mrdelurk - 13 December 2018 09:47 PM

This is what IC9 looks like on my XS8: photo

If I need to point the electrode into the brown circle, and not touch the chip leg or silver stripe, I can manage that.

As best I can tell from your photo, there appears to be significant corrosion. Solder mask over the copper foil should look uniform, and areas where there’s no mask should be copper or covered with solder.

https://en.wikipedia.org/wiki/Solder_mask

 

mrdelurk -

It’s starting to dawn on me, any hopes of me swapping out a bad JKAN-JA board or cable by itself are at present illusory due to the fact, that I simply can’t get the current 40-pin cable out. It won’t budge, even when I pull on the stiffener from both sides. So I ran a search on XS8” on ebay, looking perhaps someone sells a dropped XS8 for peanuts from which I could remove the JKAN-JA an DM boards still connected by the 40-pin cable as one.

Well, nope, but the XS8 search found something else. This awful 40-pin connector itself, it’s a “WC199000 FMN Connector”. I’m looking at the ebay picture and I still can’t tell how, if ever, this thing is supposed to come apart.

WC199000 is the Yamaha part number.

 

mrdelurk -

Further gumshoeing reduced the circle of likely manufacturing culprits to JST Manufacturing Co. They have a whole bunch of FMN connector PDFs, but not one tells how the dagnabbit thing opens.

I came to the same conclusion about the manufacturer, and found this:
https://www.jst.co.uk/productSeries.php?pid=44&cat=22

Non-ZIF LIF (Low Insertion Force) FFC connectors don’t open. LIF doesn’t mean that the force required for removal is low. I wouldn’t try too hard. Unless the photo of the board is somehow misleading, corrosion would lead me to suggest that you continue to use the XS8 via the DIGITAL OUT.

The granularity in the IC9 close up seems to be largely an artifact of the cell’s camera zoom, IMHO. Here is a photo of the whole JKAN-JA board with the same camera (photo). Even areas of 100% washed out white contain granularity, so I suspect it is a camera artifact.

But either way, you are right, I’ll use the XS8 via the SPDIF, that’s all there is to do right now (until a wrecked XS8 surfaces on eBay). I don’t think mLan will ever get a spin, even Apple abandoned FireWire for Lightning in its current Mac lineup.

mrdelurk - 14 December 2018 06:45 PM

The granularity in the IC9 close up seems to be largely an artifact of the cell’s camera zoom, IMHO. Here is a photo of the whole JKAN-JA board with the same camera (photo). Even areas of 100% washed out white contain granularity, so I suspect it is a camera artifact.

Thanks for the additional photo. It’s possible that artifacts and limited resolution might be compromising my ability to properly evaluate the condition of the board.

There are voltage readings at IC9 that could be made which, if found to be incorrect, would point to a definite failure on the JKAN-JA board. Pin 4 of IC9 (lowermost on the left as shown on the diagram I posted and your photo of that area) should be at -12 volts with respect to analog ground. Pin 8 of IC9 (uppermost on the right) should be at +12 volts. If you can safely contact those two pins with the meter probes at the chip pins or solder (black to pin 4, red to pin 8), you should see +24 volts between them. Not seeing that voltage would make a failure on the JKAN-JA board (rather than the DM board or the 40-pin cable connection) significantly more likely. If the voltage is correct, it still leaves other possible causes open.

If you don’t feel confident that such a measurement can be made without the chance of the probe tips slipping, please don’t attempt this.

Better to use the XS8 as it is than risk having things get worse than they are.

 

mrdelurk -

But either way, you are right, I’ll use the XS8 via the SPDIF, that’s all there is to do right now (until a wrecked XS8 surfaces on eBay). I don’t think mLan will ever get a spin, even Apple abandoned FireWire for Lightning in its current Mac lineup.

Apple had recognized the need for backward compatibility with FireWire, and there’s an adapter for converting a Thunderbolt port to FW. I don’t know if there’s anything available for Lightning ports.

Best of luck with your XS8.

I kept measuring IC9 Pin 4 and Pin 8 voltage for 10 minutes and getting a reading of 0. Then I remembered that I forgot to turn the XS8 on.

After turning the XS8 power on, with the multimeter dial set to 20V, I still got zilch ("-1") readings. Only when I lowered the multimeter dial to 2000 mV (aka 2V) sensitivity, I got a reading oscillating from -3 to -7 on IC9 pin 4 and -3 to -13 on IC9 pin 8. (On the blackish stubby chip legs) With both pin 4 and 8 probed simultaneously, the readout was -6

For a control test, I grabbed a working AA alkaline battery right after, and measured it with the multimeter. Yep, 1.58V reading at 20V sensitivity and 1586 mV at 2000mV sensitivity, the multimeter isn’t cuckoo. (Apparently.)

Remembering your remark how the XS8 lid should not be opened at 90 degrees, it hit me, that I had it at 90 degrees before during cleanings. I also see that a cable tiedown above the DM board has a popped out leg as a result to this day. (See red arrow and circle on this photo.) Could this lid opening have caused the XS8’s current symptoms?

Measuring between pins 4 and 8 of IC9, if things were operating correctly, should have resulted in a reading of 24 volts. With the voltmeter on a 20 volt range, that would give an over-range indication - the next highest meter range would have been required to take the measurement. However, none of the readings you got look normal.

If you can manage to make contact to pins of connector CN4 (the one with the white wires near the top right of your photo of the JKAN-JA board), you could determine if both +12V and -12V are getting to the board from the power supply. The pin numbering is identified on the board - pin 1 has an arrow pointing to it, and pin 1 and pin 6 are labeled. Pins 4 and 5 of CN4 are ground. Pin 3 is +12V, pin 6 is -12V. Sewing needles can be used as extensions if needed to get into tight spots that the meter probes won’t fit into. Be careful to avoid shorts.

If those voltages are correct, then based on what you found at IC9 there’s definitely a problem on JKAN-JA.

I see, user error, sorry about it. I repeated the measurement on IC9 with a 200V multimeter sensitivity setting. Now Pin 4 read 11V, Pin 8 read 12V, both together read 23V.

CN4: seemed like I need to unplug it to reach the pins with the probe, so I powered the XS8 off, unplugged CN4, powered the XS8 back on, and measured CN4 with the probes. Pins 3 and 6 on the footing read 0V at 200V sensitivity. Pins 3 and 6 on the incoming cable oscillated a lot before settling on 12V each at 200V sensitivity. Together they measured 24V.

Powered off XS8, replugged CN4 really well, powered XS8 back on, measured IC9 again: still only 23V for both pins. Those crafty ants inside must be stealing one Volt.

mrdelurk - 16 December 2018 02:52 AM

I see, user error, sorry about it. I repeated the measurement on IC9 with a 200V multimeter sensitivity setting. Now Pin 4 read 11V, Pin 8 read 12V, both together read 23V.

CN4: seemed like I need to unplug it to reach the pins with the probe, so I powered the XS8 off, unplugged CN4, powered the XS8 back on, and measured CN4 with the probes. Pins 3 and 6 on the footing read 0V at 200V sensitivity. Pins 3 and 6 on the incoming cable oscillated a lot before settling on 12V each at 200V sensitivity. Together they measured 24V.

Powered off XS8, replugged CN4 really well, powered XS8 back on, measured IC9 again: still only 23V for both pins. Those crafty ants inside must be stealing one Volt.

While what should be -12V at IC9 pin 4 is reading somewhat low, that normally wouldn’t be sufficient to prevent the op amp stages from functioning.

Another possibility is that the output muting circuit isn’t turning off (unmuting) properly. The voltage which controls that can be measured at transistor TR9. TR9 is located across the board from JK4 (L/MONO output jack) - that is, nearer the opposite edge of the board from that jack. The voltage of interest will be found on the tab of TR9, opposite the side that has the three pins. (See attached part of your photo - the tab is near the added white dot.) The voltage should be measured with respect to ground. When powering on the XS, under normal conditions the voltage will change from one of positive polarity initially to negative as the XS boots. Start with the XS turned off, then turn it on, and let us know what voltages and polarity you find.

Click thumbnail to see full-size image

I repeated the test at first 2x, with consistent results. At 200V multimeter sensitivity, the meaured voltage on TR9 went from 0V to -4V at power on and stayed at -4V without any fluctuation until XS8 power off.

Subsequently, I did 2 other takes, with multimeter sensitivity at 20V, just in case. This time the reading consistently went from 0V to -4.25V at power on and stayed at -4.25V without fluctuation until XS8 power off.

Maybe it could be noteworthy that with CN4 and IC9 this multimeter did not display one -nV and one +nV reading, just two readings of +nV (or -nV, depending on whether the red or the black probe was plugged into COM port).

mrdelurk - 17 December 2018 03:33 PM

I repeated the test at first 2x, with consistent results. At 200V multimeter sensitivity, the meaured voltage on TR9 went from 0V to -4V at power on and stayed at -4V without any fluctuation until XS8 power off.

Subsequently, I did 2 other takes, with multimeter sensitivity at 20V, just in case. This time the reading consistently went from 0V to -4.25V at power on and stayed at -4.25V without fluctuation until XS8 power off.

That seems OK. The negative voltage is what turns off the muting transistors, which is required for there to be output.

 

mrdelurk -

Maybe it could be noteworthy that with CN4 and IC9 this multimeter did not display one -nV and one +nV reading, just two readings of +nV (or -nV, depending on whether the red or the black probe was plugged into COM port).

Typical usage is to leave the black probe connected to the meter’s COMMON. When a measurement is made with respect to ground, the black probe is connected there. Then the red probe is connected, and with most DMMs if the voltage is negative with respect to COM, a minus sign will appear on the display. I’ve been somewhat “reading between the lines” with your measurements. So far, no obvious problems have shown up.

Here are a couple of more things to check, which will check voltage supply to the DAC chips. To be done after the XS boots…

1) Check the voltage at capacitor C56, which will verify power to the digital part of the DACs. It’s near IC9 - measure the point marked (white dot) on the first attachment with respect to ground.

2) Check In and Out voltages of the regulator IC11 - the regulated output voltage is the source of power for the DACs analog sections. IC11 is across the board from JK8 Foot Controller 1 jack.  See the second attachment. Connect meter common black probe on IC11’s tab (black dot), red probe on either of the pins marked with a white dot. Then move the red probe and check at the other white marked pin.

Let us know what you find.

Click thumbnail to see full-size image

Hello 5pinDIN,

With the XS8 running, C56 voltage was 3.3V
IC11 measured 12.2V on top and 5.0V on bottom (top being towards JK8)
Thank you for the detailed measuring instructions, I learned a lot.

mrdelurk - 17 December 2018 11:41 PM

Hello 5pinDIN,

With the XS8 running, C56 voltage was 3.3V
IC11 measured 12.2V on top and 5.0V on bottom (top being towards JK8)

Those voltages are correct. I’ll have to see if there are any other things that can readily be checked with a voltmeter. That will probably be tomorrow - my time, anyway.

 

mrdelurk -

Thank you for the detailed measuring instructions, I learned a lot.

You’re welcome. I only wish that one of these tests would have pointed to a fault. Maybe that will happen yet.

I’ve looked over the service information again, and unfortunately I don’t see anyplace else where a DC voltage reading will be particularly helpful in pinning down the problem. At this point, if I were servicing the XS8 I’d be using an oscilloscope to check digital signals, etc.

You could try running the built-in diagnostics, but I suspect they won’t reveal much we don’t already know.

Would it help anything if I bought one of these cheapo oscilloscopes to measure things with it?

mrdelurk - 19 December 2018 08:16 AM

Would it help anything if I bought one of these cheapo oscilloscopes to measure things with it?

I must say, I admire your perseverance. I’ve looked at the linked device, and it appears that the bandwidth would be insufficient for the purpose.

More to the point, I can’t easily describe what the waveforms should look like, and capturing them from a ‘scope and then posting them for me to analyze is probably beyond what could be reasonably done via this forum.

Perhaps it is worth running some of the built-in diagnostic tests. They’re detailed in the Service Manual I linked to earlier. The Test Mode should be begun as per
B. Test Entry, MANUAL MODE, on page 112 of the SM.

Test T01: EBUS ID (page 115) is the most likely to be helpful, in particular if any of the IDs are found to be NG (no good).

Tests T14: OUTPUT-L, and T15: OUTPUT-R, although designed to be used with test equipment, would provide audible output if working.

Tests T16: AD-DA (MIC) and T17: AD-DA (LINE) specify using an audio generator as the input source, but an actual mic and other audio line source can be used for a quick check.

The remaining tests aren’t really relevant to the problem, and they can be skipped over. In some cases they require special gear, and otherwise can indicate NG when nothing is actually wrong.

If you run these tests, let us know if the results reveal anything unexpected.

As I have stated in previous posts:
If any of you are not having problems with your Motif, I suggest that you do not run the Test Mode diagnostics. Please don’t go looking for trouble, it can find you without your help.

Well, this time we did get a nonstandard result for sure. To enter into test mode, I held the 3 buttons that the manual said. Promptly got the first screen, with TEST MODE written inside red circles. Then I lifted my fingers as the manual said, and got a second splash screen - and that was it. The test screen never loaded.

I repeated the procedure 5x with the same consistent result. The 4th time I waited for the test screen for 15 minutes… nada. The fifth time, I left the XS8 on, and came out to prepare this message. By now 40 minutes elapsed, let me go back and look… yep, no test screen, it’s still the second splash screen.