MZ-N910 recording error

[valentinc]

Hello everyone !

 

Have been struggling with a MZ-N910 that reads discs perfectly, but is unable to record.

Symptoms are the following:

- in analog input mode, the recording starts but after 1min or so it throws REC ERR error;

- in netMD mode, the recording starts but progress remains stuck;

- spindle speed can be heard going constantly up and down, like a CD player struggling to read the disc;

 

After lubricating the spindle motor (watch oil), I managed to get a full disc recording when unit sat upside down.

However was unable to consistently reproduce this behavior.

Before lubricating the spindle motor, unit was also throwing a MEM OVER error which has disappeared now.

 

Checking the unit for stored errors, I found this (see picture attached). However after clearing the errors, they do not re-appear.

[sfbp]

Is the unit plugged into the AC adapter? There has always been a problem with NetMD-capable machines writing disks using NiMH batteries (and there are no gumstick alkalines) because it takes a lot of power to write (heat) a disk.

Do you hear a lot of noise from the seeking? This is pretty normal for this unit.

 

The first thing I would do is to enter the service mode, do NOT do a 911 reset which requires a full servicing involving equipment you probably do not have.

 

Check the first voltage in the electrical adjustments. It's (from the 910 service manual) on page 19. Is it down about 10%? That will cause all your symptoms, right there. Check with and without the AC plug and with/without alkaline sidecar batteries (if you have a sidecar, should have been one included with the unit when new). Actually check all 4 (items 1-4, VC1/2 low/high). Don't adjust anything yet.

[valentinc]

Unit is not plugged into an AC adapter (do not have one), but used with an AA NiMH on sidecar.

Have tried recording with unit powered by AA alkaline and lab power supply (1.5V) and behavior is exactly the same.

 

Ended up measuring all voltages (except charging) and they are as follows (unit powered with 1.5V):

 

VC1 L - 2.3428V (spec 2.35V +/-5mV) - out of tolerance

VC1 H - 2.7423V (spec 2.75V +/-5mV) - out of tolerance

VC2 Lo - 1.423V (spec 1.3V +/-5mV) - out of tolerance

VC2 Hi - 2.543V (spec 2.55V +/-5mV) - out of tolerance

 

REG1 - 2.15V (spec 2.15V +10mV -5mV) - within tolerance
REG2 - 2.198V (spec 2.2V +10mV -5mV) - within tolerance
REG3L1 - 1.1936V (spec 1.2V +/-5mV) - out of tolerance
REG3L2 - 1.196V (spec 1.2V +/-5mV) - within tolerance
REG3H - 1.5442V (spec 1.55V +/-5mV) - out of tolerance

 

VREC L - 1.1215V (spec 1.12V +25mV -10mV) - within tolerance
VREC M - 1.222V (spec 1.22V +25mV -10mV) - within tolerance
VREC H - 1.4908V (spec 1.52 +60mV -10mV) - out of tolerance

 

Since the test points are on the back side of the board it's hard to test with battery, but can test with lower voltage.

 

In regard to seeking noise, are you reffering to noise during recording ?

[sfbp]

Aha! You cannot use NiMH in the sidecar. It's expecting a decent voltage from an Alkaline AA. Kevin, can you confirm I'm right by looking at the circuit which you understand and I do not?

 

The crummy 1.3 you get from NiMH won't do, recording takes too much power.

 

https://www.ebay.com/itm/186300850129

 

I don't think these voltages are seriously out, but the directions specify 3V, not 1.5 from the adapter. The one that's odd is VC2. This might be a side effect when the main input voltage is too low.

 

You can double check and see if changing the temperature compensation makes a difference. I recall it defaults to 1A which is 26C, a little warmer than we have here at the moment (a rare big freeze up). If your lab is in the garden shed or the basement/cellar and under 20 degrees C, than it probably is important to set this first.

I would absolutely NOT mess with the laser power at this stage. Maybe later but we need to follow the directions more carefully before making changes, I think.

 

The (well-known) noise is from seeking, and occurs during recording AND playback. It suggests (to my simple mind, as I don't really understand the subtleties of servo systems) perhaps current is being used to seek the head which subtracts from power at the critical moment of heating up the disk to write it. This would correspond to what you are seeing, that it seems to work and then degrades? I do not think that the time-lag you indicate is normal write-buffering.

 

I can get my 910 out later and check some of those measurements, but it won't be for at least another 10 hours.

[kgallen]

Hi,

 

I'm very impressed by the extensive commentary from @sfbp (I didn't know you had all this in you Stephen!) and the measurements and observations by @valentinc!!! I'm not sure if I can add much at this point. I will have to download the schematics and take a look at Stephen's question. I do agree that NiMH has a lower cell voltage than alkaline although without digging in I don't know what is the tolerable range of input voltages for the 910. It's going to have a boost converter in there to provide a higher initial voltage for further regulation and I would hope that would support down to 1.1V or better 1.0V. That the NiMH has a longer plateau voltage around 1.2V should be workable if the boost is as I hope.

 

Here are a couple of datasheets for NiMH and alkaline. They are a little generic but might be interesting:

 

https://data.energizer.com/pdfs/alkaline_appman.pdf

 

https://data.energizer.com/pdfs/nickelmetalhydride_appman.pdf

 

[kgallen]

Looking at the schematics you should be fine with an NiMH AA or an alkaline AA in the sidecar. All supplies end up at a couple of fancy power supply chips containing boost regulators. I can't find any proper datasheets for the chips but they look like they are there for handling cells of this type and for charging an NiMH gumstick (but not an NiMH in the sidecar).

 

Probably in service mode those voltages that are a little out of spec need tweaking - the power supply chips look like they have some programmability so you can probably adjust them in service mode (need to recheck the SM!).

[sfbp]

My strong impression is that with an NiMH in the sidecar and a gumstick (NiMH) in the body that the one inside ends up trying to charge the one outside. Is that wrong? 

 

And yes, subject to:

1. getting +3V power adapter or 1.5 *regulated* PS and

2. setting the temperature if your room is cold (probably only affects the LP later on)

 

then I agree it's probably fine to set all those voltages to spec. If that makes no difference, I would think it's time to do a self-serve calibration (all laid out in the book, basically you need a blank disk and that's about it - for HiMD it is more complicated because there are about 4 different modes to test for).

 

PS the fancy chips are there surely to compensate for the low voltage output by the NiMH? So, given that there's no fancy chip in the sidecar circuit, it probably means you should NOT put NiMH in a sidecar.

PPS (some of) the multiple voltages in Valentin's table are surely the cut-in/cut-out voltages for charging?

[kgallen]

Too many good questions @sfbp but I can’t answer them now other than agree that a higher voltage sidecar cell can couple easily to a lower voltage gumstick. From what I remember of the schematic there is just a low value resistor between them. 

[valentinc]

@sfbpThe measurements I have posted are with 1.5V at the sidecar terminals, not the DC in jack. It was easier to put some crocodile clips there.

The room temperature is 24-25°C, so don't think this would need any adjustment.

 

In regard to seeking noise, you're reffering to the spindle speed going up and down ? Is this normal ?

Haven't noticed it on other units (that actually work on recording) like N710 for example. I also don't notice it on playback. 

 

Went ahead and adjusted the voltages (the ones that were off) at 1.5V main supply just to see what happens. Wrote down the original values also. 

While initially I was able to record 2 discs fine, the situation came back to what it was initially after that.

Also noticed the MEM OVER error didn't actually disappear but it's rather intermittent.

What do MEM OVER and REC ERR actually mean ? Couldn't find any info about this online.

 

What I have noticed in reagard to voltage rails is, the VCO ones do drop quite significantly when main rail drops (for example 1.2V instead of 1.5V).

On the other hand, rest of them remain pretty stable even down to 1.1V.

Will test if giving it 3V to the DC in jack and re-adjust all the voltage to this. Will post the conclusions later on. 

 

As far as the gumstick + sidecar, indeed they are separated only by a small value resistor (20mΩ).

If both are used, one cell will charge the other until voltages become equal.

[sfbp]

There's a scratching noise when it seeks, on most of these 910s. If you don't have that noise, you are lucky. Nothing to do with spinning up/spinning down.

 

MEM OVER sounds like a buffering error. REC ERR is surely what it says. That one probably means you got to do the real alignment.

 

I'm pleased setting the basic voltages made a difference. Even if only temporary. Did you rule out mechanical stickyness? I think you did.

Sorry I haven't had time to get my 910 out and play with it - fighting fires (not literally) here, so far this week I have had 3 disk drives and a screen fail. Something to do with cold weather, maybe.

[sfbp]

On 2/10/2025 at 2:08 PM, valentinc said:

What do MEM OVER and REC ERR actually mean ? Couldn't find any info about this online.

Sony's manuals are very good. Occasionally, on the later models, they leave out something that was perfectly explained in the earlier ones. The only thing I can suggest is you look at the manual for this device's predecessors, as far back as you have to (and have time to do).

 

I would look in the 710 and the 510 (both NetMD) and perhaps the R900 and R90. Maybe the MZ-N1 (the first widely available NetMD portable) and the MZ-N10 (except that the latter uses Li-Ion, not NiMH).

Let us know how it progresses.

[BearBoy]

According to the User Manual for the MZ-N910:

 

 

[sfbp]

Didn't even think of looking there. Well done.

 

Makes me wonder if the vibration is self-generated.

[BearBoy]

Yeah, that’s what I was thinking.

[valentinc]

@BearBoy Thanks for pointing out, that gives some insight. When I asked the question of what MEM OVER and REC ERR mean,

I was wondering what is exactly triggering these errors (so I can narrow it down) but wasn't specific enough in my question. 

It's to be mentioned there is no external vibration and device is sitting flat on a desk while testing it.

So any potential vibration is certainly self-generated. However there is no vibration I can feel.

 

Still question is what's the unit actually detecting which trigger these errors ? Since from my understanding there is no feedback during recoring,

like the unit isn't reading what has been written in real time. Is this correct ?

 

@sfbp Updates I have so far are as follows: seems I missed one adjustment (which was written below charging voltages in the manual, so I didn't see it initially)

and another one was not stored in memory. After re-checking all the values and adjusting them to spec, there is a difference although not a fix.

 

Unit can, at times, record without giving the REC ERR, but more often than not it gets stuck in the ToC Edit phase even when it does record.

Also another aspect to mention is recordings done do have distorsion every 10-20seconds of play, indicating a high error rate.

 

To me this spindle speed constantly going up and down during recording seems abnormal and I think may be related to the problem ?

This seems to be different when the unit is actually recording, in the sense there's a higher duty cycle (more constant spinning compared to spin-up).

[sfbp]

Sorry to say, the builders are still here and there is sawdust everywhere, so no way am I opening up MD on my impossibly messy desk right now. I still think, from bitter experience, that 3V power supply (which is what the alignment adjustment recommends) is the simplest way of avoiding power deficits, both during alignment and for actual live recording.

 

The fact that you actually managed to record a couple of disks is very encouraging. The spin up/spin down is from what I recall, normal, but it maybe a sign of the laser being under-powered. I had something similar with my MDS-PC3 which required a very small bump in the read power. However, portables are not aligned using same methods, and the automatic adjustment sequence is the way to go.

 

Don't forget that when a disk is written, the next thing it needs to do is to read back what was written (hence getting stuck at TOC Edit). So it is just as likely to be the read power as the write power (which is a MUCH coarser adjustment, since writing required 10x the power) that is in deficit, likely because of a laser that got a bit tired.