Brake Pulse/Flasher Circuit

Status
Not open for further replies.

Thread Starter

doug3460

Joined Oct 19, 2008
87
Added next segment. Replaced trimmers + resistors with single resistors (R12,R13) at about midway point ([(50k/2)+10k)] from Ken's original) Not sure how to get pin 14 on the 4098 to connect properly. Couldn't figure out how to get a "package' to display so could just connect the pins, & since unfamiliar with design technics in general, not sure if it's done with the gates as SOP anyway.:(
 

Attachments

SgtWookie

Joined Jul 17, 2007
22,230
Added next segment. Replaced trimmers + resistors with single resistors (R12,R13) at about midway point ([(50k/2)+10k)] from Ken's original)
OK, but keep in mind that when you're testing the completed assembly, you'll still use pots in those positions to determine what values of fixed resistance to use. It doesn't matter what type the "test pot" is. The point is, you can find out what fixed value you need a whole lot quicker by using a pot than if you tried manually soldering in a bunch of different resistors.

I suggest that you change the "Value" fields to something like "0k-50k" to remind you that pots go there during testing.

Something else - in the original schematic on Tony van Roon's site:
http://www.uoguelph.ca/~antoon/circ/motflash.html
note that R2 and R5 are there to provide a minimum resistance of 10k Ohms. You must keep this in mind when you are "tweaking" the board after assembly. CMOS logic IC's have very limited current source/sink capability; if you try to use too low of a resistance (load) on their outputs, you will very likely fry them.

In your board, add a 2nd resistor in parallel to both of those that are supposed to have pots, and give them a Value of 99MEG . This will help a great deal when you want a non-standard resistance; you can wire two standard resistance values in parallel to get very close to where you want to be. The 99MEG is just a reminder that they're only there if needed for trimming; otherwise they'll be left off.

Here is a handy page that has a serial/parallel resistance calulator:
http://www.qsl.net/in3otd/parallr.html

Not sure how to get pin 14 on the 4098 to connect properly. Couldn't figure out how to get a "package' to display so could just connect the pins, & since unfamiliar with design technics in general, not sure if it's done with the gates as SOP anyway.:(
Sorry for the confusion; that's my fault for doing a crappy job of explaining how that works. :p I alluded to it in my replies #13 and #14, but never explicitly spelled it out; so here goes:

Many IC's in the Eagle libraries don't show their power/ground pins in the schematic; they are implied by type. For example, all of the 4000 series CMOS logic IC's have implicit VDD and VSS pins. If the only thing your schematic consists of is 4000 series logic, all you need is a single VDD and a single VSS supply symbol somewhere on the schematic. When you create the board, there will be "air wires" running between all of the VDD and VSS pins on the IC's, indicating that you need to run traces between them to connect them all. It's your responsibility to connect the VDD and VSS symbols to something like wirepads or off-board connectors to the source of power/ground.

Having implicit power/ground pins helps to avoid lots of clutter that would otherwise be present on the schematic. This helps one concentrate on the logic of the circuit, rather than getting distracted by the power and ground connections.

The problem with this scheme is when you are using IC's that have different names for implicit power; right now the 555 timer has V+ and GND for it's VDD/VSS. If the 555's power pins were implicit rather than explicit, we'd have to fiddle around using zero-Ohm resistors or jumpers to connect a V+ symbol with the VDD symbol and another to connect VSS to GND. You can't simply place VDD and GND on the same wire, or you will wind up with ERROR messages that will cause problems in the board (wires that don't show up, leaving critical pins disconnected).

You'd run into the same kind of problem if you tried to mix TTL (54/74 series) and CMOS (4000) series on a board; where CMOS uses VDD/VSS, TTL uses VCC and GND. Sometimes this is good; where you can run CMOS at 3v-16v, TTL requires 4.75v-5.25v for reliable operation.

Which reminds me - we should add a resistor and 15v Zener diode to the supply circuit just in case your regulator goes whacko and decides to dump 20v into your electrical system, which would turn your circuit into something resembling a small volcanic island. :eek:

OK, back to your circuit - Remove the wire from R7 to IC3D-14, and the VSS symbol you tacked on to IC3B.

Something else that hasn't been mentioned before; when you're using IC's, you should ALWAYS have a "bypass capacitor" across the power/ground terminals. These are typically 0.1uF (100nF) tantalum or ceramic capacitors. They take care of transients, or "glitches" in the supply voltage when the IC changes states. 555 timers have LARGE glitches when they change states, because for a very brief time both the high side and low side output transistors are turned on, which creates a dead short across the supply pins. So, a larger bypass cap is needed for the 555.

You are missing a junction above C4. Without that junction, the brakelight would instantly come on steady, as C4 & R12 would not be connected to the input of IC3A.

There's a bit of a problem with the input to IC3A that I'm still "noodling" over; what's the best "fix" for it. The trouble is that when C5 becomes charged or discharged and the output of IC3D toggles, the instantaneous voltage on the input of IC3A becomes either roughly double that of VDD or VSS-VDD, whichever is the case. Although R14 limits current, a CMOS input has virtually infinite impedance, so IC3A's input can be subjected to voltages well in excess of it's ratings. This kind of abuse tends to punch holes in CMOS gates, destroying them. :( A pair of clamping diodes is one fix I'm considering.

At this point, it would be a good idea if you started attaching your schematic after you make revisions; that way I can more easily double-check it for you. This will become even more important when board layout begins; attaching both the schematic and the board - otherwise, things could get out of hand pretty quickly. Besides, if you upload them here and your computer crashes or manage to completely screw your schematic/board up, you have an off-site backup :)

Also, when you've made several changes, run ERC first, and then "File/Save as..." with a new revision number/letter on the tail end of the filename. You'll wind up with a bunch of versions, but if/when you screw things up, it can be a lot easier to step back a version or two rather than having to throw everything out and start over.
 
Last edited:

Thread Starter

doug3460

Joined Oct 19, 2008
87
SgtWookie - I'll just put the pots back in w/ a 99 meg resistor. For the prototype it's fine & then will replace when doing the final. Will also make your other corrections. Concerns I have:

I think the junction between R5 & R6 creates a short (I think) between them so when the 4017 is pulsing the lines are no longer seperated. Placing a diode there only fixes the problem one way. I read something about a component that has rapid 180° switching (image is of a triangle I think) - is that type of oscillation what's needed? This is why I was thinking about putting in the logic gate to handle the: ((A+B)xNOT C)+(CxNOT(A+B)) but I haven't finished figuring that one out either.

How do I show on the schematic the connection or relationship of 4017Q6 to 4093pin14?

Apologies if this is off target - I haven't had my coffee so my brain isn't up & running.;)
 

SgtWookie

Joined Jul 17, 2007
22,230
SgtWookie - I'll just put the pots back in w/ a 99 meg resistor.
In parallel, if you would please ;)
For the prototype it's fine & then will replace when doing the final. Will also make your other corrections. Concerns I have:

I think the junction between R5 & R6 creates a short (I think) between them so when the 4017 is pulsing the lines are no longer seperated.
I didn't even notice that. I had to go back to my schematic to check that out. You did in fact muck that up. As a result, the lights would be on constantly.
Placing a diode there only fixes the problem one way. I read something about a component that has rapid 180° switching (image is of a triangle I think) - is that type of oscillation what's needed? This is why I was thinking about putting in the logic gate to handle the: ((A+B)xNOT C)+(CxNOT(A+B)) but I haven't finished figuring that one out either.

How do I show on the schematic the connection or relationship of 4017Q6 to 4093pin14?

Apologies if this is off target - I haven't had my coffee so my brain isn't up & running.;)
When are you awake/asleep, in relation to EST? Or does it vary considerably?

Not wishing to take backwards steps, but perhaps it's time to again examine exactly what is to be accomplished here.

The circuit from Tony van Roon's site could be modified to flash both a brakelight, and to do the "wig-wag" thing at the same time, with the addition of a few transistors. IF you don't mind that they are all occuring at the same time, of course.

Brakelight on, and either left or right "wig wag" light on - then brakelight off, the other "wig-wag" light on - cycle back and fourth for perhaps a second.
 
Last edited:

Thread Starter

doug3460

Joined Oct 19, 2008
87
Concept is 6 pulses alternating left/right (the 4017 circuit), then 3 or 4 pulses both lights, then steady. Please see the Signal Dynamics Wig-Wag (it shows a display - I don't think posting a link is with site rules).

My plan was for a different version than theirs for my use only. Mine includes the option to by-pass segments or the entire circuit. Of course the final schematics would be available here for the adventurous.

My sleep schedule is quirky because I do the night shift (8 years now - the life of a vampire). I usually sleep between 0830 & 1300 on work days. On off days, it's pretty much catch what sleep I can.

Simply modifying Ken's circuit would be nice (KISS) if we can achieve the desired result. I've attached the original concept flow.
 

Attachments

Thread Starter

doug3460

Joined Oct 19, 2008
87
I've added the Zener & resistor to Vdd per your recommendation. I routed IC2Q6 to a pad labeled "to Pin 14" for IC3. Returned the pots & resisters (in parallel ;)) & relabeled them. Relabeled the TIPs to 137s. Removed the short. Added a 4070 even though I only needed two gates, but had to have the exclusion logic.

:( Am disappointed in the layout of the schematic as it gets towards the amplifiers, but not quite sure how to rearrange it.

Got called in for work on Sunday, so may not get time to work the safety relay into the output lines. Will get to it.
 

Attachments

Last edited:

Thread Starter

doug3460

Joined Oct 19, 2008
87
SgtWookie - a couple things:

...when you're using ICs, you should ALWAYS have a by-pass capacitor across the power/ground terminals...
I realized after I posted the latest schematic that I had forgotten to add the 0.1uF capacitors to the remaining ICs.:mad:

There's a bit of a problem with the input to IC3A...
I also forgot the "clamping diodes" on IC3A.:(

As far as IC4 (the 4070), I think I'll need a boosting transistor there also. I get a little bewildered with the data sheets at times. I was considering going with the 74series XNOR, but they need 5V to trigger. I guess a Darlington could handle it.:rolleyes:

That triggering boost is why I tied the net into "PAD1 To pin 14" where I did - I wanted to catch the boosted current on it's way to IC1 pin4 (reset). Now that I think of it, I probably need a resistor on that line too.

I realize now that's it's more complex than I originally thought, but I am learning a lot. Thanks for your help & to all the folks who made the AAC tutorials & run the site.:)
 

SgtWookie

Joined Jul 17, 2007
22,230
Sorry I'm slow in responding; we had a very busy weekend at the National Vietnam War Museum; had a very successful fundraiser/bike run to Daytona/auction/lunch/etc with hundreds of bikers on Saturday, and multiple functions on Sunday. Exhausting but worth it.

With your new input, I think a simplification might be made for the wig-wag circuit; add more diodes on the 4017. The less complex this thing is, the better.
 

Thread Starter

doug3460

Joined Oct 19, 2008
87
SgtWookie - Not sure what you meant by:

I think a simplification might be made for the wig-wag circuit; add more diodes on the 4017.
:confused:

Also, as I've been adjusting the schematic, I started playing with the board. Guess I'm doing something wrong with the set-up since it makes multiple layers. I tried to simply duplicate Ken's original in a seperate layout so I could practice, and even there it's layering.:(

Any suggestions?
 

Thread Starter

doug3460

Joined Oct 19, 2008
87
I've attached the Eagle schematic & board for the Moffett/Roon circuit I redrew. I can't get the board to look like the one at Tony's site (no kits are available while he's moving). For some reason, the Vdd & Vss airwires aren't showing up. I tried adding some to the board itself & Eagle didn't like that. When I was playing with the board for my project circuit, the airwires show up.:mad:

What am I fat fingering?:confused:

OT: I'm going to go get parts/supplies today. I need to get a breadboard. Limited budget, but was thinking for these small projects, a mid-size board would be okay & a jumper wire kit. Suggestions/recommendations?
 

Attachments

Thread Starter

doug3460

Joined Oct 19, 2008
87
Ken - Super! And thanks so much...

Pleased to hear yours was hand wired - I intend to build yours & install it to use while I continue to work on the other one. Out of embarrasment I didn't want to post that I was going to have to hand wire it since I don't have the PCB making ability (yet) and felt it was too low-tech for this site. :p

Doug
 

KMoffett

Joined Dec 19, 2007
2,918
Doug,

I do most of my small one-off projects hand-wired. I use ExpressPCB to generate the schematic and verify the layout of the components/traces...then place the components and just duplicate the top and bottom traces on the bottom of the isolated pad perfboard. Except for higher current traces, which I place first with appropriate gauge wire, I use #30 Kynar wire-wrap wire. If I decide I want to do more of the same, I can ship off my files and have ExpressPCB make them. If I need components to fit the 0.1"x0.1" perfboard footprint, and they aren't in the library, I just modify existing ones, and add them to my custom components library.

Ken
 

Thread Starter

doug3460

Joined Oct 19, 2008
87
Attached the file for the next segment. Included the 0.1μF capicitors to the ICs needing them (IC3 & IC4). Hope they're in the correct spots. I was unsure about IC4's: the input is off IC3, so I wanted to ensure that signal was 'smoothed,' but since the unused pins of IC4 have to be tied to Vdd, didn't know if it should go w/ that resistor. :rolleyes:

Reading up on relays to add the safety (by-pass) circuit next. Final thing will be the selector switch. Hope to have it on a board later today.:eek:
 

Attachments

Thread Starter

doug3460

Joined Oct 19, 2008
87
SgtWookie, Ken, one & all...lol

Please take a gander @ the attached. I had a couple last minute (hours) SNAFUs:

I wasn't sure where to hook in the lamp to discharge Cap1 to the IC1 (the 555) when the circuit goes off. Sarge's comment was simply across VDD & VSS, but is it supposed to be lit all the time? If so, can any VDD/VSS lines be used? Can I set it to light off the reset VDD line only so it lights only when C1 needs to dischg to reset the circuit?

I decided not to include a snubber on the selector switch. The board goes under the seat, so rider & power have to be off to make a change in the switch selection.

I have tried but have not been able to figure out the way to make the safety by-pass using a NC Relay. The relay needs to be tied into the switch for the manual circuit by-pass, as well as be able to activate both outgoing lines to the dual brake lamps.

Appreciate your time for review & comments.
 

Attachments

Thread Starter

doug3460

Joined Oct 19, 2008
87
Sorry it's been a bit since I was able to update. "Life is what happens while we're busy making other plans..."

Couple quick questions I haven't been able to find answers to on the site:

1. Can I power my bread board w/ a 12V, 500mA Wal-Wart I have for charging a sealed 12V battery? Does such an item provide "clean" source for general experiment/development of the automotive type circuits?

(Bear w/ me - I haven't done all the math yet for these circuits but given their general nature, I thought 500 mA might work & I don't have the $$ at the moment to get the fancy digital one I want :().

2. What is the recommended way for placing items in 220 cases on the board? The 92 cases are easy enough to bend the legs, but a little more difficult for the 220.

3. Before jumping in with the circuit we've been working on, I was going to use Bill Marsden's 555 projects & some others from Chap 6 to build up to the more complex circuits. Are there other sources that display the board layouts that I should be looking into?

Thanks, as always, to all for your time & patience as I "learn to fish." :D
 
Last edited by a moderator:
Status
Not open for further replies.
Top