MDV2 Microdrive spinning slowly all the time.

[martyn_hill]

Hi James!

I've been thinking about your 'leaky' MDV 2 and taking another look at the schematic, I'm wondering about the possibility of leakage from the MDRW line that - on QL MDVs - is connected via D23 (and D22 for MDV 1) though the WP switch in a 'clever' arrangement to enforce write-protection. This workaround was a fudge as far I can tell to compensate for the absence of a WP pin available on the ZX8302, but that's another story.

D23 is reverse biased so that current should only flow from the WP switch - in to the MDRW line - but - If D23 is hosed, then the nominally 5V WDRW line could be feeding in to Pin 10, via a pair of 2k2 ohm resistors (R99 & R101) divided with the 1k resistor to ground (R105) - which could also explain the voltage drop - though not 3-4V (I'd expect about 1.8V.)

None of the above would apply if you measured the voltage leak with the MDV disconnected, but simply disconnecting D23 (just lift one leg) would quickly validate this possibility...

Anyway, good luck

[Plasticduck]

Hi James!

I've been thinking about your 'leaky' MDV 2 and taking another look at the schematic, I'm wondering about the possibility of leakage from the MDRW line that - on QL MDVs - is connected via D23 (and D22 for MDV 1) though the WP switch in a 'clever' arrangement to enforce write-protection. This workaround was a fudge as far I can tell to compensate for the absence of a WP pin available on the ZX8302, but that's another story.

D23 is reverse biased so that current should only flow from the WP switch - in to the MDRW line - but - If D23 is hosed, then the nominally 5V WDRW line could be feeding in to Pin 10, via a pair of 2k2 ohm resistors (R99 & R101) divided with the 1k resistor to ground (R105) - which could also explain the voltage drop - though not 3-4V (I'd expect about 1.8V.)

None of the above would apply if you measured the voltage leak with the MDV disconnected, but simply disconnecting D23 (just lift one leg) would quickly validate this possibility...

Anyway, good luck

Many thanks for the reply and having a look over the schematic for me.

Yes the voltage is there with the MDV disconnected.

I have some free time in the morning, after I have done the school run, so I will get back into it then.

I can see your logic with D23 but in this case as you stated it doesn't apply with the MDV unplugged.

I also need to trawl the forum to see of an alternative method for connecting the microdrives.

I may have to make a jumoer cable arrangement for now to avoid trashing the ends of the ribbon cables.

Thanks again. James

[martyn_hill]

Hi James - you're very welcome.

OK, one more question - when you test that voltage at Pin 10 of slot#2, was that with/without an MDV unit connected in slot#1, or either way?

Some other useful check-point suggestions for when you find the time:
a) What voltage is present at Pin 2 (COMMS OUT) of slot#1 at the time of the 3-4V at Pin 10 of slot#2?
b) What voltage at the base of TR5 (centre leg) of the slot#2 drive cct? Should be less than 0.7V
c) What voltage at the base of TR7 (centre leg) of the slot#2 drive cct? Should 'float' greater then 0.7V below the collector leg (I never remember whether its still called the collector on a PNP transistor...) EDIT: I meant the Emitter, not Collector... Anyway, the one with the arrow...
c) What 'mods' do you observe on the mainboard and/or MDV board? Relying on the schematic makes the assumption that any mods were carried-out as documented, but that may need validating...

Enjoy the problem-solving

[Plasticduck]

Thanks again for the reply.

So this morning I finally had time to look again at the QL.
My first task was to measure 'a' in the post above.
Fired it up still out of the case, measured pin 10 on the mdv2 socket, 0 volts on pin 10.
Switched it off, plugged in both microdrives and mdv2 is not spinning.
No idea what has happened but the fault has gone.
So I therefore have an intermittent fault, nightmare.

Put it all back in the case and it is still good, mdv2 nice and silent and the led is off.
However although the drives spin up and sound well, and I have cleaned the heads and pinch rollers, which look like new, it still refuses to format a cartridge.
I have bought some extra ones which have been re-felted and tested from a friend but no joy.
I have QL easel on cartridge and if I dir it on mdv2 it will say something like Easel 2.24 bad or changed medium and stop.
Once it actually brought up a list in the top part of the screen with the contents.
I tried to run it but nothing.

Time to have a good look through the rest of the forum and see if I can get these to work.

Many thanks for all your help and if it comes back I will re visit this thread and start with the suggestions kindly posted in the reply above.

[Dave]

If you physically examine the wires, you might find they're in rough shape.

Buying some turned pin socket strips, and reshaping and soldering these wires into the socket strips, to get a nice, secure connection, might help.

[bwinkel67]

Thanks again for the reply.

So this morning I finally had time to look again at the QL.
My first task was to measure 'a' in the post above.
Fired it up still out of the case, measured pin 10 on the mdv2 socket, 0 volts on pin 10.
Switched it off, plugged in both microdrives and mdv2 is not spinning.
No idea what has happened but the fault has gone.
So I therefore have an intermittent fault, nightmare.

Put it all back in the case and it is still good, mdv2 nice and silent and the led is off.
However although the drives spin up and sound well, and I have cleaned the heads and pinch rollers, which look like new, it still refuses to format a cartridge.
I have bought some extra ones which have been re-felted and tested from a friend but no joy.
I have QL easel on cartridge and if I dir it on mdv2 it will say something like Easel 2.24 bad or changed medium and stop.
Once it actually brought up a list in the top part of the screen with the contents.
I tried to run it but nothing.

Time to have a good look through the rest of the forum and see if I can get these to work.

Many thanks for all your help and if it comes back I will re visit this thread and start with the suggestions kindly posted in the reply above.

Ugh, intermittent faults are a nightmare. All you an do is hope it won't go away. I suppose it could be a bad solder joint somewhere on the board.

So one thing I have used is a tool called mdvedit. It gets past the constant spinning and unrecognizing of a drive and allows you to step through it sector by sector. I had one cart that would just sit there forever and even after re-flelting it seemed completely lost. Once I ran mdvedit I've been able to go through sector by sector (254'ish) in all to see what is there and (so far 1/8th done) have been able to reclaim a good amount . Note that you have to piece them together by block id's and each file is identified by its own id which you have to sort of figure out by content but that's not too bad.

More importantly, though, it reinforces the fact that my mdv1_ drive is working just fine (mdv2_ perhaps as well though it always works differently since sometimes it reads carts the other one can't) and it stops that incessant infinite spin of bad microdives -- I wish the original QDOS ROM had added a feature to allow the microdrive driver software to recognize the cartridge was corrupt after so many unsuccessful tries at reading a sector and given an option for recover since it's not too hard to skip over a bad sector and grab the next block. I suppose Sinclair Research believed in their product so much they didn't want to admit it was crappy and needed that. For BASIC and some document files this would have worked pretty well, just not for executables. I've skipped over a bad block in the middle of a long program listing and only lost like 4-5 lines that can be recreated, with some thought, by hand.

I have a vDrive which requires no additional hardware and works beautifully in connecting the QL to the outside world via SD card -- you use an emulator (QLAY2 for me unless I need audio then it's QLAYK) and/or qlayt.exe Windows utility and unzip QL executable to get programs to and from my QL. You can find mdvedit on Dilwyn Jones software repository site tucked away in another utility called In_Out_Kit_v1_05.zip:

http://www.dilwyn.me.uk/files/In_Out_Kit_v1_05.zip

[Plasticduck]

Thanks again for the replies.

So I managed to get an hour this morning on the QL.
The fault hasn't come back and with regards the ribbon cables I have lifted the board and drives back out and checked for continuity and they all seem good.
I am worried at this point that just as with the constant spinning there is another fault somewhere that is stopping the microdrives working as intended.

When I get more time again I will have another probe about and no doubt the other fault will come back.
With regards the mdvedit that sounds like something I would like to look into.
Having browsed the forum it seems that the hardware and software available for the QL has come on leaps and bounds.
I am torn between persevering with the microdrives and sticking in a modern SD based alternative.
I would like to get them working if possible and it may be that I have to send it away as my skills are limited and they do seem to be a popular point of failure.

Many thanks again. James.