CPC 6128 - white screen, black border

mmldrm · 16:21, 03 January 25

Quote from: Bryce on 15:11, 03 January 25Have you made a HAL adapter to swap the upper and lower banks?

I haven't made one, due to having no 20 pin sockets, but I did previously use wires to manually create the adapter (from your post, via Noel's video, and McArti0's suggestion).

I could do that again. (I'm also tempted to get a 20 pin socket to make a permanent adapter).

McArti0 · 17:09, 03 January 25

You need to be sure that after inserting the IC or adapter all connections are secure and stable. Sometimes it is thought that it is, but it is not.

Check by oscylloscope signal RAS, CAS, CLK 4 MHz, READY.

On Probe setting 10x. And trigger in oscyllo single.

Bryce · 13:34, 04 January 25

Quote from: McArti0 on 17:09, 03 January 25You need to be sure that after inserting the IC or adapter all connections are secure and stable. Sometimes it is thought that it is, but it is not.

Check by oscylloscope signal RAS, CAS, CLK 4 MHz, READY.

On Probe setting 10x. And trigger in oscyllo single.

If he's getting a border and screen area, the CPU and the CLK are working correctly.

Bryce.

mmldrm · 14:12, 04 January 25

I did all the tests, then realised that it should have been with the upper bank adapter in place.

Well, here are the results anyway. RAS, CAS, READY, CCLK (1MHz) from the 40010 and also Z80 CLK (4MHz).

I also checked the readings from one of the lower RAM bank ICs (IC134). In fact, the other RAM chips had identical results. Once I format and label the images, I'll share those too.

mmldrm · 14:27, 04 January 25

IC134 (lower RAM chip) readings

McArti0 · 16:04, 04 January 25

Quote from: mmldrm on 14:27, 04 January 25IC134 (lower RAM chip) readings

Pin15 CAS0 is null. How have You configure PAL 40030?

mmldrm · 16:19, 04 January 25

Quote from: McArti0 on 16:04, 04 January 25
Quote from: mmldrm on 14:27, 04 January 25IC134 (lower RAM chip) readings
Pin15 CAS0 is null. How have You configure PAL 40030?

The HAL (40031 here) is the original chip. No override.

mmldrm · 17:36, 04 January 25

I'll check all the pin readings on the 40031, to have a better overview of that IC.

McArti0 · 17:38, 04 January 25

When CAS on GA is present then CAS0 or CAS1 is copy CAS.

Try CAS, CAS0 and CAS1 but on PINs/Legs HAL(PAL).

mmldrm · 13:22, 06 January 25

CAS (pin 16 on GA/40010) has a nice clear binary signal.

CAS0 and CAS1 on the HAL are both still null, like I shared previously.

I see there is also NCAS on the HAL but I could find no description for it. I measured NCAS and it looks the same as CAS on GA..?

eto · 14:21, 06 January 25

Quote from: mmldrm on 13:22, 06 January 25I see there is also NCAS on the HAL but I could find no description for it. I measured NCAS and it looks the same as CAS on GA..?

NCAS on the HAL is CAS from the GateArray.

CAS0 should be identical to NCAS as long as RAM banking is not active. If the HAL is inserted and there is no CAS0 signal, then the HAL is not working.

Time to make the HAL adapter. If you don't have a socket you could also make one using pin headers - even a permanent/robust one with a perfboard.

mmldrm · 15:26, 06 January 25

Quote from: eto on 14:21, 06 January 25NCAS on the HAL is CAS from the GateArray.

CAS0 should be identical to NCAS as long as RAM banking is not active. If the HAL is inserted and there is no CAS0 signal, then the HAL is not working.

Time to make the HAL adapter. If you don't have a socket you could also make one using pin headers - even a permanent/robust one with a perfboard.

Thanks. I had previously shoved jumper wires into the HAL socket, but I have a 20pin socket arriving in the next couple of days.

Will get a HAL adapter soldered up and see if that improves things.

McArti0 · 15:29, 06 January 25

https://www.cpcwiki.eu/forum/amstrad-cpc-hardware/banked-video-ram/msg231994/#msg231994

Check all connections PAL/HAL (after insert) with others path point.

mmldrm · 15:35, 06 January 25

Quote from: McArti0 on 15:29, 06 January 25https://www.cpcwiki.eu/forum/amstrad-cpc-hardware/banked-video-ram/msg231994/#msg231994

Check all connections PAL/HAL (after insert) with others path point.

Thanks. I'll do that while I wait for the 20 pin socket to arrive.

mmldrm · 18:24, 06 January 25

In this image, "OK" means "measurements were consistent across test points". I don't know if the values themselves are correct.

Measurements marked "OK?" were consistent across test points but voltages looked bad to me, i.e. not high or low.

Finally, I couldn't see where to test pin 20 elsewhere on the board. As it is Vcc and was 5V, I figured it didn't matter too much.

McArti0 · 19:34, 06 January 25

What have you on screen after switch ON CPC?

mmldrm · 20:15, 06 January 25

Quote from: McArti0 on 19:34, 06 January 25What have you on screen after switch ON CPC?

Still the white screen with black border (the one that usually indicates a RAM issue).

eto · 20:28, 06 January 25

Pin 16 is CAS1 and 1.2V does not seem right. That should always be high unless someone has activated banking.

Pin 17 is CAS0 and LO seems totally wrong. That means CAS0 would always be set. What do you get on Pin 7 (NCAS) or Pin 31 on the GateArray (CASAD)?

Pin 2 should be identical on IC114 and IC115. If not you have a crack in the lanes. Pin 35 on the GateArray is not connected to D0 as tha Gate Array is isolated from the data bus via a latch.

McArti0 · 22:10, 06 January 25

Do it again.

https://www.cpcwiki.eu/forum/hardware-related/cpc-6128-white-screen-black-border/msg238062/#msg238062

Bryce · 14:41, 07 January 25

Quote from: eto on 20:28, 06 January 25Pin 16 is CAS1 and 1.2V does not seem right. That should always be high unless someone has activated banking.

Pin 17 is CAS0 and LO seems totally wrong. That means CAS0 would always be set. What do you get on Pin 7 (NCAS) or Pin 31 on the GateArray (CASAD)?

Pin 2 should be identical on IC114 and IC115. If not you have a crack in the lanes. Pin 35 on the GateArray is not connected to D0 as tha Gate Array is isolated from the data bus via a latch.

Pin16 comment: Why? All RAM needs to be refreshed.

Bryce.

McArti0 · 14:49, 07 January 25

Quote from: Bryce on 14:41, 07 January 25All RAM needs to be refreshed..

RAM is "RAS only refreshed". Don't need CAS.

Bryce · 14:55, 07 January 25

Quote from: McArti0 on 14:49, 07 January 25
Quote from: Bryce on 14:41, 07 January 25All RAM needs to be refreshed..
RAM is "RAS only refreshed". Don't need CAS.

Sorry, you're right. The coffee machine is broken today. I probably shouldn't be posting at all.

Bryce.

mmldrm · 17:42, 07 January 25

Quote from: eto on 20:28, 06 January 25Pin 16 is CAS1 and 1.2V does not seem right. That should always be high unless someone has activated banking.

Pin 17 is CAS0 and LO seems totally wrong. That means CAS0 would always be set. What do you get on Pin 7 (NCAS) or Pin 31 on the GateArray (CASAD)?

Pin 2 should be identical on IC114 and IC115. If not you have a crack in the lanes. Pin 35 on the GateArray is not connected to D0 as tha Gate Array is isolated from the data bus via a latch.

Both pins 16 and 17 are definitely both consistently LO. Just checked again to confirm.

NCAS (pin 7 @ HAL) is a nice clean HI/LO signal.
CASAD (pin 31 @ GA) is also a nice clean HI/LO signal.

Pin 2 on IC114 and IC115 are indeed identical.

The 20 pin socket should arrive in the next day or two - I'll make that HAL adapter and see what readings I get then.

McArti0 · 18:17, 07 January 25

Either PAL/HAL does not work or nCPU does not reach.

mmldrm · 20:34, 07 January 25

Quote from: McArti0 on 18:17, 07 January 25Either PAL/HAL does not work or nCPU does not reach.

Where does nCPU originate? I can't find any references to it. Only on the PAL pin-outs.

I thought I would try to trace the origin (still learning, here) and try to measure the value before it reaches the PAL. (Also learning to call it PAL and not HAL.)

Edit: I see it on the GA, in the circuit diagram. But I can't tell how that relates to the PAL - difficult to tell from the diagram.

News:

CPC 6128 - white screen, black border