Brake light flasher (555 - based) circuit works on breadboard, but not PCB

[domiflichi]

So I'm trying to build a brake light blinker/flasher using the schematics from this guy's website:
http://www.dudley.nu/projects/brake_light_blinker/

I've replicated the best I can (he doesn't provide part#s...just values) on a breadboard, and it works great. I ordered a set of PCBs from OSH Park that he provides files to. I did not modify the PCBs - I just submitted them as they were to OSH Park and I received them a week or 2 later. My problem is that when I tried soldering everything onto one of the circuit boards and testing it out, it doesn't work right.

On the breadboard, it does what it should - the 12V DC light bulb blinks several times, then stays on constant (until power is removed of course). But on the PCB version, when you apply power, nothing happens for 2 or 3 seconds, then the light goes on constant. So there's no blinking at all.

When I ordered the components for this circuit from Digikey I ordered a few extras of everything like I always do. So when I soldered everything on onto the PCB I used the same exact components, so I don't get why this isn't working. I was hoping someone could point me in the right direction. I am far from being an EE, so I don't even know where to start.

If it helps, I posted a brief YouTube video of me testing the breadboarded circuit a couple times, then testing the PCB circuit a couple times:

(P.S. I would have contacted the guy who designed this and ask for help, but I can't find any kind of contact form/info/area on his website)

Thanks!

 

[" Moe " morris shouse]

i have had electronics training at a trade school. but still consider my self as an armature. but two ideas come to mind. 1. check the cheapest / easiest thing first. 2. check for heat. if you find heat thats a good place to start.

 

[domiflichi]

I'm not sure what the cheapist/easiest things to do/check, but I did double-check all the values of the resistors and they check out at the right resistance.

Then I checked temperatures:
Applied power to the breadboard circuit, let it set for about 10 minutes, then took out my cheap IR thermometer. The 555 timer and the mosfet were about in the low 70s (F).
Then I did the same thing to the PCB/problem circuit. Everything was in the mid-high 70s (F). I felt everything by hand also, and nothing was hot. A couple things were warm, but that's it.

Not sure where this leaves me. Would it be the 555 IC?

 

[Audioguru]

Why didn't you show a schematic so we can see what might be wrong? The video showing the dark side of the circuit board is useless.

 

[domiflichi]

Hello,
The schematic is in the 2nd image down of the website I linked to in my original post. I didn't think to embed it here. Here it is:

Hope that helps.

 

[(*steve*)]

There's a reasonable chance that the noisy power supply in a vehicle is upsetting the circuit.

There is a similar thread running at the moment about indicators on a 6V system. A similar solution might help here too.

See here. In your case a 16V zener would probably be the right value.

 

[Audioguru]

The schematic should work fine if the polarity of C2 is correct, its polarity is not shown so maybe it is installed backwards.

 

[AnalogKid]

If the time the board is not blinking is the same as the time the prototype is blinking, then the 555 reset circuit (Q1 et al) might be ok. Next is why the 555 is not oscillating. Remove Q1. The output should blink continuously. Let us know.

ak

 

[domiflichi]

Hi again all,

I'm sorry, but I forgot to mention in my original post that the way I'm supplying power to the breadboard circuit and the PCB is I'm using a lab bench power supply, so neither circuit have been hooked up to my motorcycle yet.

I thought C2 is non-polarized though? That's what I bought and have in the breadboard and pcb circuit. Now, I do have a polarized capacitor in C3 however, even though the schematic calls for a non-polarized cap (at least that's what I thought it's supposed to be). That one I do have the right way on the breadboarded circuit, and I'm pretty darn sure I put it in the circuit the right way (there's continuity between the 'negative' side of the cap and the ground wire.

I'll have to remove Q1 tomorrow. But do I just remove it, and then try it straight away after that, or do I have to do something else after I remove it before I can power it up and test it?

 

[davenn]

Now, I do have a polarized capacitor in C3 however, even though the schematic calls for a non-polarized cap (at least that's what I thought it's supposed to be).

show a photo of your construction so we can see C2, the 1uF clearly

any other problem going from breadboard to PCB can only indicate a wiring error ... pretty much no other choice.



Dave

 

[73's de Edd]

Sir domiflichi . . . . .

Consulting your supplied schematic . . . conveniently repeated . . . just above

What I am gleaning from your demo of the stuffed board is:

ALL of the 6 components, and their connections, on the LEFT of the schema and their connecting into the 555 pin 4 are being good.

Because you can see their time constant charge up / delay, until the lamp is THEN turned fully on by their produced change of state at pin 4 of the 555 ..

The lamp, having turned on, after that initial wait, verifies the pin 8 and 1 power connections into the 555 are present and good as well.

Also, the output state of the right half of the schematics pin 3 of the 555 and the 3 components over there are functioning properly and as expected.

That then leaves the most suspect . . . . as being the trace and component path of . . R2 . . .1K with a RED down to R1 . . . 100K with a YELLOW to C2 poly/paper type to ground and the small jumper wire needed at pin 2 to 6.
Meanwhile, ‘ole C1, over there all by hisselfs says “ Don’t fergits me TOO ! ”.

These would be the parts AND INTERCONNECTING FOIL PATHS associated with producing the 555’s few short lived flashing actions.

Can we additionally get a gooood close up of the foil side of the PCB to check out the possibility of there having been an error in “ goober “ files transfer, with a foil path error possibly being made.

That’s all that comes to mind . . . at this very instant . . . .

73’s de Edd

 

[AnalogKid]

If the pcb layout is correct, simply removing Q1 with no other changes will turn the circuit into a continuous oscillator. If not, the problem is in the 555 oscillator circuit (or a bad 555).

C2 does not have to be polarized; it could be a ceramic or foil capacitor. But if it is polarized then the + end goes to pin 2.

ak

 

[Audioguru]

The transistor has nothing to turn it off, especially if the power was removed then turned on again. Fix it by connecting a 22k resistor from the base of the transistor to ground.

 

[domiflichi]

First of all,

I want to thank all of you for all your info/responses. I really appreciate it, and hope my newbiness is not driving anyone nuts.

Ok, so I removed the 'Q1' transistor from the PCB circuit, applied power, and nothing happened. Then I tried removing the 'Q1' transistor from the breadboarded circuit (just to see what would happen), and the bulb just immediately started flashing and never went on solid. Between this difference between breadboard and PCB and from what you guys are saying, it sounds like we're getting somewhere...

But then before I put a new transistor back on the PCB, I added a 20k (closest one I have to 22k) to where the base (middle pin) of the transistor would be, connected to where the 'E' pin on the transistor would go to (ground ). (See photo #14) So I applied power, and nothing happened.

Then I put a new (same one as before and the breadboard's) transistor back on, tested it, and same problem (which I guess was expected).

Not sure if this even helps at all, but another thing I did:
Reversed the polarity on breadboard C3 one way, tested, and the light just constantly flashed. Reversed the polarity on breadboard C3 another way, and the light just immediately turned on solid, didn't change.


#1 - Bare PCB - back side

#2 - Bare PCB - front side

#3 - Populated PCB - front side

#4 - Populated PCB - front side

#5 - Populated PCB - front side

#6 - Populated PCB - front side

#7 - Populated PCB - front side

#8 - Populated PCB - back side

#9

#10

#11

#12

#13

#14 - Temporary resistor in place of transistor

Well, hopefully this helps you guys help me. I was thinking since I have 3 boards, maybe I should try to make another one and see what happens? I don't know...I'm very tired so maybe I'm not thinking clearly at the moment.

(And thank you everyone again for helping me out)

 

[(*steve*)]

Images 2 thru 12 do not appear. Did you upload them?

OK, it looks like you linked to them. It is better to upload them and show the thumbnail rather than the full image.

 

[domiflichi]

Yeah that's strange - when I was writing the post, all the images were showing just fine.

I'm just going to upload all the images then.

 

[(*steve*)]

If removing Q1 doesn't make the output flash, but changing C3 does, then I suspect some misconnection.

On an unpopulated board use a multimeter to confirm continuity between points that you expect

The dark solder mask did not make it easy for me to make out the tracks on the small screen of my phone.

 

[Audioguru]

Measure the voltage at pin 4 (reset pin) of the 555. The 5.1k resistor R4 to the positive supply should make it at the positive supply voltage to allow the 555 to cause the light to flash if the transistor is turned off or is removed. The voltage will be 1V or less to cause the 555 to be reset without flashing the light.

 

[73's de Edd]

Sir domiflichi . . . . .


Ohhhhh . . .Woe is me . . . Nutz to them . . .shoulld they have chosen to use ANY other color than BLACK and Deeeeeeeeep Purple; I could have chroma filtered for acquiring better board to foil differential enhancement.

But I do LUUUUUV that 23 and 59/71ths carrot feed thru and plated holes that they done gaves you !

Now, between the top snippet, confirming power into R2, and the bottom photo covering all of the other interconnect aspects of :

Connecting 12 volts B+ to R2
Then connect R2 to R1 and 555 #7
Then connect R1 to 555 #6

Then they had the boards foil path automatically
doing your jumper path between 555 #6
and 555 #2.

Then 555 #2 connects C2 to ground.


One separate side path, is pin 555 pin 5 To C1 to ground which looks OK.


That constitutes the critical foil path connections, with you only having to confirm my drawn in GREEN marked path.

I still suspect:

ALL of the 6 components, and their connections, on the LEFT of the schema and their connecting into the 555 pin 4 are being good, inclusive of the transistor.

Because you can see their time constant charge up / delay, until the lamp is THEN successfully turned fully on, by their produced change of state at pin 4 of the 555.


Going back to that string of components from 12VDC to R2.. t.. h.. r.. o.. u.. g.. h C2 going to ground, can you confirm the short foil path that is marked in GREEN ?

Otherwise, all looks good on the boards foil paths.

Can you now check that complete flow path with an ohmmeter connected to component leads ?

BOARD vs SCHEMATIC . . . COMPARED

https://s17.postimg.org/b8818efr1/555_Timer_wiring.jpg

ASIDE:
Would you believe that I done a variant of this "Dole Stop Light " on my Rincon Contirental sedan
waaay back in the 70's . I had harvested the freebie flash circuits from 2 disposable film cameras.
The units got mounted in the tail light assemblies and were fed reduced keep alive voltage and one steering diode to trigger the flash .
When the brake light was initially activated,there was this BRILLIANT red flash and then the conventional bulbs illumination was able to be seen again.


73’s de Edd

 

[73's de Edd]

Sir domiflichi . . . . .


In a proof re-read I see that I omitted poking in additional thoughts of using a DVM in low voltage DC mode.

You set up the breadboard model and initially monitor 555 pin 4 for its initial voltage just after powering up the stage and then after its time out and state change and record both and then do the same testing with the PCB build up.
I fully expect both to be basically the same.

Then you use the working bread board model first.

On it, you do the same two changed state test, but this time you monitor:

• 555 pin 7

• 555 pin 6 or 2

• 555 pin 5

Realizing that there will be a variance during the blink cycles . . . but steady after the . . .no flash . . .changed state condition.
That should come up with six readings.
Move to the PCB model and do the same test to see what its read out voltages then might reveal to us.

73’s de Edd