Unable to format microdrives

[Bytesguy]

Hi folks, new here Picked up a QL pretty cheap on Ebay a while ago, but only just getting round to tinkering with it. It overall seems in good condition, it has been repaired in the past (there is a sticker on the bottom of it that says so). The keyboard membrane ribbon cables were cracked and not working so replaced that and all seems fine now.

The problem I am having is with the microdrives. I am using some known good cartridges (used on a 48k spectrum) but I cannot seem to format them successfully for use on the QL. Every time I run "format mdv1_" or "format mdv2_" the drives wirr to life for a while then just say "format failed".

I have cleaned the head and checked and cleaned the rubber roller (both seem fine). With the top case off, I can see the tapes are moving freely when the motor kicks on and the replacement foam pads are pressing the tape against the head fine.

I'm going to dig out my speccy microdrives later and see if those will work with the QL or not, but I was wondering if anyone has seen this before? I did a search and didn't really come up with too much for this issue.

I was able to load some software, albeit intermittently, from some pre-made cartridges, so it seems writing is the issue.

Thanks!

[Chr$]

Obvious thing that I sometimes forget, much to my own frustration: Is the write protect tab present on the right side of the cart? If it isn't, that results in unable to format errors, without actually telling you why.

And if they are missing, cover the gap carefully with some tape.

If the write protect tab is present and they are therefore writable/formatable, then it's a good idea to check the output of the 7805 regulator and the capacitance of the 2 capacitors that are easy to get to at the tops of the mdv pcb.

Then there are other things to check, but that's a start!

[Bytesguy]

Hey Chr$ thanks for the quick reply Yeah the ones I am trying to format have never had the tab snapped off. The cartridges that read (sort of) ok were ones with no tab on them (quill, abacus etc)

[tofro]

When a format fails, this could mean the erase head is faulty.
Have you tried writing to a QL-formatted drive? If that works, it is definitely the erase head, if it doesn't, it could also be the write head.

Tobias

[Bytesguy]

When a format fails, this could mean the erase head is faulty.
Have you tried writing to a QL-formatted drive? If that works, it is definitely the erase head, if it doesn't, it could also be the write head.

Tobias

Hey Tobias,

I have a box full of microdrives that came with the QL. I replaced the foam pads on all of them, I can't seem to write to them from what I can see. I'll try all of them shortly.

The only ones I have been able to read from (and it doesn't work very well) are the standard cartridges with quill, abacus etc

[Bytesguy]

I just hooked up the 4 spectrum microdrive I have and each one failed to format, I used multiple different cartridges in each too, so strange!

[tofro]

I just hooked up the 4 spectrum microdrive I have and each one failed to format, I used multiple different cartridges in each too, so strange!

What's your power supply like? Maybe the AC is missing? Can you check the +/- 12V supply?

[Bytesguy]

I just hooked up the 4 spectrum microdrive I have and each one failed to format, I used multiple different cartridges in each too, so strange!

What's your power supply like? Maybe the AC is missing? Can you check the +/- 12V supply?

I will need to see if I can find a multimeter somewhere, but everything seems to work fine apart from this format issue (the drive motors etc are all working great)

[Bytesguy]

So doing some more testing today and still unable to get things working. I am wondering if it's detecting cartridges as write protected? Maybe something is wrong with mdv1 and that is having a knock on effect with the other drives down the chain

[martyn_hill]

Hi Bytesguy!

The timing of the Format failure should give a reasonable indication of whether the WP is the cause - the format process on the QL will write for the first 7 seconds regardless as the ZX8302 has no way of knowing (literally, no connection to) the WP switch, so it tries to write a whole cartridge in this first phase, before then attempting to read-back the (special format) sectors just written.

If the error occurs _immediately_ after 7 seconds, there's a good chance the special format sectors never got written to the cartridge as the WP switch was causing the Erase current to be effectively disabled (and the ULA 'fooled' in to thinking it was really a RD operation rather than a WR.) Compare that to the Spectrum MDV for example, where the ULA IS connected to the WP switch and the FORMAT or WR operation is instead disabled in software after reading the status of the switch.

If alternatively, the error occurs 10-15 seconds later, then the initial write was 'successful' (even if marginal) and implies that the error was in reading back the 'normal format' sectors in phase 2 of the QL Format process. In fact, QDOS will object to a 'Total found' sector-count that is too HIGH or a (IIRC) a 'Good' sector-count that is way below the total-found. Either case would cause a Format failed error - but at the very _end_ of the 30 second cycle - not during it.

Of course, if either of the Erase or RD/WR heads (RD/WR are the very same heads - one per track) or if the MDV ULA itself is borderline or capput, then you probably won't be able to deduce anything conclusive from simple observation - you'd need instead to probe with a good multimeter or, better still, check the real-time signals with a simple digital oscilloscope of some of the (digital) lines coming from the ULA and comparing wit the signal traces shown in the QL Service Manual (or the updated single-sheet PDF version available on Dilwyn's site that should be easier to read than the blurred original.)

That probably sounds like a lot of messing-about just to determine whether these MDV units are worth persevering with - some members here for example might advise you to forget it and focus on more productive use of your beautiful QL... Rich Mellor often has had spare old units available for sale in the past...

In my own case, with a deep fascination with these little drives, the luxury of having spare MDV units to test against, a simple digital oscilloscope to hand and, some would argue, too much time on my hands, I've spent many an hour troubleshooting MDVs - being able to 'tune' some units to perform better, but generally to identify those that are simply 'dead.'

I've recently identified a couple with defective ULAs and undertook my first 'open-heart' transplant of a ULA on one such unit to bring it back to life. NOT for the faint of heart, though satisfying...

Good luck!
Martyn.