How does this 1971 directional flasher switch work?

[AZFIREBALL]

Main Question or Discussion Point

This is a diagram of an old 1970's directional flasher switch. Can anyone walk me thru the details of its operation?
https://drive.google.com/open?id=1OI6bL3PXfDx8lmg2eiWDCVHg9rgm0gYl

Thanks.

 

[jim hardy]

Image doesn't show. Try making it into a .jpg with Paint, save on your computer, and use "upload" button in lower right.

 

[Nik_2213]

I don't recognise that circuit.
IIRC, some used a bi-metal strip, heated to make it close to 'light', which opened the heating circuit, which cut the lights.
Clunky, very. Subject to slow start and erratic timing, very.
Still, better than lit arms which popped out of the door-pillars...
{ Never drove such, but helped push-start family jalopy downhill when the crank failed... }

 

[AZFIREBALL]

Yes Nik_2213....before the current electronic types there was the Bi-metal type.
The one shown in the diagram is used on a 1970 vintage Yamaha motorcycle I am restoring. Therefore, I would like to understand how it makes the directional lights flash at about 1 Hz. The filament bulbs are 17 watts each at 6 volts, for a total load of 34 watts. The coil feeding the cap has about 3 to 4 times the number of turns as the coil on the right, feeding the lights.

 

[Asymptotic]

I may not be thinking about this correctly, but the circuit as drawn doesn't add up.

It is based on an RC circuit in series with a relay coil. When the switch is closed the RC network will appear at first as a low resistance, the relay coil will be energized, contact set will be open, and lamps will be off. As the capacitor charges, voltage drop across it increases, current through the relay coil drops, and eventually falls enough to let the contact set return to the normally closed position turning on the lamps.

As shown here, that would be the end of it - the capacitor remains charged, relay coil de-energized, and lamps turned on through the normally closed contact set. I'm thinking the circuit isn't quite as drawn, and some way exists to discharge the capacitor to allow successive cycles.

 

[AZFIREBALL]

Asymptotic: Thank you for your consideration of my question.
You said: "... and eventually falls enough to let the contact set return to the normally closed position turning on the lamps."
Would this not then reopen the contacts due to the right side concentric coil drawing current to light the lamps?
Also I do not understand the effect the collapsing magnetic field (when the contacts open) might have on the charge in the cap as there must be an induced voltage in the RC coil when the contacts open.
BTW-The unit seems to be buzzing in between the clicks that accompany the 1 Hz off/on cycles.

<<moderator's note, bold text changed to normal text>>

 

[jim hardy]

I may not be thinking about this correctly, but the circuit as drawn doesn't add up

I agree it doesn't look right.

Here's a 'shade tree garage' explanation of the three terminal bimetal thermal type
http://www.mossmotoring.com/turn-signals/

How many terminals on this one? two or three?

 

[AZFIREBALL]

Just two. One at the switch (+ lead) and one at the lead to the lights (Green wire).
Thanks Jim!

 

[Averagesupernova]

It makes perfect sense to me. When power is first applied the fields likely cancel so the core in which the coils are wound on is not magnetized. Eventually, the capacitor charges and the fields are no longer cancelling. This attracts the armature and opens the contacts. At this point there is discharge current through the filaments which discharges the cap and continues to attract the armature holding to contacts open. Eventually the current dwindles away in which case the contacts close and it all starts over again. Very clever. The basic requirement is that at certain times there are opposing fields within the core.

 

[jim hardy]

Okay here's what i think you have, but it could be sneakier

My first guess:
an old fashioned two terminal thermal flasher but one with a second heater coil.
the R-C network applies a little extra heat to second heater coil immediately after the switch gets closed so that the first flash won't last so long. Gives a more graceful startup. Leave it to the Japanese....

I remember that on my 6 volt '53 Ford the first flash was slow, i always assumed it was because the bimetal element had to start from ambient temperature. It sped up after a flash or two as the device got warm. I've not paid attention on my newer rigs but will take note tomorrow.

If it's sneakier than that and indeed is operated by electromagnetism as you and Averagesupernova suggest, a photo would help out because we could see the magnetic path.
Can you temporarily tack in a second capacitor of roughly the same value and see if it changes flash rate ?

Interesting gizmo. Thanks, and keep us posted what you find?

old jim

 

[AZFIREBALL]

Jim:
Thanks for your input.

If it's sneakier than that and indeed is operated by electromagnetism as you and Averagesupernova suggest, a photo would help out because we could see the magnetic path.
Can you temporarily tack in a second capacitor of roughly the same value and see if it changes flash rate ?

Here are two views of a like flasher (Note different cap size). I can not add another cap as the flasher is installed (BTW-works well)

I think Averagesupernova is on the right track. There is no bi-metal in this flasher.

 

[AZFIREBALL]

It makes perfect sense to me. When power is first applied the fields likely cancel so the core in which the coils are wound on is not magnetized. Eventually, the capacitor charges and the fields are no longer cancelling. This attracts the armature and opens the contacts. At this point there is discharge current through the filaments which discharges the cap and continues to attract the armature holding to contacts open. Eventually the current dwindles away in which case the contacts close and it all starts over again. Very clever. The basic requirement is that at certain times there are opposing fields within the core.

Averagesupernova:

Thank you for your reply. I think your approach is the right one; although I do not claim to understand all the nuances involved.
The device does seem to “buzz” while the lights are on (I think).

Is it possible the in-rush current to the lights causes the contacts to open, due to high flux build-up in the core? This opening, causes the field to collapse, and the contacts close under spring pressure.
This rapid opening and closing of the contacts might be the ‘buzz’ I think I hear.

During this period, the lights stay on as the filaments do not have time to cool between the rapid off/on cycles (the on cycle may be longer than the off cycle, due to time needed for flux to build).

Also, during this same period the cap is charging and building up its flux in the core, to eventually hold the contacts open (overcoming the field made by the osculating lighting circuit) – the lights go out.

The cap is slowly discharged through the filaments allowing the contacts to close under spring pressure and the lights come back on, causing the next in-rush of current in the lighting coil; and we start over.

The one thing that still bothers me is the interaction of the two coils during rapid flux decay. How does the induced voltages in the high-turn cap coil effect the charging/discharging of the cap? I think the cap coil has about 4 times the number of turns as the lighting coil (This estimate is based on looking at the respective wire sizes.)

Is that the way you see it?

Also, please see additional images provided in my reply to ‘Old Jim’

Thank you for your consideration.

 

[AZFIREBALL]

BTW - Averagesupernova I like your avatar.
I built my first crystal radio back in 1940. Put it inside one of Dad’s old round-top lunch boxes.
Used my bed springs as the antenna. Fell asleep every night listening to that magical device!
Thanks for the memories!

 

[jim hardy]

There is no bi-metal in this flasher.

Yep. Sure looks like a magnetic structure ! You guys are on the right track - old jim

 

[jim hardy]

an observation or two..

After switch closure....
The capacitor charges while the relay contact is OPEN , through the lights
because
on one side it sees battery through 47 ohms, on other side it sees ground through the lights.

The capacitor discharges while the relay contact is closed. because It sees battery voltage on both sides.

The current required to open the contact is substantially greater than the current required to hold it open, because when it's open the air gap is smaller.

Starting at the instant switch gets closed,,,,,

Lamp current through right hand winding makes flux that opens the contact. But not immediately.
By Lenz's law, current is induced in the left hand winding to oppose increasing flux. There's @Averagesupernova 's opposing fields ..
That places some charge on the capacitor.
Flux rises because of lamp current
until the contact opens. .
Now lamp current goes away so flux tries to decrease
and current in left coil reverses to oppose decreasing flux.
Current through the capacitor and left hand coil , now reversed, holds the contact open for a few time constants. 4700 microfarads X 47 ohms is 0.22 second..
When capacitor charging current decreases to minimum relay holding current the contact closes again.

So it would be very interesting to observe voltage from the junction of left hand coil and capacitor to "gnd" . Does it ever get above 6 volts ?
If i understand this thing at all
Voltage across the 47 ohm resistor will alternate its polarity every time the relay changes state.

Check my logic please ?

......................................................................................................................................................................

btw i observed my blinkers today. On my 2002 Ford Escort the first flash is perceptibly longer. But it's driven by a relay with a half dozen wires so i dont know whether timing is set by a computer or something analog like your clever flasher. .
Changing the taillights to LED made it flash way too fast so something is measuring current.



old jim

 

[AZFIREBALL]

Thanks Old Jim for your very detailed description of the relay's operation. It would be nice to attach a multi channel scope to several points in the circuit to measure voltages, wave forms and frequencies during operation.
Would you be interested in doing some testing if I can find another flasher of this type. I do not own a scope and my grandson is riding the bike to which the one I had is now attached. So it is no longer available to me.
I have not seen any of this type flasher for sale at any of the regular places selling Yamaha parts nor at automotive distributors. It came on the new bike in 1971.

 

[jim hardy]

i'd better not take on any more projects just now. But thanks .

Is the capacitor's positive wire connected to the resistor or to the coil ?

 

[AZFIREBALL]

i'd better not take on any more projects just now. But thanks .

Is the capacitor's positive wire connected to the resistor or to the coil ?

I think it is attached to the resistor.