Ok, we are going to shift gears.

Keep in mind that ideally we would hope for a simple fix such as a shorted bypass capacitor and a blown inductor in the VDD power feed.

With your scope we are going to look for bus activity starting with the first major components, beginning with IC1.

At Bottom Right of the board, locate IC48 SN74HC04 and probe the first inverter at pins 1 and 2. Look for signal activity. This is CE' on the main ROM.

At IC6, D27C256AD-15 (XG879), probe the data bus at pins 11- 13, 15-19 and look for activity on each pin.

All we are looking for is that the pins are changing between about 0V and 5V.

 

I am on it Mr Chips, thank you, your help and support is awesome.

 

Sorry for the delay. I have been flat out. Be back with the results, ASAP.

Thanks for your patients.

Kind Regards
Steve

 

Hi after probing where you have stated, something strange is happening. The results are intermittent and changing. Something weird is going on here. This is going to be hard to right up. I am thinking of making a video. Be back later tonight with the write up.

Thanks
Steve

 

Ok, we are going to shift gears.

Keep in mind that ideally we would hope for a simple fix such as a shorted bypass capacitor and a blown inductor in the VDD power feed.

With your scope we are going to look for bus activity starting with the first major components, beginning with IC1.

At Bottom Right of the board, locate IC48 SN74HC04 and probe the first inverter at pins 1 and 2. Look for signal activity. This is CE' on the main ROM.

At IC6, D27C256AD-15 (XG879), probe the data bus at pins 11- 13, 15-19 and look for activity on each pin.

All we are looking for is that the pins are changing between about 0V and 5V.

Hello Mr Chips,

I will attempt to right this up, as clear as possible.


I put my Logic TTL Probe on IC48 (Hex Inverter) Pins 1/2 and found it was pulsing at some unknown frequency. As I was not expecting it to pulsate, I decided to measure its frequency. However, I accidentally slipped with my probe and shorted Pins 1 and 2. (I don't know for sure).

After that Pin 1 was LOW Pin 2 High. Thinking I have shorted something (Very annoyed with myself) I tried restarting the Piano several time trying to get that same pulse on Pins 1/2. I could not. (However I will come back to this below)

All remaining Pins on IC48 (Inverter) were tested with a Logic probe, results as follows.

Pins 1, 2 saw a LOW in and a HIGH out.

*All remaining 5 Inverters were identical with a HIGH In and a LOW Out.


Then I moved onto IC6 (28 Pin ROM) namely Data Lines, D0-D7.
ALL data Pins D0-D7 were LOW.
*Infact all Pins on IC6 were LOW, except Pins 20 CE and 22 OE which were HIGH.
*(I am excluding all GND and VCC)


Now this is where something strange happened. Remembering what I spoke about before where I saw Pulses on IC48 Pins 1/2.

Not seeing any logic changes on DATA Pins D0-D7 (Or any pins for that matter) I decided to hold my Logic TTL Probe on one of the Data lines on IC6 whilst booting the Piano to see if I see any data changes.

This is when it started Pulsing as it was initially when I thought I has shorted Pins 1 and 2 on the Inverter (IC48).

I then went back to IC48 Pins 1/2 and sure enough it was back to its initial state, however this time I was able to read the frequency at 200KHz. (This was a clean Square Wave with the Highs and Lows of equal duration)

Yes, the output of IC48 Pin 2 saw the same output showing on Pin 1, (No doubt inverted). What is interesting is that the data lines on IC6 then showed varying out put frequencies relative to each pin. (As opposed to before where they were all sitting LOW) The output frequencies were not pure square waves, (as you would see when Calibrating ones Scope) they looked more like Data Words. i.e Short H, Long L repeated. (Hope that makes sense) I have forgotten how one describes these frequencies. In other words the Highs and Lows are not of the same duration.

Thats all, I can think of. I have no idea what to make of any of this. I don't know which state the CE line should be in ? Pulsating, LOW or HIGH.

Thanks
Steve

 

When properly running, CE should be changing randomly LOW/HIGH. The frequency is not important (except see notes below).

In some cases, it might be better for us to find it constantly LOW, HIGH or a fixed frequency. This would indicate that something is wrong. Remember, we are trying to find a fault. It may be harder to fix something that is working right.

My job is going to be trying to keep you focused.

I put my Logic TTL Probe on IC48 (Hex Inverter) Pins 1/2 and found it was pulsing at some unknown frequency. As I was not expecting it to pulsate, I decided to measure its frequency. However, I accidentally slipped with my probe and shorted Pins 1 and 2. (I don't know for sure).

Now we don't know if the signal was ok before you started probing or if the accidental short caused an additional fault or if the intermittent behavior is actually what is going on.

Here is what I would expect you to see.
With the probe connected to IC48 pin-1 and POWER OFF, turn POWER ON.
The signal should be randomly changing, typically 1μs (my guess) pulses).
It should stay doing this all the time.

What you might see if there is a FAULT:
Intermittent followed by constant LOW, HIGH or fixed frequency.
If it is fixed frequency, we want to know the frequency.
We also want to know roughly how long it takes from POWER ON to get to this state (but not very important).

Any or all of the DATA lines might exhibit the same behaviour.

Here is the reasoning.
CE is the CHIP SELECT signal to the ROM. Every time CE is active (HIGH on pin-1 and LOW on pin-2, the MCU is reading instructions from the ROM.

Normally this should be randomly changing.
A constant LOW or HIGH means something is dead.
A fixed frequency might mean that the MCU is working but locked up in a tight program loop.

If we find random transitions on CE and DATA then we can move on to another area in the circuit.

(If we have to frequently POWER OFF/ON, see if there is a RESET button on the board. I don't think there is one. We can add one temporarily.)

 

Hi,

I am extremely angry with myself for that slip up. Something I knew could happen hence why I mounted the CRO. (I can't know for sure) Lesson is I have to be super careful.

Did you want me to continue looking at CE as you have described above, switching the unit On/Off, and noting the Input on PIN 1 of IC48?

(If I had shorted PINS 1/2 isn't the likely hood of a new fault small? Firstly the Inverter might have blown. Second A15 from IC1 was forced to change state H/L (not good) finally CE was forced to an incorrect state ?) Just trying rationalize the possible damage)

Can we work past this slip up. ???

Thanks
Steve

BTW, there is a Reset Button. Please see the Test Procedure in the Service Manual. ????

 

If we created a new fault that increases our difficulty exponentially.

Can we work past this slip up. ???

If we created a new fault, no, we cannot work past it.

Don't be too hard on yourself. We are only humans and we all make mistakes.

(If it it makes you feel any better, over the weekend I was supposed to cook BBQ ribs but cooked it at 325F instead of 225F. The ribs were cooked (burnt) in under an hour instead of taking the usual 4-5 hours to cook. Simple numerical mistake.)

We will just keep this in mind as we hunt for whatever faults may exist.

At this point we are looking to see what has locked up.
As you turn the POWER ON (or press RESET) I need you to describe what you see on:

1) all front panel displays
2) CE at pin 1 and pin 2
3) activity on the DATA bus

We are looking for both transient (on initial POWER UP) and steady state conditions.

At this point I need to take a break to plan the next line of attack after you respond with your observations.

 

While we wait for Steve to come back with some results, here is the best BBQ ribs I have ever cooked.

Marinate ribs overnight. Set a large aluminum pan with water on the BBQ grill. Another grill is placed on top of the pan for the meat. Do not use BBQ sauce. This will burn the ribs.

Preheat the BBQ to 225F. This is the most important part.

Cook the ribs at 225F. Takes 4-5 hours. Keep the lid on, opening only to add boiling water and to turn the meat once.

After cooked, baste with BBQ sauce and cook over regular grill just to brown the ribs.

Enjoy with a nice cold beer.

 

The test program can only be used when the piano is working. It is used by the technician to diagnoise the piano when all the CPUs are working but certain non-critical parts of the hardwares like LED, switches, padels, VRs, pitch and tone volumes are not right.

The test program is initiated by using certain key combinations on the keyboard during power switch ON.

The CPU reset signal for the DM board is at pin 87 of the KPU1 called /IC (initial clear). It was also connected to pin 1 of IC44. The signal was started from TR5 and the 4 components connected to the collector. I think it should be safe to wire a push button to ground from there and use it as the reset signal for the DM board.

Allen

 

The test program is initiated by using certain key combinations on the keyboard during power switch ON.

Thanks Allen.

Yes, there are a whole bunch of simple things we have not checked yet, for example, the status of IC and MRST.

Do you have info on what the test combo and sequence is?

 

Well, for CVP50, it was A1 & C7 key combo - the leftmost and the rightmost keys both hold together while the power is switched on.

[edit] it should be A-1 & C7 [/edit]

Allen

 

So how does one do that! With the left hand, right hand and "middle digit" I suppose?

 

It's possible. The power switch is next to A-1 key.

Allen

 

Hi Gentlemen. I have been able to hold down the Keys with BluTac. However, as Allen has said this shows no results, either on CE (Pin1 IC48) nor any of the other Test Routines function. (Not to mention the Display needs to be working)

BTW, Thanks Allen for that info on making a Reset.

To answer your 3 questions Mr Chips. As follows.

1. No there is no display at all. Never has been. Zero LEDs light up. However the Floppy LED does but I think thats part of its internal (Floppy hardware seems similar when a PC boots) routine when it gets power. The light soon goes out when no floppy is detected. (Or for whatever reason)

2. I have use a Logic TTL Probe and found the following on IC48 Pin1 and 2. (Low/high respectively)


3. All Data pins are LOW on IC6 no sign of activity with the TTL Probe.

However I did got back over all this Pinouts with a Scope. As follows.

Pin 1 CE IC48 showed some noise at the very high frequency. I am calling it noise because its a a very high frequency and I feel that it maybe simply IC electron noise.
Regardless I have taken a Pic of this and will up this shortly. (Its random chatter shown at around 100mV P-P @ 100ns/Div shows it clearly on the CRO) See image. http://flic.kr/p/cN22Qw

This noise was also present on the Data lines and again I could NOT see any change of logic states. They are all LOW.

I can't help feel that we are onto something here, as I stated before, I saw a clear 200KHz frequency from IC48 Pin1 CE. Now, regardless of how many times I reboot (Via Power switch) its simply stays low. No transients were observed.

Pffffw

Thanks
Steve


Doesn't this suggest that IC1 (KPU) A15, (Pin 84) has a problem ????

BTW, there is a voltage divider with what looks like a Zener (Ref F1 Page 49) that I have not checked.

 

Ok. We will have to back up.
Usually, here are the first things one would do.

1) Visual inspection. Looking for blown fuses, burnt components, resistors etc. Leaking or bulging capacitors. Discoloured PCB, frayed wires, bad connectors, solder joints, burnt smell.

2) Power on test. Burnt smell. Hot components. Noise, buzzes etc.

3) Voltage tests. Test all known voltage test points.

4) Now that we know a little bit more about this case, use the scope and find IC' and MRST' in the schematic. See TR5 (left of IC1 in the schematic).

Measure status level at IC', MRST', base and collector of TR5 (ok to measure at resistors). At this point we are not interested in noise (unless it is excessive, greater than 100mV). We are interested in DC voltages, 0V or 5V +/- 0.1V.

(Don't get distracted by 100mV noise. You need to know what is important and what is a distraction. Though sometimes a tiny artifact can be essential. It is important to stick with the main reason for testing a particular junction while keeping at the back of your mind any unusual artifacts.)

I will try my best to keep the dialog going so that you can understand the test procedures while at the same time I have to keep on track of my own line of attack.

 

Ok I have taken all this on board. To be honest, I don't know some of those acronyms but I will ask when I get to that.

However, now this is just a suggestion, wouldn't it be wise if possible to hunt down the fault in the Front Panel, getting the 2 x 7 LED Segment working so we can then run the Test Procedures as outlined in the service Manual ?

Thanks
Steve

In the mean time I will work though your Points.
But right now, I could do with one of your BBQ Ribs.


BTW, Thanks for your help, I am learning something and I need to remember that this is part of this exercise.

 

No. To run the test procedures we have to get the main CPU running first.

Don't distract me. I need to know the voltage level at IC44 pin-1.

I have not looked at the photos to find the best place to probe IC'. This signal should be at 5V.

 

No. To run the test procedures we have to get the main CPU running first.

Don't distract me. I need to know the voltage level at IC44 pin-1.

I have not looked at the photos to find the best place to probe IC'. This signal should be at 5V.

4.97V measured with DVM.

 

2. I have use a Logic TTL Probe and found the following on IC48 Pin1 and 2. (Low/high respectively)


3. All Data pins are LOW on IC6 no sign of activity with the TTL Probe.

You cannot use a TTL probe to help us with this level of trouble shooting

Doesn't this suggest that IC1 (KPU) A15, (Pin 84) has a problem ????

Yes. If you accidentally shorted IC48, pins 1 & 2, SN74HC04N and the signal at pin-1 is now dead, "Houston we have a problem".

Unfortunately we have only one driver for A15 and that is coming from IC1.
Our only hope is that IC48 is the culprit (wishful thinking). If A15 is dead in IC1 we are skunked.

*********** wait ******************
You will have to unsolder IC48. Clip it off for now by snipping each pin.
Leave it off and test voltage on the board at pin 1.
*********************************

This may seem rather drastic but the prognosis is not looking good.

$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
Edit: I have reread your previous post #65.
If at any time you see any activity on the scope at IC48 pin-1, A15 must still be alive. Don't remove IC48.

Let me know if this is the case.
(I have to go do some other stuff now.)
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$