Announcing the Teensy Saber open source sound board

[profezzorn]

Ran into another thing that I am a little confused about.  Where exactly did you put your 50-100ohm resistor for the PL9823 blade.  The schematic would seem to indicate that it goes in the blade itself (but your pictures only showed what I think was the ID resistor).  I was thinking of just placing the resistor on the hilt side of the aviation connector.  Since I will likely not be doing an star blade, I can't see the downside to this.  Thoughts??

Are you planning to charge through the blade connector, or some other way?
If you put the resistor in the hilt, it will add to the blade ID resistance, which will require tweaking the blades[]
table a little, but other than that it should work just fine.

If you only have a single blade and a separate charge port (meaning you only have one entry in your blades[] array), then you can skip the blade ID resistor completely.

[Fletch]

I understand your explanation.  I just was having trouble thinking through all the options.  I am honestly only planning on doing the PL9823 string blade at this moment.  In other words, I am not sure if I have plans to do any other blades at this point but having options is never bad, right?

I do want charge my battery through the blade connector.  Again, I am a little confused about the statement you made about doing the simple blade charger "can be helpful as running the electronics can interfere with the charging in some cases."  I guess I would like to keep it simple and just do the simple charger (less to wire up), but I am also curious about the cases where it would cause problems?

[profezzorn]

I understand your explanation.  I just was having trouble thinking through all the options.  I am honestly only planning on doing the PL9823 string blade at this moment.  In other words, I am not sure if I have plans to do any other blades at this point but having options is never bad, right?

I do want charge my battery through the blade connector.  Again, I am a little confused about the statement you made about doing the simple blade charger "can be helpful as running the electronics can interfere with the charging in some cases."  I guess I would like to keep it simple and just do the simple charger (less to wire up), but I am also curious about the cases where it would cause problems?

Options are never bad. If you want options, go ahead and include the blade ID resistor. :)

The "fancy" charger is cool, because it can show the battery level and play a sound when you start charging, however, there are some drawbacks. The first one is that the power draw from the electronics slows down the charging a bit. For a 500mA charger, it may take 20% longer to charge.  The second problem is that the charger may not actually turn off when the battery is full, because the extra draw makes it seem like the batter is not *quite* full. The workaround for this is to not leave the charger plugged in for long periods of time. Constantly charging the battery will eventually damage the battery.  The third problem is that if the charger does turn off, the electronics will not. Depending on how smart your charger is, the charger will either turn on again after a while, or it will not. If it turns on again, you're essentially back to "the second problem". If it does not, the battery will discharge down to the point where the cut-off circuit cuts the power to the electronics, but at that point the battery will be empty again.  Again, the solution is to disconnect the charger while the battery is full.

Since you really shouldn't leave li-ion battery connected to a charger for a long time anyways (it ages the battery), these drawbacks aren't a big deal for me, but they can be annoying as you have to pay a little more attention to the charging process.

[Fletch]

That make sense... thanks a lot.  Things are coming together.  Thanks again for your work on this and making it available for the hobbyist just trying to do something cool and new.  I will post a picture of our creations when we are done.  We only have a few more days to wrap this up before the new movie.   

[profezzorn]

TensySabers are now available for sale here:
Teensy Saber Kit V3 - Open Source Sound Board - The Saber Armory

[Fletch]

OK...we are having some problems.  Everything was going so well.  We actually had 2 board stacks working and then disaster struck and we don't know what to do.  So...here is the history.

1.  We ordered some V2 boards just before you came out with the V3 version.
2.  So we thought, hey, let's get try the latest and greatest so we got a couple of the V3
3.  We got the V3s working and all seemed to be good.  Except for 1 of the V3 stacks stopped playing audio. We had been handling that one V3 so much that we thought it might be damaged.  So we opted to get the other V2s running.  No big deal right?
4.  On to the V2s.  We assembled them and turned them on.  Everything looked good except that we did not have any audio coming out. 
5.  We tried a different speaker and tried the beep command.  No luck. 
6.  We are getting 5V to the Teensy so I am assuming that the 5V regulator is working.
7.  I am confident that the chip is put on correctly.  If you are looking at the top of the board with the FETs on the right hand side...the audio chip is mounted such that pin 1 is in the top left hand corner. 
8.  Again we tried the serial commands dir and beep.  I was able to see the wav files using dir but nothing from the beep.  No feedback.  I was not even able to use the command amp to turn on the amplifier.  All I got back was "Whut? :amp".  When it boots up, it says that the amp is off.  We tried to see whether the enable pin was working when the blade is turned off and it seemed to.  I want to say it jumped up to ~3V and the serial said it was playing the sound file.

We are out of ideas to try and hope that you can help point us in another direction...software, places to probe, ways to check... Anything?  :)

Thanks in advance!!

[profezzorn]

OK...we are having some problems.  Everything was going so well.  We actually had 2 board stacks working and then disaster struck and we don't know what to do.  So...here is the history.

1.  We ordered some V2 boards just before you came out with the V3 version.
2.  So we thought, hey, let's get try the latest and greatest so we got a couple of the V3
3.  We got the V3s working and all seemed to be good.  Except for 1 of the V3 stacks stopped playing audio. We had been handling that one V3 so much that we thought it might be damaged.  So we opted to get the other V2s running.  No big deal right?
4.  On to the V2s.  We assembled them and turned them on.  Everything looked good except that we did not have any audio coming out. 
5.  We tried a different speaker and tried the beep command.  No luck. 
6.  We are getting 5V to the Teensy so I am assuming that the 5V regulator is working.
7.  I am confident that the chip is put on correctly.  If you are looking at the top of the board with the FETs on the right hand side...the audio chip is mounted such that pin 1 is in the top left hand corner. 
8.  Again we tried the serial commands dir and beep.  I was able to see the wav files using dir but nothing from the beep.  No feedback.  I was not even able to use the command amp to turn on the amplifier.  All I got back was "Whut? :amp".  When it boots up, it says that the amp is off.  We tried to see whether the enable pin was working when the blade is turned off and it seemed to.  I want to say it jumped up to ~3V and the serial said it was playing the sound file.

We are out of ideas to try and hope that you can help point us in another direction...software, places to probe, ways to check... Anything?  :)

Thanks in advance!!

First of all, you need to type "amp on" or "amp off", just typing "amp" won't do anything.
Second, make sure that your config file(s) include the right board config (v2_config.h or v3_config.h, or it won't work)
If you have an oscilloscope, you can hook it up instead of a speaker to see if the amp is doing anything.
If you do not have an oscilloscope, you can try a multimeter set to measure alternating current.
For the V2 boards, you should be able to measure the input to the amp from the analog out on the teensy as well.
If you're not getting audio out from A14, then the problem is definitely on the teensy or software side.

When I've had these kinds of problems, it's usually because the amp didn't quite connect right, so I have to
de-solder it, apply a tiny amount of solder paste and put it back on again.

[Fletch]

OK...we just put a scope on A14.  Then we thought A14 can't be the right pin to test since it is NOT connected to the saber board.  We consulted the pictures a little closer and realized that we don't have a connection.  Stay tuned...that may have been the problem.  Did i mention we were stupid?

[Fletch]

Yes...we are stupid.  Hooking up A14 to the saber board is kind of important.  :)  It is workinnnnnggggggg! (drop to young anakin)

For the other V3 board, can i probe the teensy like the V2?  You know...just to make sure that the signal is going out to the saber board.

[Fletch]

One more thing that is weird.  When I power the blade using the serial pow command or use a switch, the blade almost immediately turns off and the serial power says "battery low, turning off:  2.60" and it shows that the audio is still playing.  It just keeps repeating the following "Playing font02   channels: 1 rate:  22050 bits:  16.  and then the battery voltage measures 3.4. 

any ideas on this one?

[Fletch]

one more bit of info...

on the V2 board I connect the GND pin on the saber directly to the BAT negative and it works just fine.  When I go through the blade hooking up the GND on the saber to pin 2 on the blade.  It acts like a kill switch (by design) but give me that funky issue where the blade turns off and it gives me a low battery turning off message (and actually turns off the blade but continues to tell me that the wave file is playing in an endless loop).

[profezzorn]

Yes...we are stupid.  Hooking up A14 to the saber board is kind of important.  :)  It is workinnnnnggggggg! (drop to young anakin)

For the other V3 board, can i probe the teensy like the V2?  You know...just to make sure that the signal is going out to the saber board.

You can, but the V3 board uses I2S (which is digital) to send data to the amplifier. You should see things going on on pin 9, 22 and 23 when it's sending data, but it won't look like audio..
It's fairly likely that the V3 that used to work but doesn't anymore has a blown amp, but it *could* be anything. The V3 amp is fairly durable though, so maybe it's something else.

[profezzorn]

one more bit of info...

on the V2 board I connect the GND pin on the saber directly to the BAT negative and it works just fine.  When I go through the blade hooking up the GND on the saber to pin 2 on the blade.  It acts like a kill switch (by design) but give me that funky issue where the blade turns off and it gives me a low battery turning off message (and actually turns off the blade but continues to tell me that the wave file is playing in an endless loop).

That's weird.  Usually these sort of things are caused by weak batteries, but if it works well when hooked up directly, that's probably not the issue.
It kind of sounds like you have high resistance in the connector somehow. What kind of connector are you using?
Is BATT- also hooked up to through the blade connector, or is it hooked up directly to the battery?

This kind of thing should be easy to figure out though, as you can just measure voltage between GND and various points to see where the weird voltage drop occurs.
The voltage detection is pretty simple, there is a 2M - 220k voltage divider hooked up to pin 14, which I just run analogread on, so the expectation is that the voltage on pin 14 is about 1/11th of the actual battery voltage.

[Fletch]

That's weird.  Usually these sort of things are caused by weak batteries, but if it works well when hooked up directly, that's probably not the issue.
It kind of sounds like you have high resistance in the connector somehow. What kind of connector are you using?
Is BATT- also hooked up to through the blade connector, or is it hooked up directly to the battery?

This kind of thing should be easy to figure out though, as you can just measure voltage between GND and various points to see where the weird voltage drop occurs.
The voltage detection is pretty simple, there is a 2M - 220k voltage divider hooked up to pin 14, which I just run analogread on, so the expectation is that the voltage on pin 14 is about 1/11th of the actual battery voltage.

I have everything hooked up to an external power supply at the moment.  So the battery isn't even a factor.  I am using the aviation connector that you specified along with the blade ID resistor and the other resistor to prevent current leakage into the LED string.  We measured the resistance through the blade connector and it was 1.5ohms.  I think we measure ~.28V on pin 14...then I do the pow command and it drops to ~.15V.  When I bypass the blade and connect directly to Batter neg. the pin 14 measurement is ~.31V and when I turn the blade on it drops to ~.27V.  Could there be too much resistance in the wire?  1.5ohms does not sound like that much.

BATT- is hooked up to pin 1 on the aviation connector.  Then pin 2 comes back to the GND on the saber.  I also have BATT- hooked up to BATT- on the saber.  I assume this is correct.

[profezzorn]

That's weird.  Usually these sort of things are caused by weak batteries, but if it works well when hooked up directly, that's probably not the issue.
It kind of sounds like you have high resistance in the connector somehow. What kind of connector are you using?
Is BATT- also hooked up to through the blade connector, or is it hooked up directly to the battery?

This kind of thing should be easy to figure out though, as you can just measure voltage between GND and various points to see where the weird voltage drop occurs.
The voltage detection is pretty simple, there is a 2M - 220k voltage divider hooked up to pin 14, which I just run analogread on, so the expectation is that the voltage on pin 14 is about 1/11th of the actual battery voltage.

I have everything hooked up to an external power supply at the moment.  So the battery isn't even a factor.  I am using the aviation connector that you specified along with the blade ID resistor and the other resistor to prevent current leakage into the LED string.  We measured the resistance through the blade connector and it was 1.5ohms.  I think we measure ~.28V on pin 14...then I do the pow command and it drops to ~.15V.  When I bypass the blade and connect directly to Batter neg. the pin 14 measurement is ~.31V and when I turn the blade on it drops to ~.27V.  Could there be too much resistance in the wire?  1.5ohms does not sound like that much.

1.5 ohms is actually very high for a connector, but it shouldn't matter much unless you put power through it. Do you have BATT- hooked up to the blade connector or to the power supply?

How much power can the power supply provide? A 3A power supply is not enough to power a blade....
Anyways, you need to keep measuring at different points, now measure at BATT+, BATT-, both sides of the blade connector and at the power supply. It sounds like the power drop is in the connector, but you should verify that. BATT- and GND are supposed to be close, so there shouldn't be much voltage difference there. BATT+ should have the battery voltage.