VBAT Power

Status
Not open for further replies.

leonkristal

New member
My understanding is that the problem associated with the Teensy not coming up properly is that the diodes between the 3.3V and VBAT give to low a voltage when the oscillators are turned on. I am not using an external VBAT so the voltage is coming from the 3.3V regulator through a diode. If I short out the diode will this alleviate the problem? Would I create any other problems by doing this?
 
Could you provide a reference to the issue you are concerned about. I've read here that some power supplies, when turned on do not ramp up the output fast enough, or in a non-linear fashion, and cause POR problems. This can be fixed by applying a larger capacitor as others have mentioned in their posts. Another method would be to use a POR circuit that holds RESET at ground until the power supply gets over 2.5 to 3v then waits for an additional 200-400mS and releases RESET to its operating state. Use something like an APX803L with an open drain output. This is important as a Push-Pull output will prevent you from programming the Teensy. I found that out the hard way.

RESET drive by open collector only. https://forum.pjrc.com/threads/34108-Teensy-resethttp://
 
Indeed reference - what Teensy - and is this about an unseen problem based on forum posts or has it been observed?

IIRC:: Prior to TeensyDuino 1.46 release the startup sequence of T_3.6 allowed for some slow start external supplies to interfere with normal startup. The T_3.6 startup ordering was altered in TD 1.46 with the general expectation that was resolved.

But AFAIK this doesn't relate to VBat which is indicated on the Teensy Card as taking a 3V supply that keeps the RTC portion active when Teensy otherwise unpowered.
 
Diode Drop

Indeed reference - what Teensy - and is this about an unseen problem based on forum posts or has it been observed?

IIRC:: Prior to TeensyDuino 1.46 release the startup sequence of T_3.6 allowed for some slow start external supplies to interfere with normal startup. The T_3.6 startup ordering was altered in TD 1.46 with the general expectation that was resolved.

But AFAIK this doesn't relate to VBat which is indicated on the Teensy Card as taking a 3V supply that keeps the RTC portion active when Teensy otherwise unpowered.

My understanding of the issue from Paul is that the problem is that the VBAT voltage when powered by the 3.3V regulator on the board is a diode drop behind the rest of the chip. The code turns on the oscillator based on the 3.3V but portions of the chip are based on the VBAT voltage so the chip doesn't operate properly. By shorting out the diode I would hope this problem would be alleviated without adding delays in the software, adding supervisory chips, etc.

We are using a supervisory chip which holds the reset line low for ~350mS after the regulator hits 3V, we are using version 1.4.6 of the code, and our rise time on the power supply is in the low mS range, and we are still having the Teensy not come up properly. What we see is that the Teensy will come up some number of times and then stop working. Power cycles do not bring the Teensy back up. We then hold the reset line low for about 10 seconds and we can then reprogram through the USB and it works again.
 
Does the problem startup happen if you program your Teensy with something very simple, like the 01.Basics > Blink example?
 
Paul, The software engineer, Gerard should be able to answer your questions about the software. If anything comes up about the hardware I'll answer it. The supervisory chip we placed on the 3.3V, ground and reset line is the MCP130 from Microchip. We soldered it directly to the center pins on the Teensy board.
 
Status
Not open for further replies.
Back
Top