Wig wag (alternating) automotive flasher relay for Lionel 154 crossing

Here are spec

A standard 2 prong mechanical flasher for a motorcycle/small equipment with a low amp load may also work.  Best when using Auxillary voltage vs. track on AC.   I did that in a bell-ringing tender that did not have the heat strip.  With an accessory; fit problems are eliminated.  Adding a 2nd accessory for the current draw needed may also help.   If there is not enough current draw it flashes slowly if at all.

Sorry I don't have the id/number for the flasher anymore.

If an old-school traditional automotive flasher based on a thermal strip (no "electronics") I'd think it would work on AC too. In other words maybe it could be directly powered with 14V Accessory AC.

@stan2004 posted:

If an old-school traditional automotive flasher based on a thermal strip (no "electronics") I'd think it would work on AC too. In other words maybe it could be directly powered with 14V Accessory AC.

Agreed, but the timing will likely be off, in other words the flash rate will likely not be the same with AC as on DC for the same voltage (nominally 12 V).  I'm thinking it would need in about 17 VAC to match.

It's worth a try -- if you don't mind the click - click - click when it's operating.

Mike

@Mellow Hudson Mike posted:

Agreed, but the timing will likely be off, in other words the flash rate will likely not be the same with AC as on DC for the same voltage (nominally 12 V).  I'm thinking it would need in about 17 VAC to match.

Curious as to your reasoning.  My reasoning is a traditional bimetal thermal strip warms up through resistive heating, bends to make/break electrical contact, cools, bends back to break/make electrical contact, lather-rinse-repeat.

A resistor is agnostic to whether the applied voltage is AC or DC.  So 12V DC and 12V AC should heat it equally well and the timing should be identical.

Based on the somewhat "loose" specs, I do think this is an old-school thermal strip type flasher.  It is almost inconceivable that an electronic (digital) implementation would have a flash rate that might be from 60 - 120 flashes per minute...and have a duty-cycle that might be from 35%-75%.

I would fully expect an "electronic" implementation to require DC voltage.

@dogdoc posted:

...Could get a bridge rectifier but cost goes up.

If the purpose of the bridge rectifier is to convert AC to DC...then the output of the bridge rectifier will not be "flat" constant DC like you'd expect from a battery.  The DC would be "pulsed" DC with peaks and valleys like the original AC voltage.  I can't imagine this would change the behavior of a thermal bimetal strip flasher...other than you'd lose about 1 Volt going thru the bridge.

I realize it's easy for me to say to just hook it up to train transformer accessory AC and see what happens.

Separately, while not as simple to hook up as the 3-prong wig-wag automotive flasher, I've previously posted many ways to wig-wag flash the 154 if you're comfortable with some DIY wiring and such.  For example, this OGR thread had the following video for a 154 flasher for less than $10.   It's digital so you can program the flash rate to EXACTLY 60.0 flashes per minute...or 60.1 flashes per minute...etc.  Or you can program the duty cycle to be EXACTLY 50% so that each lamp is on exactly half the time.  Of course you still have the issue of the clicking relay sound though the automotive flasher presumably has a similar clicking sound.

@stan2004....

Your circuit demo videos are the best! If a picture is worth a thousand words, your videos are worth a million!

I have one on the way to try for the heck of it. I will wire it up on my test station to try. If I can figure out how to post a short video of its function I will.
The flasher Stan posted looks fun to make as I like modest electrical projects.I may try that as well

@dogdoc posted:

I have one on the way to try for the heck of it. I will wire it up on my test station to try. If I can figure out how to post a short video of its function I will.
The flasher Stan posted looks fun to make as I like modest electrical projects.I may try that as well

The flasher is an interesting approach to the wig-wag.  If you do happen to try both ways, I'd be curious to hear your opinion about which "clicker" is quieter. My guess would be the relay in Stan's design.  Those bi-metal type flashers can be kinda noisy IMO.

I take it astable multivibrators are no longer in fashion?

@SteveH posted:

Those bi-metal type flashers can be kinda noisy IMO.

Yes.  They're required to be by law, Federal Motor Vehicle Safety Standard #108 (FMVSS 108):

Mike

@Mellow Hudson Mike posted:

Yes.  They're required to be by law, Federal Motor Vehicle Safety Standard #108 (FMVSS 108):

Mike

That makes sense, we wouldn't want people driving in the left freeway lane with their left turn signal on for miles...

@Mellow Hudson Mike posted:

Yes.  They're required to be by law, Federal Motor Vehicle Safety Standard #108 (FMVSS 108):

So I tried to look up what the requirement is that a flasher be "noisy" in what I think is FMVSS 108.

Very lengthy document so I tried searching for words like "audible" and "decibels" and similar terms, but I could not find anything referring to how "loud" a visual flasher mechanism must be.  Can you elaborate what you are referring to?

I am thinking of how quiet electric cars synthesize some kind of sound to warn pedestrians, or I can imagine a Federal requirement for how loud or noisy a backup warning sound must be for commercial vehicles and the like.  I'm curious why a wig-wag flasher mechanism for a vehicle must be "noisy"!

I found this in the FMV code Mike cited near the bottom of the page (after doing similar page searches for SPL, decibels, noise, sound, audible, loud, etc. with no success):

It references SAE standard J590b.  When I searched elsewhere on the web for that information, there was a hefty fee to download, so I let it go.

I think the main reason for the "noisy" requirement is so that there is an audible indicator as well as a flasing light on the dash to remind drivers that their turn signal is activated.

Or they are simply due to the technology used?

@dogdoc posted:

Or they are simply due to the technology used?

No, there's definitely a requirement to have a certain level of noise, i.e. the click, so that a driver knows that the turn signal is on without taking their eyes off the road.

Today we do it electronically.  Most of the time it's a digitized click (just like RailSounds or ProtoSound 1/2/3 using a digitizing whistle/horn, bell and/or chuff), although sometimes a tone, but even though it's electronic and modern the amplitude in decibels (db) must still match the requirement in FMVSS 108, from the early 1960's.

I'll see if I can find the exact requirement and bring it back here.

Mike

@A. Wells posted:

I take it astable multivibrators are no longer in fashion?

There have been many OGR threads on the astable multivibrator - most revolving around this "classic" circuit:

For whatever reason, on OGR this approach seems to fall under the DIY umbrella for guys that don't mind working at the component level, soldering, and so on.  @Rod Stewart posted a project based on this classic astable multivibrator with circuit board, component list, etc.

There is no relay so it is silent.

And as I'm sure you're all aware, this classic Lionel accessory has had a built-in flasher circuit since the late 1990's.

Jon

@dogdoc posted:

Here are spec

Looking at the spec and wiring diagram, it looks like all the current through the thermal element also passes through the lamps, as there is no separate ground connection to the flasher. The spec calls for a minimum of one 1156 lamp (around 2A). It is possible that the small lamps in the 154 will not draw enough current to make the flasher work, but please try it and report back!

Just my $.02.

@SteveH posted:

...

I think the main reason for the "noisy" requirement is so that there is an audible indicator as well as a flasing light on the dash to remind drivers that their turn signal is activated.

I'm still struggling with why the "noise" spec would apply to the flasher mechanism itself...as opposed to a spec on how loud the indicator must be at the driver's ear.  That is, car manufacturers go to extraordinary lengths to make interiors quiet and insulated from road or engine sounds.  So to specify the loudness of a clicking relay for a driver to hear seems odd since this means the relay must be mounted near the driver, say, behind the instrument panel which seems unduly restrictive.  In the specific case of a wig-wag alternating flasher, as I see it you would have to run 2 "hot" wires from the relay to, say, the rear of the car where the 2 alternating loads are located.  You would save a lot of wire (and weight which is gold in a car) if the relay was located near the bulbs since only 1 "hot" wire needs to run from the on/off control switch to the relay located near the 2 alternating loads.

The spec dates from the 1960's, so back then interior quiet was not as much of a concern as it is now with modern vehicles.

As for the location of the flasher mechanism, a central location under the dash where a faulty unit can be somewhat easily replaced used to be the norm.  Also since there is only one flasher mechanism and at least 4, often more, signaling lamps at the front and rear of the vehicle, a central location does somewhat reduce wiring. I'm also assuming, at least in older vehicles, that the lamp selection switching is done in the turn signal and hazard switches.

I agree that having the flasher mechanism closer to the lamps would further reduce wiring, however maintaining flash synchronization between the front and rear lamps would be more involved than a bi-metal flash control device might simply allow.

@stan2004 posted:

There have been many OGR threads on the astable multivibrator - most revolving around this "classic" circuit:

For whatever reason, on OGR this approach seems to fall under the DIY umbrella for guys that don't mind working at the component level, soldering, and so on.  @Rod Stewart posted a project based on this classic astable multivibrator with circuit board, component list, etc.

There is no relay so it is silent.

The irony in this is I received my package of 2N3904's from Mouser yesterday.  Not for multivibrator purposes but for a random number generator.

They do make electronic flashers to retrofit the old electromechanical ones when changing to LEDs in classic cars. I've used the in my 1969 Corvair as the LEDs don't draw enough current to operate. They do have an additional ground wire. The Corvair had two flashers, one stuck high up under the dash so you could hear the clicking.

@Scotie posted:

...The Corvair had two flashers...

Maybe one for turn signal (2 lamps) and another for hazards (4 lamps = twice the load) so that in either case the flash rate is consistent with FMV and SAE specifications.  This would also maintain synchronization among the lights.

I remember those old flashers well. I would replace the standard flasher with a heavy duty one in customers vehicles that pulled trailers with them.  This would stop the turn signals from flashing so fast.

As long as you don't exceed their max the electronic ones do not change rate with load, that's one reason for changing them when you switch to LED turn/parking/hazard lights.

Wig wag flasher works well but buzz evident. As loud as my layout is I don’t think it will matter under the layout. May could insulate it in something.

@dogdoc posted:

Wig wag flasher works well but buzz evident. As loud as my layout is I don’t think it will matter under the layout. May could insulate it in something.

Interesting, thank you for reporting back.

The buzzing is REALLY interesting.  I couldn't believe my ears as it sounded like line-frequency buzzing of old-school O-gauge vibrating/buzzing accessories.  Could the flasher relay actually be buzzing from the AC frequency?

So I analyzed the audio from the posted video and sure enough, it is buzzing at 120 Hz or what appears to be every "peak" of the AC voltage.  I can't believe it!  That means the relay's wiper or reed is wiggling 120 times per second from variations in the voltage?

When you hear the buzzing, the right lamp does NOT fully turn off.  That suggests the relay wiper is banging against the contact 120 times per second providing intermittent power to the right bulb and hence that's why is does not completely turn off.  On the other half of the cycle (when there is NO buzzing), the left lamp completely turns OFF!

Someone explain to me what the ?&$#@ is going on here!

It's hard to believe that a bimetallic strip can wiggle (heat and cool, bending back and forth) 120 times per second.

@stan2004 posted:

Someone explain to me what the ?&$#@ is going on here!

It's hard to believe that a bimetallic strip can wiggle (heat and cool) 120 times per second.

Yes, I caught that too. The manufacturer's catalog page tells the story, that is an electronic flasher, not a thermal flasher. They do not offer a thermal wig-wag.

Because the specified voltage drop is only 500 mV, they are apparently running the low-current electronics (including a relay) through the unlit lamp. Because the lamps in question are so small, they try to light anyway.

@PLCProf posted:

Yes, I caught that too. The manufacturer's catalog page tells the story, that is an electronic flasher, not a thermal flasher. They do not offer a thermal wig-wag.

Because the specified voltage drop is only 500 mV, they are apparently running the low-current electronics (including a relay) through the unlit lamp. Because the lamps in question are so small, they try to light anyway.

Ah ha.  Good detective work! 

Of course now it's even more curious why there's a "minimum" load of 2 Amps or whatever per your previous observation...and why there's such a massive frequency and duty-cycle range if the "timing" is performed electronically rather than thermally?!

You guys are above my pay grade electronically but those are interesting observations. Bottom line is the relay is cheap (<10.00 before shipping) and very simple to hook up and you could hook up several flashers to one relay if desired. The negative of course is the 60 hertz buzz but under a layout I don’t think it would be noticeable especially with a 494 beacon and 192 control tower going. They buzz like the dickens. Toy Lionel type layouts(like I have) are very loud anyway. A bridge rectifier may quiet the buzz? I may have one somewhere to try but then the cost goes up.

dogdoc

The buzz is at 120 Hz...not 60 Hz.  That suggests to me (my opinion) that there might be a bridge-rectifier already inside the electronic flasher module.

Aside from the buzzing noise, whether maskable or not, I see the bigger issue as wear-and-tear on the relay contact.  A "typical" electro-mechanical relay might be rated for, say, 1 million cycles.  Seems like a lot right?  Well, if the relay is banging away 120 times per second 1 million cycles is "only" a little more than 1 hour of operation.  I realize a typical crossing flasher might be on only a few seconds here, a few seconds there.  But in the big scheme of things, 1 hour (or so) of operation does not seem like a lot especially since it's relatively inexpensive to supply constant DC.  My 2 cents...

This is made for emergency vehicle flashers so it should be designed to run a long time as they may be on for hours at some accidents or incidents. Of course you are not out much if it fails. The 120 hz buzz would be gone on dc I assume?

I can't imagine how it could buzz at 120 Hz if you apply smooth DC.  But simply using a bridge-rectifier after the train transformer does not provide smooth DC.  That is, with just a bridge-rectifier you get pulsed DC which I believe will still buzz.  But give it a try and let us know.

Then, if it indeed requires smooth DC we can discuss your options - such as an AC-to-DC voltage converter module, bringing in external DC from a wall-wart, etc.

Ok, a little more playing around. I connected a 400 volt 8 amp bridge rectifier to an ac transformer with the following result:

asyou can see that it worked great with no buz. I had that rectifier laying around but I found some online for 3 to 5 bucks. This relay and rectifier are a simple and inexpensive way to set these old flashers up and you could run multiple off the same rectifier and relay. My layout is small (5x9)so I usually set the flashers to run continuously for the effect. My trains are running by them all the time anyway. I like a lot of toy like Lionel action not so concerned with realism.

Pleasantly surprised it does not buzz with just a bridge rectifier...and works over a wide AC voltage range too!

Thanks for the update!

Addendum: not on data sheet but on Amazon site selling EF-30WW for $13.56, it says rated life is 400 Hours.  So at 60 flashes per minute, that's about 1.5 million relay cycles.  Plenty of play time for a train layout!

OK, now just to be silly, I want to go rectifier - relay- flasher and add a bell to mechanically ding whenever either light flashes, similar to the prototype.   Any suggestions that don't involve transistors, integrated circuits and so forth?  Emphasis on ruggedness and simplicity, as this will be going on the museum layout and will be subjected to little Gomez Addamses leaning on the button...

So far I've tried wiring a Marx bell in series with the 154 ground from a bridge rectifier and alternating  sending power to the lights to simulate the flasher.  The lights work fine, but there's not enough juice to ring the bell.   The answer which comes to mind is feeding the bell from the DC flasher terminals and adding diodes so the lights don't blink together, but if someone can come up with a low voltage mini-gong... 

Thanks!

Mitch