• Please review our updated Terms and Rules here

Powertran Cortex

Sorry Nige, didn't mean to impose on you - especially on someone else's behalf! I didn't realise your Cortex was in an 'unknown' state. If Andy's Cortex has been stuck in a cupboard for donkey's years then I'm sure another couple of weeks won't matter if you're up to you eyes with things at the moment. ;-)

Stuart.
 
Stuart - I'll go and have another check to see if I've got it wrong.
The LED numbering and circuit board labels match your description but I might have the LED wires in the wrong position on the plug.
I don't have the original construction info for the LEDs so maybe I always had it connected incorrectly.
With regard to the eproms from Nige, I'm pretty sure that they aren't the cause of the problem because I get similar issues when using the forth eproms.
And the forth eproms have been recently read/copied by Nige and the code used in the emulator.

Nige - Stuart's right, there's no rush to get my Cortex fixed. I'm greatful for the help but you sound pretty busy at the moment so I'm more than happy to wait.
 
Stuart - Just checked my front panel LEDs and I do have the plug configured as follows:
PCB IAQ = RUN
PCB MAP = BASIC
PCB TIME = MEM
PCB IDLE = IDLE
PCB 5V = All LEDs
So it appears that it has always been wrong and all I've succeeded in doing is misleading you - So I suppose my original post should read:
With the reset button held in, the "BASIC" light is lit. When the reset button is released the BASIC light stays on, the RUN light comes on for approx 1 second and then I get a continuous tone plus BASIC led.
I now feeling guilty about taking up everyones time.
 
I've checked the psu (a bit of a pain because someone cut the mains lead off), and shorts on the motherboard, both ok. however I've got to the stage where I need an aerial lead, which is in the shed, and it's sleeting.
It's almost but not entirely tempting, but it's quite a way to the shed across a couple of slippy planks over the ex-pond, in the dark. so i'm staying warm till tomorrow morning.

It's always a fight between doing the washing/washing up, feeding the children/cat, putting another bit of ring-main/plumbing/plasterboard in, being sociable or fixing an "orange glowy thing" (wife's reference to Toshiba3100 etc), as the latter is something I enjoy & me being the result of a protestant upbringing, the latter usually misses out ;-(
 
It works!!! .... well almost, colour/vertical sync are a bit out of sorts on the UHF adapter. I haven't tried component video out yet. but I think we have a test-bed machine. Ran a sample basic program, fine.

Now to put a bit more floor down & earn some "ram testing" time.
 
nige the hippy said:
It works!!! .... well almost, colour/vertical sync are a bit out of sorts on the UHF adapter. I haven't tried component video out yet. but I think we have a test-bed machine. Ran a sample basic program, fine.

Now to put a bit more floor down & earn some "ram testing" time.
Click to expand...

Great news! The vertical sync issue may be due to a TMS9928 being fitted instead of a TMS9929. The 9928 has NTSC video timing while the 9929 is PAL. I had an old Fergusson TX9 TV and there was a simple resistor change to get it to sync up to the 30 frames/second.
 
Stuart, I've had a quick check with the logic probe ( set to TTL/LS and PULSE ) and got the following results:
IC47 - CS2 = High initially, High when RESET held in and pulses for short duration when RESET released - Pulsing stops when RUN led goes out.
IC46 - CS2 = High initially, High when RESET held in and pulses for short duration when RESET released - Pulsing stops when RUN led goes out.
IC45 - CS2 = High initially, High when RESET held in and pulses for LONGER duration when RESET released - Pulsing stops when RUN led goes out.

ADDRESS LINES
The address line results can be divided into groups:
A1, A2, A3, A4, A6, A7
Without touching the reset button these are continually pulsing.(high, low and pulse lit)
With the reset button held in, these address lines are LOW and start pulsing when RESET is released and then remain pulsing.
A0, A5
Without touching the reset button these are showing LOW with the occasional pulse LED every 3- 4 seconds
With the reset button held in, these address line are LOW and start pulsing (high, low and pulse lit) when RESET is released, but only for the length of time that the RUN led is lit, then they return to their normal state of LOW with the occasional pulse LED every 3- 4 seconds.
The question is, have I got the correct settings on the logic probe ? ?

Nige - Great news about your Cortex - I could always try punishing mine by cutting its mains lead off !!
 
tms9995 said:
Great news! The vertical sync issue may be due to a TMS9928 being fitted instead of a TMS9929. The 9928 has NTSC video timing while the 9929 is PAL. I had an old Fergusson TX9 TV and there was a simple resistor change to get it to sync up to the 30 frames/second.
Click to expand...
You know you might be right! my eyesight's gone terrible in the past 5 years or so! I'll take another look. If you're not right, That area of the motherboard is a little crustier than the rest.
However I've a microvitec cub, and a commodore 1084 monitor, It should be a case of making up adapter leads.
I can still try the other dram chips in to see if they're ok.
 
tms9995 said:
Great news! The vertical sync issue may be due to a TMS9928 being fitted instead of a TMS9929. The 9928 has NTSC video timing while the 9929 is PAL. I had an old Fergusson TX9 TV and there was a simple resistor change to get it to sync up to the 30 frames/second.
Click to expand...
There are also two mods in the area of the sync separator circuit shown on the latest circuit diagram. This might be to improve the vertical sync?
 
Andy.

(Logic probe settings sound fine).

The EPROM chip selects sound fine - so that is another area of the circuit that seems to be working OK.

Looks like you probed the address lines on the RAMs (A0 - A7), rather than the EPROMs (A0 - A12)?
 
AAAAARRGGHH no component video out (in spite of the hole on the backplate) it looks like there were issues with the sync, as there are hand drawn mods on the circuit diagram. There's monochrome composite out from the chip, so all is not lost. but I think it's out with the 'scope tomorrow, and a good root around.

p.s Don't answer this one, Andy's problems need the attention.... incidentally it couldn't be a damaged eprom socket could it?
 
Sorry Stuart - It's just me being a plonker again.

The info I gave you was for the Eprom pins A0 - A7 but I didn't spot A8 - A12 on the other side.
I'll have to give you the next thrilling instalment for A8 - A12 tomorrow.

Are my results what you would expect for the first 8 address lines?
 
They sound sort of OK! The activity when you release the reset button is good, but I'm not sure where this pulsing every 3-4 seconds is coming from.

When you release the reset button, the EPROM data is copied to RAM, so you should see activity on all the EPROM address lines for that 1 second. When the buzzer starts, the processor goes into a tight 'loop of death', so I would expect some lines to be static high, some static low, and a few pulsing. Not sure where the 3-4 second pulses are coming from - the processor shouldn't even be responding to interrupts.
 
nige the hippy said:
it couldn't be a damaged eprom socket could it?
Click to expand...

It's a possibility. IC47 is working well enough to at least read the reset vector from, and IC45 is working well enough to at least read the first part of the boot code.

Andy, can you pop the probe round all the address pins on *all* the EPROMs and check you have activity when you pop the reset button? Also check the *CS1 pin on each EPROM - should be pulsing even after the buzzer starts (I think). The PD pin on each should be static low. Check the power supply pins.

Nige - what might be a good way to check the data pins? Power off then check the resistance between the data pins on adjacent sockets?
 
Stuart,
I've re-checked all the pins and listed the indications below. The random pulse I was seeing every 3-4 seconds on A0 + A5 seems to have disappeared - Maybe it was my probe randomly triggering. Also the data pulse on A11 seems slower or a different format to that seen on other address pins - If this means anything.

ADDRESS LINES
The address line results can be divided into groups:
A1, A2, A3, A4, A6, A7, A12
Without touching the reset button these are continually pulsing.(high, low and pulse lit)
With the reset button held in, these address lines are LOW and start pulsing when RESET is released and then remain pulsing.
A0, A5, A8, A9, A10, A11
Without touching the reset button these are showing LOW
With the reset button held in, these address line are LOW and start pulsing (high, low and pulse lit) when RESET is released, but only for the length of time that the RUN led is lit, then they return to their normal state of LOW. * (ON A11 THE DATA PULSE SEEMS TO BE SLOWER THAN THE OTHERS)

CS1 (x3) = Initially pulsing High / Low -- Reset button in = High --- Button released = Pulsing High / Low
CS2 / IC47 = High initially, High when RESET held in and pulses for short duration when RESET released - Pulsing stops when RUN led goes out.
CS2 / IC46 = High initially, High when RESET held in and pulses for short duration when RESET released - Pulsing stops when RUN led goes out.
CS2 / IC45 = High initially, High when RESET held in and pulses for LONGER duration when RESET released - Pulsing stops when RUN led goes out.

PD (x3) Continuous Low = Uneffected by reset
Voltage between pin 14 and pins 1/26/28 = 4.85v - All Eproms
Q1 to Q8 = Initially High -- Reset button in = High --- Button released = Pulsing High / Low for duration of RUN led then continually High.

I also had a quick check of :
Pins 11 to 18 of IC44 = Initially no indication on probe -- Reset button in = No indication -- When the Reset Button is released, these are active for duration of RUN led then back to no indication.
Pins 5,6,7,9,10,11,12,13 of all the RAM chips = Initially pulsing High / Low -- Reset button in = Low --- Button released = Pulsing High / Low (High LED brighter for duration of RUN led)

The continuity check between the Eprom pins was also all good.

Andy
 
Last edited:
Well you have seem to have activity on all the lines - nothing has failed to pull a line permanently high or low.

What I would suggest is that you now wait until Nige has laid a bit more flooring and has earnt (!) the time to test his spare set of RAMs and post these down to you. Try these with your Forth EPROMs. It is possible that Forth isn't working because you have a bad RAM chip, and Basic isn't working because of a dodgy EPROM checksum (I don't think the Basic code is actually reading anything from RAM up to the point where the buzzer starts, and so it wouldn't notice a bad RAM chip). Then perhaps your Basic EPROMs back to Nige to test in his Cortex. If your Basic EPROMs are proven good - then on to plan B - whatever that is!

Stuart.
 
OK Stuart, Sounds like a plan !
Once again, thanks for all your time and effort, I didn't have a clue what to check.

Nige, I appreciate you rushing around and getting your Cortex up and running, but remember there's no rush, I can wait as long as you want.
It sounds like you've already got a list of jobs as long as your arm !
 
I haven't plugged the scope in, but I thought "what the hell..." so I've pulled one of the DRAMs Lo & behold it just sits there being dead with the "Basic" light lit, and the "Run" light glowing dully (does that sound familiar?). I popped one of the replacement Drams in, and it fires up fine, so they work in the machine as a valid substitute.
I can't test drams on my eprom programmer (it only does static ram), and I'm unwilling to do a lot of dram swapping on my cortex, so I'll pop 9 chips in the post in the hope that at least 8 will work. You probably only need one, so that gives you a set of 8 chips that will probably work to put in something else.
I suggest you swap them in one at a time so you know which one is faulty.
Of course, it could be something else!

Edit:- Just popped them in the post, but they probably won't go till monday morning now (there used to be a sunday collection from the box opposite :( )
 
Last edited:
You must log in or register to reply here.
Back
Top