| previous by date | index | next by date |
| previous in topic | topic list | next in topic |
On Mon, May 5, 2014 at 9:01 PM, Bob Grieb <bobgrieb@yahoo.com> wrote:Hi Jeremy,
I'm a little bit of a "scope snob", so maybe not the best one to ask. Certainly the
Hantek device would be a lot better than nothing. I would suggest troubleshooting
instead of just replacing a chip and hoping. There are lots of things that could cause
the MCU to not do the right thing. If you buy some sort of scope, it will be useful for
other things as well.
Higher bandwidth would be better, but having storage is nice. The general rule is that
for looking at a square wave, to get a good display you need bandwidth that is ten times the
frequency. So you may be OK with the signals in the PolySix, since the CPU divides the
clk by 15 or so for the instruction cycle.
How much are the replacement CPU's for the PolySix on ebay? I haven't noticed them.
You would need a chip that has the Korg code in it, of course, not just any 8049 chip.
Bob
Sent: Monday, May 5, 2014 6:01 PM
Subject: Re: [PolySix] I killed my PolySix, help!
Well, I've been looking into a scope, but really the only thing within my budget would be something like this:I know you recommended a scope with at least 60MHz bandwidth; would this one not get the job done? My only other idea at this point is replacing the MCU chip on the off chance that it went bad, replacements are really cheap on ebay...On Fri, May 2, 2014 at 2:13 PM, Bob Grieb <bobgrieb@yahoo.com> wrote:It would certainly make things a lot easier. Troubleshooting a processor-based system with a DMM is pretty tough.
Hard to say if 4.7 instead of 4.82 means you are getting reset pulses. It's possible.
A low AC reading on the DC supplies is good. Normally the ripple at the input to
the regulator would be removed by the regulator. The exception is when the filter cap
has lost much of it's value due to age and the voltage at the input to the regulator is dropping below the minimum value that it needs. A 7805 5V regulator needs something like 6 to 6.5V to put out 5V. (Depends on current, temp, etc) Anyway, having a scope
available would be a big help for troubleshooting. Even an old boat-anchor type with at least 60 MHz bandwidth would be a lot better than nothing. Digital storage is very useful in some situations, but adds cost.
Bob
Sent: Friday, May 2, 2014 1:22 PM
Subject: Re: [PolySix] I killed my PolySix, help!
I'm getting 4.7V on pin 4 of IC22. Not sure how to measure the ripple with the DMM, I have it in AC mode, but I just get a low reading (2v or so) which quickly goes down to zero. Looks like now might be a good time to invest in a scope...On Thu, May 1, 2014 at 11:07 PM, Bob Grieb <bobgrieb@yahoo.com> wrote:OK. Seems like there may be an issue with the 5V supply.
Possibly a bad filter cap in the power supply, or maybe something
that is drawing lots of extra current. You can try putting your meter
in AC mode and measuring the supplies to see how much ripple
is one them. I don't think the CPU on the KLM367A bd is running.
Reset to that chip is pin 4. If the 5V is too low, or is pulsing low due
to excessive ripple, it may be resetting that CPU over and over.
I would see what you measure with the DMM on pin 4 of IC22, being
careful not to short to any adjacent pins. The reset circuit is looking
at the 5V supply, so if that is too low, the CPU will be held in reset.
Bob
From: Jeremy Pisani <jeremy.pisani@gmail.com>
To: polysix <PolySix@yahoogroups.com>
Sent: Thursday, May 1, 2014 9:04 PM
Subject: Re: [PolySix] I killed my PolySix, help!
Thanks again for your help. Here's the values I got after doing the power supply calibration:+15V: 15V-15V: -15.06V+5V: 4.82V-5V: -5.01VVR2 is all the way clockwise, I can reduce the voltage, but can't get more than 4.82V out of it.Tried the D/A calibration procedure in the service manual, but the LEDs stayed the same.On Thu, May 1, 2014 at 6:20 PM, Bob Grieb <bobgrieb@yahoo.com> wrote:Sometimes a DC voltage will read low because it has a lot of ripple
on it. If you have a bad filter cap, it could cause a volt of ripple
on the 5V supply, which would cause it to read low. If the -5V supply
is designed to match the +5V supply, then it will drop too.
A scope would make checking for ripple very easy.
Bob
From: Jeremy Pisani <jeremy.pisani@gmail.com>Subject: Re: [PolySix] I killed my PolySix, help!
To: PolySix@yahoogroups.com
Sent: Thursday, May 1, 2014 3:47 PM
Hey Bob,Thanks for the reply! Haven't had a chance to look at the PolySix since I sent this, but hopefully I'll be able to get back to work on it in a day or two.I haven't attempted to get into the DAC calibration mode; if I remember correctly, the service manual asks you to connect a 56k dummy resistor to the 2 prong connector on the power supply, and I don't have any test clips for the time being...will give it a shot once I pick some up.To answer a couple of questions you had: I don't have an oscilloscope, unfortunately, but I'm definitely going to pick one up, as I have another synth here that I'd like to calibrate as well (Yamaha CS-5). The multimeter I have is good quality and practically new, so I'm pretty convinced the readings are accurate. Any idea why the voltages would be so low?On Mon, Apr 28, 2014 at 10:21 PM, Bob Grieb <bobgrieb@yahoo.com> wrote:The LEDs that you say are working are controlled by the key assigner MCU.
The other LEDs are controlled by the MCU on the KLM367 bd. Can you get into
the DAC calibration mode by flipping the switch on the KLM367? At least that would show that the MCU is executing code. (I'm not suggesting that you calibrate the DAC, just see if the CPU will enter that mode) Do you have an oscilloscope? The 4.2V value seems low. How much confidence do you have in your DMM?
Many of the chips you are probing are analog, and have power on different pins. Please
just focus on the digital IC's. Maybe you need to download the service manual and take a look at the schematics? Usually the problems with the KLM367 have to do with IC31, 30, and connections to the front panel on CN06. If you have a scope, what is the waveform like on pin 32 of IC22? (Be careful not to short to an adjacent pin when probing)
Good luck,
Bob
From: "jeremy.pisani@gmail.com" <jeremy.pisani@gmail.com>
To: PolySix@yahoogroups.com
Sent: Monday, April 28, 2014 6:56 PM
Subject: [PolySix] I killed my PolySix, help!
So, I did a stupid thing. I got my PolySix a few years back, and it had the leaky battery issue. I sent the board off to be fixed by Old Crow's synth shop, and after that it was working great, except for the buttons being finicky, and the Hold button didn't work at all. After a lot of procrastinating, I finally ordered some new switches and went to work desoldering & replacing the buttons.At first I checked to make sure the synth was still functioning correctly after I replaced each button, but after 5 or so I felt confident enough to keep going without checking (I'm kicking myself as I write this). When all was said & done I powered it up and it's behaving like it did when I first got it. When powered on, the Write & Manual buttons light up, as well as the Bank C, Poly, and Program 3 & 5-8 buttons. I can switch between Poly/Unison/Chord Memory & Hold buttons with no problem, but it doesn't affect the sound. None of the knobs seem to affect anything either.I did a lot of the desoldering without putting something over the circuit boards to protect them from solder flecks, and I'm guessing maybe something got fried when I powered it on, but honestly, I don't know. I did a little poking around with a multimeter & got the following results:Power supply voltage:4.2/-4.214.6/-14.6All ICs on the KLM-376A are reading 4.8v except for the following:IC14: 3.5vIC20: 3.6vIC21: 4.1vIC25, IC32: 4.1vIC34: 4.6vContinuity between the previously repaired traces seems to be fine. Any ideas?