A0,A1,A2 from Z80 to CompactFlash, and also the /EXTIOSEL signal back to the TRS80. This signal is generated by U3 as /IOR OR /ADDRSEL; U4 is just serving as an amplifier to overcome the 150R pull-up in the TRS80
Click to expand...

With test.com running, using a 3 state logic probe (high / low / pulse):-

  • /EXTIOSEL is pulsing at U4 pin3
  • /ADDRSEL is pulsing at U1 pin 19
  • /ACTIVITY is high all the time (U4 pin 9). Pulling it low causes the Activity LED to light up.
  • IORD is pulsing (measured at U2 pin 1 Direction)
  • IOWR is high (U3 Pin 6)
  • IDE A0 - A2 all pulsing at U4 pins 12, 14, 16 respectively
  • IDE *RESET is high at U4 pin 18
  • *DASP is high at U4 pin 11
  • *M1 is pulsing at U4 pin 19
  • IDE D0 - D7 all pulsing at U2 pins 11-18
  • U4 pin 17 pulsing (output of U3 pin 8 )
  • ADR 0-7 all pulsing at RN1
  • A3-A7 all pulsing at U1 pins 3, 12, 14, 16, 18 respectively
  • D0-D7 all pulsing at U2 pins 2-9 repectively
  • At RN1: A0-A7 all pulsing. Pin 1 is high (Vcc)
  • At RN3: IDE D8-D15 all high is this because we are only using D0-7?
  • At RD2: 1-Vcc:high, 2-/IN: pulse, 3-/OUT:high, 4-/RESET:high, 5-IORDY:high, 6-/IOREQ: pulse, 7-NC:high, 8-U1-ADDR-SEL1:low, 9-/M1: pulse
 
Last edited:
Nuts.

Just spotted the IC at U4 is a 74HCT241N and it's supposed to be a 74ACT241N. I bet that is the problem - any comments? The PCB actually says it must be ACT in the screen printed notes. Arrgggh! (etc...)

I hope James has one in stock - I have asked for a replacement, as this is what came in the kit...

Fingers crossed this'll fix it!
 
Oh! Yes certainly that will stop it running. I'll get the correct chip out to you ASAP.

Sorry about this.
 
You must log in or register to reply here.
Back
Top