Casiotone CT 6500 - Turns On , No Sound

[Aaron Wright]

I decided to go ahead and order some replacements for the three big caps on the power board, the three TL082s, the 4051, the 4053, and the NJM13600 just to be safe. Hopefully some progress will be made next week after installing those.

 

I'll probably finally reassemble this thing and get it off of my dining room table tomorrow, so if there's anything anyone would like me to try before I put it away to wait for the new parts, I'm all ears.

 

As a side note, I'm kind of curious to know what anyone else's waveforms look like on the 5V and 15V rails (or on the output with no keys being pressed) on a similar keyboard, since after doing some more reading up on smoothing capacitors, I realize I'd expect at least a similar shape to what I was getting (but on a lesser scale) during normal operation due to how smoothing capacitors work on fully rectified AC. It's possible that the hum isn't even an issue for me anymore, since it's barely audible through either the speakers or the headphones now. Regardless, new capacitors are on their way.

[IanB]

Well, there's really no harm in swapping those power caps after all these years, they can't be in great condition even if they aren't the problem. (By the way, there's a bit of a mania on the internet for "recapping" everything, which I don't agree with). There really shouldn't be any significant waveforms on the power rails, they're meant to be DC! The less ripple the better.

 

For now, can't think of anything else to suggest, keep us updated Aaron.

[Aaron Wright]

The big power supply capacitors, the 4051, and the 4053 came in today, but as of installing and testing them, no progress has been made.

 

The power supply still produces a similar (perhaps now even sharper) waveform, here:

 

 

And there is still no sound coming from the output of the TL082 directly above the 4053, although I actually do see some scratchy remnants of the waveform when I play keys. However, it's inaudible over the buzz.

 

There's one strange thing I noticed when testing the input of the TL082 (pin 5 of the 4053), which is that if I strike the same key several times in succession, the key will only register for about a one second period, then nothing will register for another second, and so on. Strangely, this doesn't apply to holding a single key down (the key just continues to play, but only if I strike it during one of its "capturing intervals"). This only happens at pin 5 (and pin 12, on the opposite side) of the 4053, and this happened before I replaced it too. Those two pins are only connected to the TL082 above it, so I'm not sure what's causing the regular drop in keys, or this behavior in general. When listening on the other pins, I can't find anything that correlates with this drop in new note sounds. Furthermore, if I'm holding down a key, I still can only press additional keys during the capturing interval, otherwise they will not play through the 4053's output, even though it is audible on its input. All these "dropped keys" somehow only apply to the output of the 4053, since they are still playing on the input side.

 

I hope I explained that well enough, since it's pretty confusing.

 

I also noticed that there's a little 'blip' in the voltage across the main board whenever I strike a key, even if no sound is produced.

 

And another thing! When I play two keys at the same time and listen to the sound on the 4053's input, sometimes the keys will play harmoniously like they should, other times the combination sounds noisy and almost overdriven, and this seems to correspond to the drops in the output, so I guess that's the only thing that correlates with the drops that I've found so far. Seems like a timing issue somewhere. I'm wondering if maybe the choppy voltage is periodically syncing up with something logic-related on a clock pulse and causing something to go wrong. Could the rectifiers be at fault?

 

I decided to test the two rectifiers' waveforms. Here are my results:

 

5V Rectifier ( + )

 

24V Rectifier ( + )

 

24V Rectifier ( - )

 

24V Rectifier (between + and - )

 

I'm not sure what to make of this, whether this is normal or abnormal, but it doesn't seem quite right to me. I should note, the 60Hz AC coming into the rectifiers is a lovely perfect square wave.

 

In the next week or so, the 13600 and the replacement TL082s will get here, and hopefully I'll be able to get some sound out of the speakers from the drums, at least. For now, the vexing continues. As usual, any help or insight is appreciated. If anyone with a similar keyboard wants to prod their rectifiers (or at least come up with a less funny way to say that), that would at least provide a clear basis of comparison. For now, I'm returning to the service manual to try out its waveform tests.

Edit: I also may very well reflow all the solder joints on the power board just to be super extra sure none of them are faulty.

Edited January 6, 2022 by Aaron Wright
added a little more info

[IanB]

I'm thinking we're getting lost in the weeds here maybe. Something I forgot when looking at the scope traces is that they aren't a real oscilloscope but being read by a device's audio input, so high voltages will not pass through properly, producing a clipped "square" result and it's hard to know what the scale is anyway. I'd just check there's about the right DC voltage on the power rails with a meter and move on (if your meter can properly filter out DC when set to an AC range, that might show you the approximate amplitude of any ripple, but unless that's huge the thing should work).

 

The one second on off of those pins on the 4053 makes no sense to me at all and I am pondering it. It's way too slow for the function of the chip, which is to sample and hold the DAC outputs (so it's meant to be operating at the sample rate).

 

I was just about to post this when it occurred to me that possibly the power supply is resetting itself due to detecting a fault, Casio had a habit of designing comprehensive power supplies in that regard, so the 1 second might be "power up, restart, power up, restart"... That's just a guess.

 

edit I think  in retrospect that last thing was total cobblers. It has occurred to me that this weird on off behaviour might possibly have something to do with the keyboard, can you confirm whether you get a proper output from the MIDI out, if you've got a MIDI interface run it into your PC and check it with MIDI-OX. You could also try sending the CT6500 some MIDI data and see what you get in the audio section of the PCB.

 

This is all really hard without a proper oscilloscope to look at what's going on digital signal wise.

Edited January 7, 2022 by IanB
Reconsidered

[Aaron Wright]

I tried hooking it up with MIDI, and it works perfectly. No dropped keys or anything. I think you were on to something, however. If I have a device plugged into the midi out port, and I play a drum sequence, the voltage suddenly gets distorted across the board, producing a loud buzz. I mentioned in an earlier post how there would be a blip in the voltage whenever a key was played, and I think it may be connected to midi output. If I disconnect the amplifier board's power entirely, the distortion no longer happens, and the blips in voltage no longer occur when I press a key. Strangely, the buzzing only happens when I have a midi cable plugged into the output. the cable doesn't even have to be plugged into anything at the other end.

 

Also, the voltage has always been reading correctly on my multimeter, so no problems there.

 

I had the idea that maybe the board was resetting itself as well, but thought that couldn't be true since held keys continue to play.

 

HOWEVER, I tried holding a key down while playing drums and listening to the drums' output, and when holding down any keys, the drum beats would start over abruptly when halfway through their cycle, not corresponding with the timing of the cyclical detection of keys in conjunction with the 4053 as mentioned before, but apparently the phenomenon are connected. Very strange.

 

I also replaced the 13600 today, but all behavior remained the same. Drum sounds are making it to the 13600, but not out of it. I wonder if perhaps this is related to the volume control circuit on the amp board, but I have no ideas for testing this yet.

 

Also, a fun thing I found out: if you disconnect the power cable from the amplifier board, the ground of the board is somehow connected to the drum sounds, so I can hold the positive end of my "oscilloscope" probe and touch the heatsink for the big amp and I'll start hearing the drums on my computer.

 

Anyway, a recap of what's known:

 

POWER BOARD (M5157-PS1 & M5157-PS2):

• 120hz buzz from rectified voltage, audible when probing the board and listening to the waveform, and making waveform analysis difficult (might not be an actual problem since all the voltages are still correct on a multimeter in the end).

 

MAIN BOARD (M5157-MAIM):

• intermittent sound past the 4053 (which has been replaced and the issue persists), in which new key sounds only register within one-second windows which are spaced one second apart (approximately).
  • held keys will continue to play their sounds as long as I start holding them during a so-called capturing interval.
  • if any keys are held, drum beats will reset halfway through their cycle, independent of the capturing interval.
  • the input to the 4053 has all the missing sounds.
• no sound at all out the TL082 op amp directly above the 4053.

AMP BOARD (M5157-ASIM):

• no sound out from the 13600 op amp, even after replacing it.
• when power is supplied to the amp board, any keys pressed will cause a blip in overall board voltage.
• when any midi cable is plugged in to MIDI OUT and a drum sequence is played, the constant midi messages cause repeated blips in the board voltage, adding a loud buzz to the sound of all points with positive voltage on any of the boards.

Yes, a real oscilloscope would sure come in handy right about now. As would a service manual for the CT-6500 in particular, since there are some differences in the main board and the amp board between the CZ-5000 service manual and the CT-6500 (such as a completely different headphone amp in the CZ-5000 that is not present in the CT-6500, and the layout of the entire main board being different), but I might be a few decades too late to find one of those.

 

For now, just waiting for the new TL082s to arrive, though I think the best those could do is finally let some sound out of the main board.

 

If anyone wants any high quality photos of parts of the board, or video or audio recordings of the sounds it makes in various places, I'd be happy to provide them.

 

------------

 

As I was about to sign off on this post, I decided to do some poking around on the outputs of the D933 chip in the bottom left, and found some blips that matched up with the keys being detected that make it through the 4053. I've traced them back through several chips, and I think I'm going to have to draw a diagram to keep track of them all. I have not yet been able to determine the source of the blips, but I have some theories.

 

While I was tracing those, I also discovered that several lines on the right side of the board make a different sort of buzz from the normal DC voltage on the board, and they have a blip that sounds like a loss of voltage once every second or so. To differentiate, I'll call the first sorts of blips (which include key presses, but only the ones in the capture interval) as "key blips" (a very technical term) and the other kind as "power blips" (for lack of a more descriptive term).

 

The key blips seem to originate in the 933 chips in the bottom left, specifically, pin 3 on both of them produces the same blips for each key pressed, cutting off for a second, then resuming detection. Without any real documentation on the UPD933AC, only the low-resolution labels on the service manual scans, it's hard to say what that's doing. I had to reference the CZ-1 manual for these, since that's the only one with the same chips. Pin 1 has all the blips for all keys pressed at all times, which is what I would expect from pin 3. The fact that pin 3 on both 933s aren't connected suggests to me that neither of them are broken, that there's something with the input that's affecting the output, but I'm not sure what it is.

 

Perhaps more importantly, though, the reset pins on the 933s seem to be being triggered once a second, or at least the signal is brought low on those pins once a second. If that were true, that might explain the issues. That reset pin is connected to pin 44 of the leftmost uPD7811G, which is the RD (read strobe) pin, which the datasheet indicates goes low "to gate data from the external devices onto the data bus." It's also connected to pin 12 of a 74LS08 to the right, a quadruple two-input AND gate, though that's an input, so it seems that the uPD7811G is the source. However, it seems like this is expected behavior, otherwise why would the read strobe pin be connected to the 933s' reset pins? I was probably searching in the wrong direction.

 

I have not been able to trace the power blips to anything concrete, the whole right side of the board seems to be plagued with them.

 

I wonder if perhaps one of the crystal oscillators could have been damaged in whatever impact broke the volume panel board.

 

Anyway, now I've stayed up until almost 6 AM trying to find a definitive source for either the 1Hz blips or the dropped inputs, to no avail. Sorry for the info dump, it looks like I might have to become an expert on these components if I want to get this thing working right. Tomorrow I'll try to spend some valuable not-middle-of-the-night hours doing more investigative work. Until next time!

[pianokeyjoe]

A question? How did you "fix" the broken volume board? And again, this sounds redundant but alot of your issues really DO sound like a bad PSU AND AMP board. Sadly my 2 units that ARE parts units are too far buried into my future studio to get them out and follow along with you. My GOOD unit how ever is within arms reach so I may pull that one out and take a gander at what the PSU outputs are supposed to me and amp board output. I know this is not recommended but in the old days I used to touch the amp section of the DC powered Casio keyboards and Yamaha portasound keyboards and could hear potential microphone input points buzzing like you hear when messing with an electric guitar plugged into an amp. Goodness. I hope you got a steal of a deal on this unit?

[Aaron Wright]

19 hours ago, pianokeyjoe said:

A question? How did you "fix" the broken volume board? And again, this sounds redundant but alot of your issues really DO sound like a bad PSU AND AMP board. Sadly my 2 units that ARE parts units are too far buried into my future studio to get them out and follow along with you. My GOOD unit how ever is within arms reach so I may pull that one out and take a gander at what the PSU outputs are supposed to me and amp board output. I know this is not recommended but in the old days I used to touch the amp section of the DC powered Casio keyboards and Yamaha portasound keyboards and could hear potential microphone input points buzzing like you hear when messing with an electric guitar plugged into an amp. Goodness. I hope you got a steal of a deal on this unit?

I fixed the volume board by soldering some wires in to connect every solder joint that had cracked traces between them. I confirmed my work with continuity tests and by making sure the resistance values were changing between the inputs/outputs in accordance with moving the potentiometers around. You might be right about those boards, but at least everything in them is replaceable! I can assure you, without naming a price, that the deal I got makes it very hard to complain, even if getting sound out of it might end up being a lost cause.

 

The other day I accidentally knocked one of the soldered wires (from the master volume input to the corresponding point on the potentiometer) off by rubbing it against part of the other half of the case, but unless I'm mistaken, that doesn't seem to be an issue when the sound isn't even making it through the little op amp on the amplifier board yet, and I didn't feel like setting up my soldering equipment again so I'm leaving it alone for now. It seems that the master volume slider adjusts the volume *after* it is put through the final amplifier (or possibly just before that)

 

If I truly can't get anything fixed, I'll probably look into a more modern power supply solution and lifting audio directly from the main board and amplifying it myself, though I'd lose the stereo chorus and other effects from the secondary board in that case. That's the worst-case scenario though.

 

Edit: I just reached my post limit for the day, so I'll have to report ground test results tomorrow @IanB, and I'll finally take the time to test midi input. I got sidetracked yesterday by the buzzing on the voltage when midi out was plugged in.

Edited January 9, 2022 by Aaron Wright
added more info to circumvent daily post limit, lol

[IanB]

This is in one sense curiouser and curiouser, on the other hand the range of problems suggests a generalised problem rather than some specific component failure.

 

I take it when you refer to the MIDI working fine, that is the MIDI output rather than input. However if that is the case we have some data; the keyboard is working fine. That means the main CPU (7811) is fine and its crystal is fine. I'm beginning to suspect a ground problem. With the whole thing powered down, and using a continuity test on your meter, check that all the chips have a connection to ground/zero volts.

 

Also, the CPU crystal is working and if you're getting anything out of the 933s, their crystal must be okay too. And crystals are unlikely to be broken by an impact anyway.

Edited January 8, 2022 by IanB

[Aaron Wright]

I tested the continuity, and every single chip in the keyboard is properly grounded, though the case ground seems to be a "lifted ground" as every ground point I tested has a resistance of exactly 101.7 ohms. I also forgot to mention that both MIDI input and output work, with the caveats mentioned of strange feedback to the board voltage, but otherwise perfect functionality.

 

I tried replacing the TL082 where the output was stopping, but it did nothing. I did manage to lift off some pads when desoldering though, that was (not) fun.

 

Very interestingly, I found that if I disconnect the large black and yellow ribbon cable between the main board and the stereo chorus board (edit: not actually the case!), the TL082 actually does have sound coming out of it, but for some reason having that cable plugged in kills the sound completely. The dropped keys issue is still present, of course. If I disconnected the cable in the top left that carries the audio output from the TL082 to the stereo chorus board, the sound quality from the TL082 improves dramatically. Something must be going wrong on the stereo chorus board, but I'm not sure what.

 

Here's what's really exciting: I discovered that the outputs from the master and sub CPUs are alternating on each 1Hz cycle. The two sides of the TL082 carry sound from the two CPUs, and I just didn't notice until I got really deep into the CZ-1 service manual. I think there might not actually be a problem! I think the two different CPUs are probably supposed to be combined in the stereo chorus board, which seems to be the point of failure. It figures that that's the one board I haven't unscrewed, examined (apart from testing the ground of all the chips and looking for loose parts), or otherwise messed with, so that figures. Also of note, when the plug carrying audio to the stereo board is disconnected, the audio going in is both clear and free of buzzing, so I think the buzzing voltage from the CPU and the 1Hz ticking are no longer problems.

 

 

 

Right after writing that and concluding that the stereo chorus board might be the point of failure, I did some testing in different configurations with cables connected or disconnected, and found that the only thing that actually triggers the loss of sound is connecting the power cable for the amplifier board. This leads me to conclude that either the power supply or the amplifier board is the point of failure. I suspect something in the amp board, maybe the STK4332 itself? The great news is, I'm actually getting sound out of it with the amp power disconnected! Line out works like a charm. Kind of ironic how simple it was to get sound out of this thing in the end. There is still an audible buzzing over the line out, though, so perhaps there is something minor going on in the stereo chorus board after all, or it could just be more amplifier antics.

 

Now, I have some qualms remaining about the keyboard that I might need someone with a working similar keyboard to help me out with. Most of the time, if I press more than one key at the same time, it sounds rather discordant and overdriven, even though it doesn't actually appear to be clipping anywhere in the waveform. I feel like this isn't normal behavior, and I should be able to play some dang chords.

 

Secondly, if I'm holding any keys down, any drums playing will suddenly seem to have a fill-in in the middle of its cycle and will restart, which might be intentional behavior, but I don't have access to those pages of the owner's manual to find out. If I set it to "synchro" and trigger it with a key press, I can continue holding down that key and the drums will continue their cycle as usual, it's only when I hold down another key after that point that they will start to do their weird thing and reset. Similarly, if I hold down some keys and then press start, the drums will play uninterrupted.

 

If either of those are indeed abnormal behavior, they sound like a pain to diagnose, but I pretty much got this thing entirely for the sound, so I may as well do my best to improve what I can.

 

I'm pretty ecstatic that this thing isn't a lost cause for sound output, and the real issue seems to be just around the corner. Hopefully soon everything will be in full working order.

Edited January 9, 2022 by Aaron Wright
fixed typo

[IanB]

The drums thing I can answer; mine does that as well. I thought it was faulty too, but it actually depends on how you play and seems to be a "feature" to make the drums more "Interesting". Total PITA frankly.

 

DIscordant and overdriven? That's not right. I take it you've tried turning the various volumes down to see if that affects it? It might be a problem resulting from that broken controls board that is causing the signal level to be wrong?

[pianokeyjoe]

AMEN!! You got it! Ok.. sort of. As I told you before, AMP board and PSU are culprits here. The Stereo Chorus is NOT really Stereo but it is what makes the Keyboard SOUND like it is stereo. The AMP is mono. Ok, you need to fix the amp board. The Chorus is taking in the mono signal and giving you the alternating pseudo stereo output. Remember that you DO have a chorus on/off switch.. try that too, besides the unplugging of cables trick. You may slap your head at some point in disbelief.

The dropping keys? What do you mean by that? You mean some keys are dead or quirky? That may be the rubber contacts, clean them. The "FUZZING sound when you play more than one note at a time IS normal when the FILTERS inside(trim pots) are adjusted just so. You will need to adjust the filter pot, check your manual for the CZ1. It is the same. IF the fuzzing is really LOUD, that is you amp board acting up like my Parts unit. That loud fuzzing is brought down in volume and filtered down to almost clean by a filter circuit somewhere between the Sound gen output and AMP input(general input of the amp or audio board, not the POWER AMP that drives the Speakers but the Line level section that drives EVERY audio out).

 

Like lanB said, the rhythm section has a feature that when you play keys in a certain sequence the rhythm tries to auto fill in order to give you the illusion of a live drummer kinda improvising along with your playing. That feature only existed in the CPS201/CPS210(uber rare midi version),CT6000,CT5500,CT6500, and one more 73 note version I saw in Argentina that I do not know the model number for but is like the CT5500 but longer keybed. Later versions of Casio keyboards dropped that feature and older versions were more organ like(MT65/CT405).

 

If the Fuzz is at the same volume level as the drums or accomp section, then you are dealing with the filter trim pot adjustment issue. I had that issue with the CZ5000. Also I found out on a CZ230S that the fuzz came in before it totally died. Like when your batteries are dying kinda of effect. For my CZ230S it was the battery acid damage all over the main board.

 

Check for that trim pot as it is a known issue. Google Casio CZ5000 fuzzy sound.

[IanB]

@pianokeyjoeThe problem with the "clever" drum feature is you can't turn the dang thing OFF!

[IanB]

At the risk of self promotion, I just uploaded a YouTube video which features the sound, if not the image, of my CT6500, might be useful for audio reference. The sound being used it the default patch with Chorus on, "Sustain/Reverb" on and some modulation.

 

[pianokeyjoe]

@IanBI just found and subbed to your youtube channel. Real glad you posted this video mate! Thankyou!

[Aaron Wright]

@pianokeyjoe the chorus feature definitely does work, so I don't think anything is amiss there. I am aware that the sound out of the main board is mono and is only given a fake stereo effect by the chorus board, I just didn't realize that the op amps were carrying separate audio from the main and sub CPUs all the way to the stereo chorus board. The "dropping keys" turned out not to be an issue as I later wrote in my last post, since I was just listening to the output of one CPU at a time, not realizing they were processing keys and sounds on an alternating 1Hz cycle. So, from the output, it sounded like keys were only being detected on every other second, but in reality, the undetected keys' sound was coming out of the other output of the board, which I didn't even realize was a possibility until I got to the part of the CZ-1 manual documenting that part of the audio generation.

 

I've been fiddling with the DAC offset voltage as per the service manual (turning the VR2 potentiometer by the BA9221 and TL082-3), but can't get it to affect anything. It's odd- it does not affect the voltage between pins 13 or 19 of the 9221 and pin 7 of the TL082 unless I disconnect the stereo chorus board. Even then, though, there is still a "fuzzy" sound at the outputs of the main board when more than one key is pressed (around two-thirds of the time, at least), despite the offset voltage being set appropriately. Messing with that pot while playing keys produces no changes either. The fact that the stereo chorus board being connected affects the quality of the sound output of the main board and the DAC offset voltage leads me to think there could be a problem on that board after all, but that wouldn't quite explain why the DAC trim offset voltage seems to affect literally nothing at all.

 

@IanB that's excellent. I have a smaller yamaha portasound pss-450 which I modified to work as a midi keyboard and was thinking of doing something just like that to avoid having to find a place for this table-consuming monster.

 

Also, as seems to be the trend with my posts, just before sending this, I checked something else and realized: I wasn't actually plugged into one of the line out jacks, I was plugged into the volume control pedal jack. I've had the part of the unit with the port labels face-down and got them backwards. Evidently, line out is very quiet and has a lot of white noise. But then again, the amplifier board's power cable still isn't plugged in, so it could be that some amplification is ordinarily supposed to take place there.

Edited January 10, 2022 by Aaron Wright

[IanB]

I really didn't know about this 1Hz multiplexing between the CPUs. Seems weirdly slow. I'll have to have a more thorough read of the service manuals!

[Aaron Wright]

15 hours ago, IanB said:

I really didn't know about this 1Hz multiplexing between the CPUs. Seems weirdly slow. I'll have to have a more thorough read of the service manuals!

It doesn't seem to indicate any specific timing in the service manual. I only found out from these pages that there were two alternate waveforms at all:

 

Anyway, next step is to determine why my DAC offset voltage is misbehaving. It's strange that the manual doesn't go into any detail about the role of the offset voltage in the timing of simultaneous keys.

[IanB]

Ah, okay, the thing is though that isn't operating at 1 Hz. Each sample and hold is operating at about 20kHz. It's demultiplexing the DAC output (which is at 40kHz). The DAC's output is being used by both music LSIs, so the DAC's output goes (LSI 1, sample1), (LSI 2,sample 1), (LSI 1, sample 2), (LSI 2, sample 2), (LSI 1, sample 3)... and so on. Sample and hold 1 then captures the output as an analogue voltage when it's LSI 1 (controlled by "Signal SH from Master LSI") and S&H 2 captures the output when it's LSI 2 ("Signal SH from Slave LSI). So each S&H is operating at the sample rate of 20kHz.

 

So I still don't know what this 1Hz thing is. If it's swapping over from LSI 1 to LSI 2 once per second, that is just plain wrong and would give you 1 second of audio from one set of keys then 1 second of audio from others. If it is doing that, there's a problem with the CPU control of the music LSIs.

 

The DAC offset voltage has nothing to do with timing, it is literally an offset voltage added to the DAC output to centre its voltage at the right point, and if it's wrong it would cause clipping of either high or low DAC values (which is why you'd get "fuzzy distortion").

 

Also "The Block eliminates a high frequency noise called as "Glitch" which appears at the end of the stepped waveform" is rather unclear English. What it actually means is that it only enables a sample (and hold) of the DAC output voltage when the DAC has settled on a sample, thus cutting out spurious voltages as it switches its output.

Edited January 11, 2022 by IanB

[pianokeyjoe]

6 hours ago, Aaron Wright said:

It doesn't seem to indicate any specific timing in the service manual. I only found out from these pages that there were two alternate waveforms at all:

 

Anyway, next step is to determine why my DAC offset voltage is misbehaving. It's strange that the manual doesn't go into any detail about the role of the offset voltage in the timing of simultaneous keys.

@IanB Well sir, this circuit explains to some degree, the thick Brass ensemble or Synth Brass sound I like so much that can only be found on hardware Casio CZ and SK1 keyboards. I knew there was something unique going on!

[IanB]

Just a quick note on the CT's voices. Casio call oscillators "lines" so I'll use their terminology.

 

Each 933 has 8 lines. On a CZ101 (one 933) this allows 8 single Line voices, or 4 dual line voices (thick voices are normally dual line of course). The CZ3000 etc have two 933s enabling 16 single Line voices or 8 dual line voices. You can further layer two dual Line patches to get four 4 line voices ("Tone Mix"). The CT6500 has two 933s, and 16 Lines.

 

In normal keyboard mode you get eight dual line voices (no additional poly for single line, the patches are mostly dual line anyway). In split keyboard mode you get 4 dual Line voices on each half of the keyboard. In auto-accompaniment mode, or in multi-channel MIDI mode, it assigns:

 

Keyboard/Ch 1: 8 Lines (4 voices, dual Line)

Bass/Ch 2: 2 Lines (1 voice, dual Line)

Chord/Ch 3: 4 Lines (4 voices, single Line)

Obligato/Ch 4: 2 Lines (1 voice, dual Line)

 

In this mode I presume that one 933 is doing the 8 lines for the keyboard and the other 933 is doing the accompaniment/channels 2-4 but I've never tested this.

 

The drums (Channel 5) are analogue and nothing to do with the 933s.

 

At one time I was considering trying to write a new ROM to give more flexibility (e.g. 16 single line voices, or 8 dual line Monosynths running different patches, etc), but documentation on the 7811 microcontrollers is sparse and incomplete and it looked very much like a rabbit hole I'd regret falling down so I abandoned that project. I did also consider just controlling the 933s with a complete different controller (e.g a microcontroller or Raspberry Pi) but, same again. Not everyone's life may be too short for such a project, but I think mine is

Edited January 11, 2022 by IanB

[Aaron Wright]

4 hours ago, IanB said:

Ah, okay, the thing is though that isn't operating at 1 Hz. Each sample and hold is operating at about 20kHz. It's demultiplexing the DAC output (which is at 40kHz). The DAC's output is being used by both music LSIs, so the DAC's output goes (LSI 1, sample1), (LSI 2,sample 1), (LSI 1, sample 2), (LSI 2, sample 2), (LSI 1, sample 3)... and so on. Sample and hold 1 then captures the output as an analogue voltage when it's LSI 1 (controlled by "Signal SH from Master LSI") and S&H 2 captures the output when it's LSI 2 ("Signal SH from Slave LSI). So each S&H is operating at the sample rate of 20kHz.

 

So I still don't know what this 1Hz thing is. If it's swapping over from LSI 1 to LSI 2 once per second, that is just plain wrong and would give you 1 second of audio from one set of keys then 1 second of audio from others. If it is doing that, there's a problem with the CPU control of the music LSIs.

 

The DAC offset voltage has nothing to do with timing, it is literally an offset voltage added to the DAC output to centre its voltage at the right point, and if it's wrong it would cause clipping of either high or low DAC values (which is why you'd get "fuzzy distortion").

 

Also "The Block eliminates a high frequency noise called as "Glitch" which appears at the end of the stepped waveform" is rather unclear English. What it actually means is that it only enables a sample (and hold) of the DAC output voltage when the DAC has settled on a sample, thus cutting out spurious voltages as it switches its output.

 

I didn't mean to imply that the 933s, CPUs, or the sample and hold circuit were running at 1Hz, only that by listening to the two separate waveforms output by the final TL082, they do indeed alternate, in a sense, once per second. I'll try to explain the phenomenon in as clear of detail as I can so that there's no further confusion.

 

If I listen to one output of the TL082 and press some keys, sound will only be produced by those keys and come out of that output if I press those keys during a one-second interval. It switches every second between capturing keys and not capturing keys. If I hold a key which it has captured, the sound will continue to come from whatever output the key started on, even during periods when it is not capturing new keys.

 

If I listen to the other output, then any keys I pressed that didn't play from the first output will play there instead. Apparently they're combined in the stereo chorus board in the end, so there's no loss of sound at all once that happens.

 

It's not that each 933 seems to be producing sound on alternating seconds, only that they seem to be capturing keys on alternating seconds. I did understand what the manual was describing, I was just citing it as the only clue I got about multiple waveforms being output in the end, and advising you not to look too closely for anything regarding a 1Hz trading behavior between the 933s, since that isn't mentioned anywhere in the manual and looking for it would be fruitless.

 

 

I was going off on a bit of a tangent with the DAC offset voltage, but it's good to know that that's not responsible for the overdriven sound of pressing multiple keys. I was just trying to find a potentiometer that actually affected the sounds per @pianokeyjoe 's suggestion, but the DAC was the only trim potentiometer adjustment documented in the manuals. Then, I found the voltages were not within spec and got carried away. There are only 6 potentiometers on these boards (not including the "tune" dial on the back): one that adjusts DAC offset voltage, two that control the Low-Pass-Filters on the stereo chorus board, two that affect drum sounds, and one on the amplifier board that I have no clue about. The issue does seem to be present right at the output of the DAC, though.

 

I don't have any leads anymore as to the source of the overdriven sounds of pressing multiple keys. Here's a recording of what I'm hearing, in case that helps. I'm pressing two keys at the same time, over and over, and it only sounds the way it's supposed to three out of 8 times. Very odd.

 

Edit: it doesn't seem to be affected by where I'm listening in on the sound, whether it's from the DAC output or the input to the amp board after the stereo chorus effects.

 

Edit 2: that is correct @IanB . The only thing that's technically 1Hz is the sync signal between the CPUs. I have to mention this in an edit because I've reached my 24hr post limit yet again.

Edited January 11, 2022 by Aaron Wright

[IanB]

@Aaron Wright Okay, so let's check I understand you correctly. We have a period of 2 seconds (so to be pedantic it's .5Hz  ) which I'll call Up and Down for want of something better. When you press a key, the CT's CPU assigns it to one of the two 933s- if it is during the Up period, it is assigned to 933 A, if it's pressed during the Down period, it's assigned to 933 B. So it actually starts sounding immediately, on whichever 933 it is assigned to- but if you are only listening to one output and it's assigned to the other, you won't get to hear it until the signal switches over. Right?

[IanB]

Regarding the strange sound- it may be that it's fine if both keys are sounding on the same 933 but not if they are on both 933s, or vice versa (fine on different, bad on the same). You could disconnect one of the op amp channels at a time and see if you're still getting overdrive. BTW if you haven't socketed the TL082, now would be a good time to do that for ease of fiddling.

[pianokeyjoe]

Man oh man oh man! I heard the sound example my friends and I am telling ya, that is the same issue I had when I bought my first CZ5000 back in 2010. Back then I was instructed by some one online via EBAY since I bought it on Ebay, and was ready to return it, when I got the instructions to trim a pot. I was told it is NOT documented at the time, but that that pot adjusted DC offset and was very very sensitive in its adjustment and that that is what was making that fuzzy sound and needed to be adjusted in one direction or the other until the sound cleared up. BUT, that was on the CZ5000 and CZ3000 and CZ2000S(Yes, the one with built in speakers and built in memory battery). Sooo, I honestly do not know what else to tell ya as the CT6500 seems to be the same hardware but slight differences in it from the CZ series? I really have to dig up my parts unit to really tinker around with it to find out what is what myself. Your fuzzing sound is at normal volume though. My parts unit fuzzing sound is super duper LOUD!! And overdriven. When I replaced the AMP board, I got my sound to be normal..

[IanB]

@Aaron WrightHopefully you'll be out of N00b Quarantine soon!