Coming Soon: Teensy-LC (low cost Teensy)

Status
Not open for further replies.
bperrybap said:
Anybody got pointers to the two datasheets?

--- bill
Click to expand...

here are datasheets
http://www.pjrc.com/teensy/K20P64M50SF0.pdf

http://cache.freescale.com/files/microcontrollers/doc/data_sheet/KL26P64M48SF5.pdf?fasp=1

I have observed, internal pullups on LC are around 43K, but only 32K on teensy 3 (your mileage may vary).
if your logic requires pullups you might use external 2.2k pullups to reduce RC time (a scope would be handy).
I guess you could also try 5k series resistor and hopefully not stress the teensy (though lacking EE credentials, I shouldn't dispense electrical tips)
 
manitou said:
I guess you could also try 5k series resistor and hopefully not stress the teensy (though lacking EE credentials, I shouldn't dispense electrical tips)
Click to expand...

Wouldn't the 2.2k resistors stress the Teensy more than 5k resistors? FWIW, I have used 2.2k resistors happily with the Teensy 3 series (I2C and so on)
 
Constantin said:
Wouldn't the 2.2k resistors stress the Teensy more than 5k resistors? FWIW, I have used 2.2k resistors happily with the Teensy 3 series (I2C and so on)
Click to expand...

The 5k resistor is in series with 5v. I intended the 2.2k pullup to pull to 3.3v. If the logic is like I2C where pullups are required, then it would seem the series resistor woudldn't be required, because the device is only pulling the line low. I suspect the 5v device is actively driving the pin to 5v, so INPUT_PULLUP is probably not required...
 
A series resistor in a GPIO input line is not a good thing.. Simple math can tell you if the input signal, after the resistor, can achieve a logic 0. Or worse, the voltage for a logical 0 hovers near the threshold.
 
Just getting ready to look at the datasheet for the LC.
To stop the guessing, pullups are not needed.
The LCD drives the data signal lines both ways (+5/GND) when the data lines are outputs.
(technically it isn't 5v but the voltage on the USB since everything is powered from USB
so it is a tad lower than 5v)

I'll write some special test code that will allow me to take voltage measurements
on the signals with a voltmeter so I can at lest see the static input voltage levels
(before and after the series resistor)
and compare them between Teensy 3.0 and Teensy LC to see if there is any
difference external to the chips.

If I don't see anything obvious in the datasheet or after doing some math,
it may be time to borrow a scope or it may be a good excuse just go buy a digital scope.
I always like new toys :)

--- bill
 
Teensy-LC will begin shipping on March 12.

That's 2 days later than the estimated release date, originally published almost a month ago.
 
ok, tracking my 5v pin read/write issue down a bit more.
writes to the GLCD are working as expected.
The issue is occasionally pin reads don't read the expected level correctly.
It isn't random and is very repeatable which is actually great.
It appears that the pin read issues track back to two pins and primarily one pin.
The two pins are digital pins 16 and 17 with most of the issues occurring on digital pin 17.

I did notice that those two pins are special and pin 17 has some sort of level shifter on it that goes to another hole.
is there any sort of special s/w handling that has to be done on these pins?

Is there a schematic for the Teensy LC so I can look at the loading on digital pin 17 by the level converter?

I'm off to write some special code to freeze right in the middle of a failed read so I can look at the actual
voltages on all the signal lines and see what is happening.

--- bill
 
@bperrybap

Pin 17 is the 5V drive and has a level converter package on it that will probably have a tendency to pull high or low. Rather than trying to freeze the LCD writes would suggest using resistors to pull it and an adjacent pin high and low and see if there is any difference in steady state voltages. Unfortunately it's possible that while the level converter should have a high steady state input impedance it may be delaying the response from the LCD drive circuit that will only how up with a oscilloscope and feeding it a pulse train at the LCD data rate.
 
I know about the level converter, which is why I'd like to see a schematic
to see what type of load it puts on the ARM pin.
I've increased h/w timing delays to the millisecond range with no changes so I don't think there is any sort
of slew rate issue for the signal nor any type of realtime signal disturbance as the signals are stable
on the input pins for several milliseconds before the ARM is reading the pin state.

In the absence of a further information or a schematic,
I think freezing the code is good thing to do right now.
I want to get a voltage reading for all the signals, while the LCD is driving the data bus as I've never done this yet.
I'll look at both sides of the series resistor to see where the voltages are and compare pin 17 to the others.

Since it is so predictable I can add special code to freeze the code where the arm has seen incorrect data
and then look at the voltages on the signals to see what is going on.
I think it will be useful to see the what the voltages are vs what the ARM is "seeing"
when the LCD is driving the bus.
 
No schematic yet, probably won't be ready for another week or two.

The logic level converter is a SN74LV1T125 chip with high impedance input, with only a few pF capacitance. It is NOT a mosfet transistor with pullup resistors.
 
Ok, so I trapped the bad read.
The data on the pins was correct when everything was stopped.
The highs were 4.88 volts and the lows were 0 on both sides of the resistors.
I'm definitely beyond the digital world.
The code normally tells the LCD to drive the data bus then goes to read it
around 320ns later.
It works with the Teensy 3.0 when using 10k serial resistors.
It doesn't work with the Teensy LC.
It works when the delay is increased to 150us so there is some sort of
AC type issue involved.

A delay of that long is not really usable as it is severe impact to performance.

Anyway the only thing I can try now is to use a smaller resistor to see if that improves things.

I'm still not sure why this is an issue for Teensy LC and not for Teensy 3.0
When I looked at the datasheet the AC characteristics looked the same
so I'm not sure what is happening.
This is where a scope would come in handy.

-- bill
 
Teensy-LC is officially released!! :)

All Teensy-LC pre-orders (without pins) were shipped today. The few orders with pins, placed within the last 48 hours, will ship tomorrow.

Plenty of Teensy-LC boards are in stock.

A final Teensyduino 1.21 release will be made within the next day or two, for official non-beta Teensy-LC software support.
 
Congratulations! I can't tell you how much rather I'd be at home testing the LC than traveling the country for the month of March. Visits to manufacturers are great fun but being away from home isn't.
 
I've updated the Snooze v5 low power library and now it works pretty well with the LC. I also updated the examples to reflect the differences between Teensy 3.x and LC. Let me know if you encounter any problems.
 
PaulStoffregen said:
Teensy-LC is officially released!! :)

All Teensy-LC pre-orders (without pins) were shipped today. The few orders with pins, placed within the last 48 hours, will ship tomorrow.

Plenty of Teensy-LC boards are in stock.

A final Teensyduino 1.21 release will be made within the next day or two, for official non-beta Teensy-LC software support.
Click to expand...

Sweet, congrats Paul!

I'm really sorry that I haven't been helpful during the beta test phase but I still have to receive my beta boards...
 
Saw that mine were shipped yesterday, looking forward to them Paul! Already got a bunch of pogo pins for one of them (to make it a test device for my applications).
 
Last edited:
Great news! My beta testing didn't feel much like beta testing at all, most of the problems were Micro$ofts fault :) so I'm sure you will have great success with this product.

At the price, It largely makes more sense to use a teensy-LC instead of multiplexer breakout boards whether analog or digital. I was hoping to finish my Teenstrument completely before the LC release, to have the first finished open source LC based project but I've been on holiday for the past week and frankly, no, im not sorry!! =D

I'll keep cracking on anyway and get it finished.

Time for a holiday for Paul now...
 
Stupid question: what pin has the 74LV1T125 connected? Is it digital pin 17 (i.e 17/A3/SDA0/PWM) or the one at the bottom of the board between 12 and 3.3V? They're both labeled 17. The store page says, "A 74LV1T125 buffer is connected to pin 17, with the increased output voltage available on another pin."
 
the one labelled as just '17' is the raw I/O from the IC at 3.3V. The one on the end between 12 and 3.3V is an output only at Vin (5V if running from USB) from the level converter.
 
Got my TeensyLC's in the mail yesterday!

Is there an update the Arduino environment and/or to the loader application to support them?

12:51:32: HID/win32: vid:045E pid:076C ver:0083
12:51:32: HID/win32: vid:16C0 pid:0478 ver:0102
12:51:32: HID/win32: usage_page:FF9C, usage:0020
12:51:32: Device came online, code_size = 65536
12:51:32: Board is: Unknown board, version 1.02
12:51:32: File "Blink.cpp.hex". 10616 bytes, 16% used
12:51:32: File "Blink.cpp.hex". 10616 bytes, 16% used
12:51:32: elf size appears to be 262144
12:51:32: elf binary data matches hex file
12:51:32: Code size from .elf file = 262144
12:51:32: Incompatible file, showing warning dialog
12:51:32: HID/win32: HidD_GetPreparsedData ok, device still online :)
12:58:35: Verbose Info event

Thanks,

DougM
 
Status
Not open for further replies.
Back
Top