MTX512 fault: Blowing fuse.

About original Memotech hardware.
stephen_usher
Posts: 205
Joined: 27 Nov 2016 19:58

MTX512 fault: Blowing fuse.

Post by stephen_usher » 04 Dec 2016 13:06

I recently bought an MTX512 on eBay which the vendor had previously tested but arrived in a non-working state as the internal fuse had blown.

Having replaced C39 and C40 just in case I could get the machine operational by bridging the fuse, though the rectifying diodes got a bit warm.

I replaced the voltage regulator (REG2) and tried again. This time there was a smell and the system was non-operational. The TIP was quite hot.

I then put the old regulator back and removed the TIP. The fuse does not now blow but there's only 2.5V on the +5V line.

I've now replaced the TIP with a new one and again the fuse is blowing. I've not tried bridging the fuse and testing just in case more magic smoke ensues.

None of the capacitors look as though they've popped and I can't find any shorts with the multimeter.

Any ideas on where to look next?

User avatar
Dave
Posts: 756
Joined: 11 Aug 2012 18:16
Contact:

Re: MTX512 fault: Blowing fuse.

Post by Dave » 04 Dec 2016 13:24

Hi Stephen,

I guess that you don't need any comments on the "wisdom" of bypassing the fuse :D

The burning smell is not a good sign, the problem could be pretty much anywhere on the board. A good, easy anyway, starting point would be to remove all of the socketed chips and try that first.

After that, the most likely suspect components are probably VRAM, then RAM, but we can look in more detail once you have eliminated the socketed components

regards
Dave

User avatar
1024MAK
Posts: 569
Joined: 24 Dec 2012 03:01
Location: Looking forward to summer, in Somerset, UK

Re: MTX512 fault: Blowing fuse.

Post by 1024MAK » 04 Dec 2016 13:27

The 7805 voltage regulator and the TIP transistor work together to regulate the voltage for the +5V line. If the fuse is blowing, too much current is flowing, which indicates that something is breaking down and drawing too much current.

But first, if your multimeter has a diode test function, test all the rectifier diodes. Then check and test the 5.6V Zener diode that is across the +5V rail and the 0V GND rail. I forget it's ID, but it's between the modulator and the heatsink with the regulators and TIP. If the 5V line goes above the turn on voltage of this Zener diode (around 5.6V) it will attempt to clamp the voltage, often by partly failing short circuit.

Mark

User avatar
1024MAK
Posts: 569
Joined: 24 Dec 2012 03:01
Location: Looking forward to summer, in Somerset, UK

Re: MTX512 fault: Blowing fuse.

Post by 1024MAK » 04 Dec 2016 13:30

Unlikely to be the 4116 VRAM, as the main supply rail for these is the +12V. The +5V is just for the outside world interface circuity. Unless the -5V is not working, in which case, one or more 4116 may have self destructed.

Mark

stephen_usher
Posts: 205
Joined: 27 Nov 2016 19:58

Re: MTX512 fault: Blowing fuse.

Post by stephen_usher » 04 Dec 2016 14:32

1024MAK wrote:The 7805 voltage regulator and the TIP transistor work together to regulate the voltage for the +5V line. If the fuse is blowing, too much current is flowing, which indicates that something is breaking down and drawing too much current.

But first, if your multimeter has a diode test function, test all the rectifier diodes. Then check and test the 5.6V Zener diode that is across the +5V rail and the 0V GND rail. I forget it's ID, but it's between the modulator and the heatsink with the regulators and TIP. If the 5V line goes above the turn on voltage of this Zener diode (around 5.6V) it will attempt to clamp the voltage, often by partly failing short circuit.

Mark
Yeah, I know where the Zener diode is. Having a closer look, that may have died after the test with the new regulator.

That's the next thing to replace.

stephen_usher
Posts: 205
Joined: 27 Nov 2016 19:58

Re: MTX512 fault: Blowing fuse.

Post by stephen_usher » 04 Dec 2016 14:36

Dave wrote:Hi Stephen,

I guess that you don't need any comments on the "wisdom" of bypassing the fuse :D

The burning smell is not a good sign, the problem could be pretty much anywhere on the board. A good, easy anyway, starting point would be to remove all of the socketed chips and try that first.

After that, the most likely suspect components are probably VRAM, then RAM, but we can look in more detail once you have eliminated the socketed components

regards
Dave
I know... but as the circuit has a belt and braces approach to this, i.e. the Zener diode, I thought it a reasonable risk. It worked fine with all the original components, except for the elevated rectifier diode temperatures. (The +5V line ran at 4.7V during this test.)

Not sure why the change of regulator (modern LN7805CP as replacement for the LN7805C) made a difference.

stephen_usher
Posts: 205
Joined: 27 Nov 2016 19:58

Re: MTX512 fault: Blowing fuse.

Post by stephen_usher » 11 Jan 2017 22:58

OK, so after replacing the power transistor and the zenor diode (which is what let out the magic smoke) I'm back to where I was.

It still blows the fuse.

But... with a meter where the fuse should be I'm measuring 0.86 amps and the machine is operating seemingly normally.

As you can imagine, the heat sink is getting a bit toasty but that's the only component (other than the rectifying diodes after a while) which is getting warm.

Any idea what my next step might be?

Possibly C11 or C57? They're the only two electrolytics left on the +5V line. (Tricky to desolder those two.)

stephen_usher
Posts: 205
Joined: 27 Nov 2016 19:58

Re: MTX512 fault: Blowing fuse.

Post by stephen_usher » 12 Jan 2017 00:22

Removing C57 made no difference. Oh well, the tricky C11 will be next. I'll get replacement capacitors from Maplin's at the weekend and replace them both anyway.

stephen_usher
Posts: 205
Joined: 27 Nov 2016 19:58

Re: MTX512 fault: Blowing fuse.

Post by stephen_usher » 14 Jan 2017 18:32

So, the capacitors have now been changed and there's no difference.

I decided to run a longer powered test to see if I could find and chips which were getting warmer than normal:

Image

As it turns out the video chip and all the video memory chips are getting hot.

Seeing as I've already run tests with the PAL video board removed it can't be that. I've also tried swapping out the video chip with a spare I have with no change.

Getting a bit stumped now.

stephen_usher
Posts: 205
Joined: 27 Nov 2016 19:58

Re: MTX512 fault: Blowing fuse.

Post by stephen_usher » 14 Jan 2017 21:07

OK. What an idiot!

I decided to bring out my MTX500 (which has a fault so bad that it doesn't even reset the video chip) to test that as a reference to determine what the amperage through the fuse should be...

The current was even higher than on the MTX512! Then I checked the fuse side by side with the fuses I'd bought. The new fuses were one tenth the rating, so no wonder they blew. (It doesn't explain why the original fuse blew though.)

So, I've been trying to trace a non-fault.

D'oh!

Now all I have to do is sort out the keyboard switches and someday find out what's wrong with the MTX500 (which died way back in 1988 when it was connected to a BBC Micro via a serial port and just stopped working in the middle of running a BASIC program).

Post Reply