kontiki timer 08m2 supply voltage issue

S

steviejay

Member
Hi

After several attempts and lots of guidance thru the forum I have an ongoing issue with timer for my kontiki(baitlauncer torpedo)resetting. It works fine until its under load in the water then cuts out randomly. To test I fitted and independent 3v CR2032 button cell to the timer side(theres no room for anything bigger) and it went all the way no problems. It seems the motor current draw under load leaves the regulated side short maybe. I beefed up the capacitor to 2200uF as suggested here, included snubber across the relay. Attached is the schematic and code if I can get it into the right place.
There must be a way to maintain the supply as there are others on the market that work from the 12v supply
I did NOT do anything with the brownout setting,
Any suggestions appreciated
Steve


Code: 
setint %00001000, %00001000,c

main:
sertxd ("back at main ",13,10) 'message to PC

pause 50
high c.4		'fast flasing LED alerts ready mode
pause 50
low c.4
pause 50
high c.4
pause 50
low c.4
pause 50
high c.4
pause 50
low c.4
pause 50
high c.4
pause 50
low c.4
pause 50
wait 1
sertxd ("reading reed switch",13,10)
if pin1 =1 then fxtimerreadyA   
if pin1 =0 then goto main   
goto main

fxtimerreadyA:	'remove magnet fxtimerA will activate or keep it and move to fxtimerreadyB
for b0 = 1 to 5	'use b0 variable memory space for a loop flash 5 times		  
high C.4		'output pin C.4 on	
pause 250		
low C.4			
pause 250			
next b0
sertxd ("reading reed switch",13,10)
if pin1=1  then fxtimerreadyB  
if pin1=0 then goto fxtimerA  		
				
fxtimerreadyB:	'remove magnet fxtimerB will activate or keep it and move to next timer
sertxd ("in fxtimerreadyB ",13,10)
for b3 = 1 to 5	'loop the LED	  
high C.4			
pause 500		
low C.4			
pause 500			
next b3
sertxd ("reading reed switch",13,10)
if pin1=1 then fxtimerreadyC  
if pin1=0 then goto fxtimerB  

fxtimerreadyC:
sertxd ("in fxtimerreadyC ",13,10)
for b5 = 1 to 5	'use variable b5 to loop LED 	  
high C.4			
pause 1000		
low C.4			
pause 1000			
next b5
sertxd ("reading reed switch",13,10)
if pin1=1 then goto main  
if pin1=0 then goto fxtimerC  

fxtimerA:		'the first timer programme
sertxd ("in fxtimerA mode ",13,10)
high c.4		'LED stays on
high c.2		'activate transistor and so relay and motor
for b1 = 1 to 10 	'10 loops 
pause 60000 	'wait 60 seconds 
next b1 
wait 1
low c.4
low c.2
sertxd ("finished timer ",13,10)
pause 500
goto main		'go back to main

fxtimerB:
sertxd ("in fxtimerB ",13,10)
high c.4
high c.2
for b1 = 1 to 20  '20 loops 
pause 60000 	'wait 60 seconds 
next b1 
wait 1
low c.4
low c.2
sertxd ("finished timer ",13,10)
wait 1
goto main

fxtimerC:
sertxd ("in fxtimerc ",13,10)
high c.4
high c.2
for b1 = 1 to 25 	'25 loops 
pause 60000 	'wait 60 seconds 
next b1 
wait 1
low c.4
low c.2
sertxd ("finished timer ",13,10)
pause 500			
wait 1
goto main

interrupt:		
sertxd ("interrupting ",13,10)
if pin3 = 1 then interrupt	'if reed switch on pin3 is high(on)
low C.2
low c.4
wait 2
setint %00001000, %00001000
reset
goto main


timer schematic with 08m2.jpg
 
B

bluejets

Senior Member
I think you will have to supply some more detail on battery size and type and motor draw current under load.
 
M

MPep

Senior Member
Simple: change the relay +ve feed to the other side of the diode.
This way you've got a diode to charge the timer circuit but the capacitors cannot discharge when the motor load is on.
 
Bill.b

Bill.b

Senior Member
The 10k resistor in the first reed switch should goto ground not directly to c.1. should this not be the same as the second reed switch.

there is no ground reference for this switch. In the open position c.1 will be floating.

Bill
 
S

steviejay

Member
08m2 timer as built notated.jpgtimer schematic with 08m2corrected.jpgthanks everyone
I do have the pulldown resistor in ok, its drawn incorrecty sorry..(Visio has a mind of its own), and I also have the diode AFTER the relay + feed..and so glad you noticed Ive simplified to an 08m2..less space just different flash rates instead of 3 LEDs
attached corrected schematic and as built photo

if it helps...I did anothe rlaunch today and it ran the full timer..on the 3v separate supply
and also..when I first had new batteries the timer functioned...as theyve worn in, say 20-30 charges Ive had this issue

thanks again
Steve
 
O

oracacle

Senior Member
what happens if you put a fairly large cap across the motor pins?

it would be very handy if we knew a bit more about the bettery that supplies the power, the more about the motor, at least a coil resistance, V/R=I will give the current draw to see if you are drawing too much juice from the battery.

to give you an idea i have a 2.5ah for my motor and all the circuitry. the motor can draw nearly 2 amps continueos under heavy load, along with nearly half amp for the driver, picaxe and various other thing that are ther
 
R

rossko57

Senior Member
steviejay said:
...but Ive had a 20 amp fuse that hasnt blown...
Click to expand...
That could be because the heavy motor load collapses the battery voltage so much that a less-than-maximum current is drawn. What's the primary battery here? Does the timer supply have seperate wires directly to battery terminals, to minimise volts drop (as opposed to sharing the heavy motor wires)

I rather think your only easy recourse here is bigger capacitors upstream of the 7805, big enough to sustain the timer power for the duration of the motor run.

A little redesign could shift the LED power source to the unregulated side, to reduce the regulated draw by a few mA
 
S

steviejay

Member
thanks
the timer circuit piggy backs onto the motor wire spade connectors that connect to the switch/battery connection batteries are two 7.2Ah 12v
Steve
 
M

MPep

Senior Member
Another suggestion: Run separate power supply cables to your timer circuit direct from the battery.

Please post a photo of what you've got, as well as inform others what a kontiki looks like. :D
 
S

steviejay

Member
kontiki torpedo complete.jpgkontiki torpedo stern.jpgattached the wiring space..and the torpedo(kontiki), not much room for more wires..sliding the batteries in is tight..and the timer has to fit in the tight side space...
thanks for taking the time to think about this
steve
 
A

AllyCat

Senior Member
Hi Steve,

At those currents I'd be looking for "parasitic" components causing the problem - stray capacitance, or very possibly a "common impedance" (resistance and inductance) in the earth rail.

However, if an auxiliary battery definitely does cure the problem, then maybe you can reduce the loading on the PICaxe supply rail, rather than increase the capacitances. The PICaxe (processor) only needs less than 1 mA, but the external components may be draining more than 20 mA (two 330 ohm resistors across a substantial part of 5 volts).

This is one case where a "darlington" transistor pair (or a logic-level FET) in place of the single BD139 would be of benefit (and increase the drive resistor to perhaps 10k). Also, an "emitter follower" transistor (with collector to the battery rail), or a "low-side switch", to take the LED current directly from the battery.

Cheers, Alan.
 
S

steviejay

Member
thanks
bearing in mind the timer worked fine for the first 20-30 runs, then just started cutting out, so rebuilt another, thinking maybe it was the interrupt reed getting vibrated into closing...same problem...Im now onto Mark V

Are you suggesting that not switching the led on at the same time as the transistor may have an effect...?
Ill have to read up on darlington
thanks Alan
Steve
 
A

AllyCat

Senior Member
Hi Steve,

It might just be that the battery voltage is getting lower, so there is less time before the regulator "falls out" of stabilisation when the battery is heavily loaded. So reducing the current drain, e.g. not lighting the LED (if it drains current from any of the supply decoupling capacitors), may help.

There was a very recent thread on Darlingtons, but IMHO wikipedia as in its #3 is usually the best source of information.

Cheers, Alan.
 
You must log in or register to reply here.
Top