Brown out issues persist eve nwith new prop , what am I doing wrong
JetMan
Posts: 39
Last week I had started a thread entitled new guy is lost , I was having issues as to where every time I sent a very simple program to the chip it would execute a quick pulse output on whatever pins I specified and then halt. This was originally in a breadboard circuit design that I had made to control a jet engine and contained numerous components. Yet the program I had written was to simply turn on p12 and p14 indefinitely and that's it , I also wrote a program to toggle these two pins every second or so using the rcfast clock speed . Since then I had ordered a new prop under suspicion that maybe I had a bad ic . now I have made a simple circuit on a solder-less board with nothing but two led's tied to these pins , a 4mhz resonator (3-pin) and a very simple power delivery circuit using a fixed 3.3vdc regulator with all of the necessary caps. Still having the exact SAME behavior . I've tried both chips , I've used the 4mhz osc. and not used the 4mhz osc. when using the osc. i get nothing at all , when using the rcfast mode i at least get a quick blink on the led's. this is becoming very frustrating as I have gone through all of the basic steps of trouble shooting and have hooked my oscilloscope up to the 3.3vdc to make sure no weird noise was present . I have even attempted to power the circuit from a battery to eliminate any strange noise from my bench to power supply.
Being somewhat new to this forum I don't know how often the Parallax folks get involved in discussions although I had many a well seasoned vets respond to my last post with good suggestions.
Below are the several different bits of code I have tried to use with no luck.
Second rendition
number 3
[noparse][[/noparse]code]
'
'
CON
_xinfreq = 4_000_000
_clkmode = xtal1 + pll16x
PUB MotorOnOff
dira[noparse][[/noparse]14..12] := 111
repeat
!outa[noparse][[/noparse]14..12]
waitcnt (clkfreq/2 + cnt)
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Landon Leigh
Being somewhat new to this forum I don't know how often the Parallax folks get involved in discussions although I had many a well seasoned vets respond to my last post with good suggestions.
Below are the several different bits of code I have tried to use with no luck.
'
'
CON
_clkmode = RCFAST
PUB MotorOnOff
dira[noparse][[/noparse]14..12] := 111
repeat
outa[noparse][[/noparse]14..12] := 111
waitcnt (clkfreq/2 + cnt)
outa[noparse][[/noparse]14..12] := 000
Second rendition
'
'
CON
_clkmode = RCFAST
PUB MotorOnOff
dira[noparse][[/noparse]14..12] := 111
repeat
!outa[noparse][[/noparse]14..12]
waitcnt (clkfreq/2 + cnt)
number 3
[noparse][[/noparse]code]
'
'
CON
_xinfreq = 4_000_000
_clkmode = xtal1 + pll16x
PUB MotorOnOff
dira[noparse][[/noparse]14..12] := 111
repeat
!outa[noparse][[/noparse]14..12]
waitcnt (clkfreq/2 + cnt)
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Landon Leigh
Comments
' ' CON _xinfreq = 4_000_000 _clkmode = xtal1 + pll16x PUB MotorOnOff dira[noparse][[/noparse]14..12] := 111 repeat !outa[noparse][[/noparse]14..12] waitcnt (clkfreq/2 + cnt)▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Landon Leigh
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
Sorry had to draw quick ,
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Landon Leigh
See the data sheet for the LM1094: you are using the wrong capacitors. 10 uF tantalums are specified.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
Post Edited (Leon) : 4/14/2010 6:19:16 PM GMT
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Landon Leigh
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Landon Leigh
Are you sure that it is brown outs that are causing you grief? If you tie BoE to VDD instead then it will disable the detection, you will need a 10K (or so) pull up on the RST pin though.
The serial cct could be reseting when you do not want it to, disconnect that to see.
I get spurious resets from switching on/off my 'scope, but only if the serial lead is in. The earth connection is enough.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Style and grace : Nil point
The crystal is actually a 4mhz 3 pin resonator from parallax , I just didn't have that symbol in my software so i dropped in the prefab crystal with caps. Sp there a re no caps just resonator and ground connection . Believe you me I have tried the basics of disconnecting the serial cable and all that jazz , tried it on the UPS and with out . Tried it with a toroidal power filter transformer and without . tried it on battery only with the laptop and the circuit and every other permutation.
I have tried disabling the boe but didn't realize i needed a pull up on the rst pin so it wouldn't connect. I'll try that now.
hover - I tried many times without the resonator as it just arrived this afternoon , then I have now tried it with the resonator and it gets worse , with the resonator attached and programmed i get no response , without it I at least get a quick pulse.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Landon Leigh
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Landon Leigh
Jim
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Landon Leigh
I'm at a total loss ?!?!?!?!
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Landon Leigh
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Landon Leigh
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Landon Leigh
My Parallax schematic shows a 10nf there. Not sure if that makes a difference.
Jim
Jim
dira[noparse][[/noparse]14..12] := %111
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
*Peter*
Heh, it still works. 111 in binary is 110_1111.
Rich H
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
The Simple Servo Tester, a kit from Gadget Gangster.
Attached is the document showing the .1uf , tech support mentioned that sometimes it causes issues . I don't know why it is there as I am no that well versed in circuit design. Maybe it should be omitted from the document altogether ?
Now it's on to new issues
I have two Very Large DC motors and an ignition coil that are all part of the functional schematic and the coil , when active immediately resets the chip. On occasion the emf from the motors starting and stopping cause enough interference to reset it as well.
Keep in mind:
1.) All inductive loads are powered from a separate power source entirely , which shares a common ground with the logic level supply.
2.) All devices are controlled through opto's which in turn eventually drive power transistors or mosfets.
3.) Prop has 0.1uf caps strapping both sets of power pins
4.) 3 qty. internal voltage reg's (12vdc, 5vdc, 3.3vdc) have 1000uf filter caps and 0.1uf noise suppression caps strapped .
I'm not really sure what else to do at this point to suppress the noise induction ?
Not that it is needed but I have hooked up the scope to the 3.3vdc power rail and there are definitely distinct noise patterns showing up for each device.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Landon Leigh
Perhaps if you opened up and posted a picture and schematics rather then trying to dole out the information a morsel at a time then you might get some real help.
1) sounds good but the manner in which the grounds are connected is important too
2) Overkill perhaps but no problem there. I drive very heavy inductive loads without this isolation.
3) The caps should be as close to the power pins as possible.
4) 1000uF is fine on the inputs but the outputs should really have something like 10uF tantalums for better response on momentary surges. Charging a 1000uf on startup leads to unregulated ramping plus if the regulator can't supply the peak currents then it won't be regulating as the cap is discharging. The 10uF is really to improve the load transient response of the regulator.
Is this an actual PCB with ground plane/pour or just a breadboard?
Beware the test instrument! What you see may be due to how the scope is hooked up. Ideally you would have a totally floating scope to test for noise but unfortunately many can create "noise" through earth loops.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
*Peter*
It makes a freaking difference!
It's not the Propeller, it's not Parallax, it's you, man.
Just to be clear, that capacitor in the schematic you posted is a 0.01uF and not a 0.1uF.
The circuit was 'lifted' from the front end of our BS2 product line and has withstood the test of time in that respect, so any real problems with that particular circuit would have shown up a long time ago.
If you are truly seeing noise on the power rail (even though isolated with the exception of ground) then a diode capacitor filter before and after your regulators would help to filter that out.
Diode/capacitor regulator filter
...but I suspect, your 'resets' are coming from having the serial cable plugged in while you are testing. The cable itself, or any length of wire for that matter, will act as an antenna and become susceptible to any external RF noise. You might be able to 'harden' the ability for the Propeller to reset by connecting an external pullup resistor between the RESn and VDD. The exact value is uncertain without testing, but too LOW of a value and the reset circuit wont be able to reset the Propeller. Too HIGH of a value and there is little to no effect from what you are observing.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe
IC Layout Engineer
Parallax, Inc.
Yeah , the 0.1uf was my clerical error , the cap I had in circuit was a .01uf but has since been removed. This has helped as before I could not get a download to complete without the chip immediately resetting.
I currently have caps as noted installed on all three regulator circuits yet I have no diodes so that I'll have to try.
I am aware of the random resets that may occur from leaving the download cable attached but these happen without the cable attached and at random. I do have a 10k pull-up on the reset line but as you said there are limitations on how LOW I can go there so I suppose I must look a little closer.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Landon Leigh
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Style and grace : Nil point