zener diode help

twhitehousescat

that wrote: »

if you use the zener calculation web site, with a 24v relay that approx consumes 65.6 ma (12v one comsumes about 100-160ma from ebay) for 36 volt supply which is rms. if the supply is 30-36 volts then guess it is workable.

If not enough voltage is generate the light may cut out, so freewheeling and yeild signs, plus easing off in traffic

We do not know the voltage range of the alternator from idle and max is 36v? rms
The circuit alone could consume over 30w, that resistor will get hot.


in my mind taking a slice of the voltage phase is a better way. Possibly connecting to the rev output is still better?

I did look at taking a signal from the bikes simple CDI unit , however the ign coil has a permanent live running to it (with ign on) and the coil is triggered by a negative pulse generated by the ign pickup and cdi

Heedtheadvice

OP Try the following a design based upon your original idea:
https://ibb.co/vjqk1Mh
(hope that image share works, it is a first for me!)


Description.

Assumed the quoted 12v ac output from one phase of the alternator is measured as an rms voltage with respect to the battery negative terminal.


D1 rectifies the AC to produce DC that is smoothed by the capacitor C1 to produce a minimum of approx 17v. for detection when the alternator is producing output ( e.g at tickover). R2 provides a discharge path to quickly remove the detection voltage should the engine stop. This acts in milliseconds.

R1 provides a feed to Zener diode Zd1 that conducts when it's threshold is reached at about 10V corresponding to 17V across the capacitor and 12v r.m.s at the alternator phase output. This conduction flows via the transistor base emitter junction turning on the collector emitter path and thus turning on the relay.
Edit 1: addition: The relay coil has power provided by the battery so the electrical conditions around it and the transistor collector are fairly constant and predictable. It may be best to provide this battery connection via the ignition switch and a fuse in case of any failure of the control circuit.



Should the alternator output rise to high levels (e.g. 35v r.m.s. at high revs) then Zd2 provides a path to limit the voltage at zener Zd1 and thus current through the transistor (base-emitter junction) to protect it.


If a relay without internal diode protection is used then D2 is essential to prevent high switching spikes being generated by the relay coil.


Smoothing the DC voltage prevents the relay switching off on the half cycles of rectified AC that result in zero voltage. This could happen at low speeds causing relay chatter and early failure. At high speeds that is unlikely as, although the circuit would still switch but the relay would be too slow to respond. It would not be good design practice however so is best if the smoothing is included.




Footnote.

This is a basic quick design and not all the component characteristics have been assessed not has it been built or developed in real world conditions so a bit of test and tweaking may be essential. Hopefully not too much and I have not made any significant errors or omissions. Try it at your own risk!! It may be advisable to add some very small value capacitors at point in the circuit to reduce generating interference. Construction would no doubt benefit from some potting compound around the component leads to reduce the risk of failure from excessive vibration once a design is finalised. No guarantee is given that the design will conform to any legal requirements for on road use such as the road traffic acts!!
Edit 2: It should be noted that, if the control fails in a mode that switches off the headlights whilst travelling then most probably it could lead to a safety issue.

forgotmyname

Puzzled why your messing with AC and several diodes when you already have a DC system available???

All you need is a simple circuit to enable the light when the alternator/dynamo is charging the battery.

Trying to re-invent a smart charge relay?

polymaff

re: The original circuit idea:


I wonder what min/max/typical frequency the alternator AC is at?

You don't want a relay that continuously chatters.

Unless you have shares in the relay manufacturer.

twhitehousescat

forgotmyname wrote: »

Puzzled why your messing with AC and several diodes when you already have a DC system available???

All you need is a simple circuit to enable the light when the alternator/dynamo is charging the battery.

Trying to re-invent a smart charge relay?

where have I got DC ?

All you need is a simple circuit to enable the light when the alternator/dynamo is charging the battery.

correct , can you suggest one?


I have looked at split/smart chargers , a: they are quite large , b: they are intended for much larger batteries than mine , and with different charging characteristics than mine

Heedtheadvice

Forgotmyname and polymaff are both quite right.
The easiest option is to monitor battery charge (easy option is battery voltage) or charging current into the battery. I was going to post similar comments in the darlier attempted post I lost.


The voltage method with appropriate level sensing would prevent discharging the battery too far to affect starting capability under normal circumstances.
The current method would only bring the lights on when the battery was being charged.


All the methods in this thread have drawbacks. The voltage method falls down when battery capacity does not relate proportionally to voltage at the terminals and small changes in voltage equate to larger capacity changes - hence more complex circuits to avoid, for example, temperature induced variations. It seems sensible to set the switching voltage to close to the charging voltage provided by the regulator.....but lights out if that falls a bit.


Current sensing is also a little more complex for reliable operation. The current from the regulator does not depend solely on charging the battery but also on any other load such as lights!


Given the aim seems to be to only switch on lights when charging (to prevent battery discharge a d thus no start capacity) neither the above fully produce that safely. A combination of them will do.


However the original design idea, those suggested/modified (including mine) do not meet a sensible need to prevent battery discharge, rule out lights out circumstances for some failure modes. Again a combination approach would seem to be a solution sith maybe early warning indication. Ghose are all a greater scope than the simple solution requested. Commercial solutions are probably nearest the ideal, not those historical methods employed by some manufacturers mentioned by the OP.

twhitehousescat

DC sensing would not work , applying the brake and indicators whilst at standstill could drop the voltage in the small battery quite a lot

it needs some sort of method to know when the engine is running , not battery voltage

I have one of those on another bike https://www.ebay.co.uk/itm/LITHIUM-MOTORCYCLE-BIKE-Battery-Charge-monitor-level-voltmeter-monitor-Shorai/191741139645?hash=item2ca4a99abd:g:LMIAAOxydUJTOtBO:rk:34:pf:0

the led changes colour dependent on voltage , when you put lights and brake + flashers on , it can go from green = charge to red = no charge , before battery reservoir fills up again

hence my insistence on getting a reading from the geni , which is either charging (sending power to ref/rec then battery) or not

polymaff

Is this too simple?


If the intent is to have the headlight on if and only if the alternator is running (post #6), either:


1. Connect an similar rectifier arrangement with inputs in parallel with the alternator rectifier's inputs, and with its output connected to the headlight, or,


2. Break the feed from the alternator rectifier to all of the other circuitry, insert a schottky diode in the break, cathode toward the other circuitry then connect a similar schottky diode, anode to the anode of the first schottky diode and cathode to the headlight.


NB This is assuming circuitry as in post #18

forgotmyname

twhitehousescat wrote: »

where have I got DC ?

All you need is a simple circuit to enable the light when the alternator/dynamo is charging the battery.

correct , can you suggest one?


I have looked at split/smart chargers , a: they are quite large , b: they are intended for much larger batteries than mine , and with different charging characteristics than mine

Where have you got DC? The battery circuit is DC. Only the alternator upto the diode pack is AC, is the diode pack external or internal? Everything else is DC.

Smart charge relay large? 25mm x 25mm x 10mm thick is that large? If you need it smaller remove the input connectors and solder directly to the board, only 25x18x7mm.

twhitehousescat

the reg/rect is a combined unit

please link me to one that size and to tech specs for it