Jump to content
midibob

Tales from Midibob's workshop MPU4

Recommended Posts

Posted (edited)

Yes the diagram says 6809B so it should be a B version which is rated up to 8Mhz the crystal being 6.88Mhz,, the standard up to 4Mhz and the A version up to 6Mhz

So if you are using the B cpu then you need the B version of the 6821 for the E clock to be in spec, which as you say isn't mentioned in the schematic.

 

You might get away with a non B 6821 but technically it's out of spec.

Edited by Wizard
  • Like 2

Share this post


Link to post
Share on other sites

Thank you for that.

 

Clears the differences between versions then, i didn't know that the pia version had to match the cpu either.

  • Like 1

Share this post


Link to post
Share on other sites

ah i may have just been lucky then! the cpu speed is divide by 4 the crystal, so it actually runs at 1.72mhz where as Maygay M1 also uses a 68B09p and runs at 2Mhz as has a 8mhz xtal, I knew mpu4 run slower which is why if I get a 68B09p processor that won't run in M1 I try it in MPU4 to see if it works t the slower speed where it should definitely work!

  • Like 1

Share this post


Link to post
Share on other sites

Again, some interesting informative info, thank you.

Share this post


Link to post
Share on other sites
Posted (edited)

The 6809 quoted frequencies are for the cystal, the 6821 quoted speeds are for the E clock, so 68B21 is up to 2Mhz, the standard 6821 is only good up to 1Mhz

 

The E clock on the MPU4 runs at 1.72Mhz so you do need a 68B21

Edited by Wizard

Share this post


Link to post
Share on other sites

Some more MPU4's have come into the workshop.

The first one had all the 1nF triac capacitors corroded along with all the triac resistors and the 4 multplex switch feed resistors.

Most markings were all worn off on the top sides and their values were mostly 50% out of tolerence. The 1/4W ones fell apart on removal as did the little caps. All in all 28 resistors and 8 caps, quite a tally for an MPU4.
The 8 triac driver transistors needed changing too as they all had green legs also the 2 RAM and battery regulator transistors.

After that little lot it was time for a power up.
Expecting this one to give me a hard time I was pleasantly surprised when everthing worked. In fact the only other thing was the volume pot, which was seized. A new one sorted that.

 

Next board I started by swapping out the 8 triac driver transistors. I do this as a matter of course now unless I've been beaten to it as they do tend to leak due to corrosion even if they look good. The usual sign for this is a payslide that momentarily energises on boot up.
On power up it halted with an RS232 error, one of the motor drives was stuck on and a column of lamps too. Luckily the RS232 error was due to IC26 (MC1488) which wasn't outputting any signal. The motor drive stuck was due to IC27 (ULN2803) which had a bad output. Lastly the stuck lamp column was due to IC13) another ULN2803.

All that was then needed was a couple of rivets for the battery connections as these had rotted.

One other board I won't be bringing back to life as it was too far gone. Close your eyes if you've a nervous disposition.:lol:

004_3.jpeg.c85d70d4249f226d1c10b2e923a1e645.jpeg

Share this post


Link to post
Share on other sites

Nice to see you back on the mpu4s bob and yes dont blame you one bit for giving the one board a wide birth 😲🤐

Share this post


Link to post
Share on other sites

Jesus that looks proper nailed!

Share this post


Link to post
Share on other sites

Nice work Bob.

Amazing how you get these old boards

running again.

That picture sends shivers down the ole spine. 

Share this post


Link to post
Share on other sites

A few more MPU4's just completed.

This one booted but only 4 of the LED's came on, a quick blip of an alarm, a row of lamps, all the triacs bar the last one and would then freeze! Also no alpha which is par for the course. At one point whilst moving the board the alpha burst into life and the other 4 LED's lit up. I could get this to happen by flexing the board slightly. Trying to trace it by turning the board over stopped it happening, so I was back to square one.

Hmm, was it a bad chip, usually something like that you'd suspect a broken track or a corroded one. No evidence was found of that so I thought I'd start with IC8 as that controls the triacs. On switch on the triacs were all on so I couldn't tell if it was doing the triac test or not? Before removing IC8 the 4049 chip below it looked really badly corroded so I thought I'd get that removed first in case there were any corroded tracks underneath. The solder had gone to that grey crumbly stuff and no matter how much flux and resoldering it wouldn't come out. The only way is to cut all the legs off and remove each one separately. All was good underneath so a new one was fitted. Luckily the 68B21 (IC8) came out a bit easier and with a new one fitted I was greeted with the triac test working, all the LEDs and the alpha.:clap:

Everything now appeared to be OK except the last triac. This was due to a short circuit zener diode.

The next one needed a new RAM chip before it would boot, the alpha wasn't running either. Not due to corrosion this time but a short circuit zener on the reset pin. On boot up it was alarming but until I had the alpha going I had no idea why. Anyway with it now functioning I was greeted with IRQ SLOW ALARM!
Blimey what causes that? I had a light bulb moment and remembered something from another tech that said this is controlled from the 6840 and with a new one fitted it passed all the tests.:)

Last one for now....
This one tried to boot  but there was no alpha, no sound, no LED display or triac test. To save on all the grisly details IC1 (6840 again) and IC8 (68B21 again) had both failed.

Share this post


Link to post
Share on other sites

Awesome stuff bob your brain must be ready to explode 🤓😁.top notch work as usual and so good to read .well done 👏

Share this post


Link to post
Share on other sites

×