Par-36 4509 Regulator settings help required.

[Raoul_Duke]

Hi, I am having a 5 speed Incan regulator set up at the moment for a 6D maglite with a Five Mega 4" head and wondered if any of you has any suggestions on the drive levels for the 4509 lamp.

Its going to be switched straight onto high with no softstart, so dont want 100% to be too high

I hear that this had been driven of 4 li-ion cells ( 16.8V peak unloaded, and I guess about ~14.4V under load~), but not sure if it was softstarted or not ..... but most of the time mine would be run at one of the 4 lower levels......the lowest being whatever temp the halogen cycle can do its thing....If anybody knows what temp that is.....Cause I wouldn't know where to start on that...but trying to avoid a realy underdriven lamp....


So....... where would you space your voltage levels ( 1 low / 5 high & also initial turn on V)???

 

[Justin Case]

I set my JM-PhD-D1 regulator to 13.0V and Medium soft start speed to run my GE 4509. I don't think that the 4509 is a halogen lamp. Do you have the 4509 or the Q4509, which is a halogen lamp?

 

[Raoul_Duke]

LOL...I have neither...and never checked to see if it was halogen or not...just assumed...:green:
I'd been reading up on both ( and a bunch of other lamps) which is where the confusion probably set in ( struggling to source the Q4509 for a decent price.) Unsure now if running it at a realy low value will be detrimental to lamp life.

I will be getting one of both of these...but unsure which is best...

Still if we are using the default specs as the high setting ( 13V ) what would be good lower levels to aim for as I have 5 lower levels to select the percentages for....Oh and yes its a Jimmy M Regulator...


Thanks for the responce.

 

[Justin Case]

Hard to say which version of the 4509 to get. The 4509 is listed by GE as an aircraft landing lamp. The 4509X is listed as a marine lamp. Both appear to have identical specs for wattage, candlepower, beam pattern, and filament life. The GE catalog info doesn't seem to explain what differentiates an aircraft light from a marine light. Maybe the marine light has some additional corrosion protection on the terminals?

For a regular incan, I would imagine that any lower level would work fine in terms of maintaining (or extending) filament life. The specific settings depend on what your needs are in terms of brightness.

For a halogen, I'd probably go no lower than about 70% of the rated voltage. This is a seat of the pants estimate, not based on any actual bench or operational test data. Basically, you need to keep a halogen bulb hot enough to run the halogen cycle (typically, that temp is 250C at the glass envelope). That means driving the bulb hard enough as well as running at a sufficient duty cycle.

If you are using a JimmyM regulator, why aren't you using the soft start?

 

[A380]

I'm using 2 packs of 11AA in paralel with no soft-start. Measured 13.2v on the lamp when on. Lamp: GE Q4509.
Hope it helps.

 

[Justin Case]

This thread has some empirical data on the 4509 and Q4509. It seems that the OP was able to drive a 4509 successfully with 14 subC cells plus an NTC thermistor. But not so with a Q4509. Even using 13 subC cells means 15.6V, which seems high.

With the 4509's already short 25 hr life at the rated 13V, even a small amount of overdrive will result in a very short calculated filament life. For 14V, the calculated filament life drops to 30%-40% of the rated value (i.e., 10hrs or less). For 15.6V, the calculated filament life is 5%-10%. Extrapolations can get dicey and the estimates unreliable when you overdrive a lamp that much. IMO, luck of the draw starts to play an unwelcome role.

If I were to try overdriving a 4509, I'd probably go up to 13.6V or so, which might retain about 50% of the rated filament life. I probably wouldn't overdrive a Q4509 at all because of my perception of heat issues blackening the lamp's reflector.

Since the 4509s are relatively inexpensive (at least for some of us), if I wanted a real barn burner out of the 4509, I might consider testing one at 14.0V just to see what happens. But I personally would not go higher than that.

 

[LuxLuthor]

RD, I almost did my destructive testing of the Martek 4509Q last night, but it was raining, so couldn't take the power supply out...maybe tonight.

 

[lctorana]

My 2c' worth:

I feel that the 12.8 and 13V lamps are made for some sort of vehicle use (tractor, boat, plane etc), and my own opinion is that the "real" design parameters will be designed around a 6-cell accumulator driven by an alternator.

Hence one would expect the key design center voltages to be 12.6 and 13.8 with the ignition off and on respectively.

So, if ever I deigned to stoop to the indignity of a multi-level light, those levels would be:
Low) 12.6V (or perhaps 12.0V)
Medium) 13V (the published operating voltage)
High) 13.8V (or whatever you think the alternator would drive the bulb at in the vehicle)

Happy to change this thinking after seeing the Great Man's test results.

 

[Raoul_Duke]

thanks for the replys...interesting stuff...though I must admit to not being up to speed on the alternator/ cell / vehicle voltages.

I am going to be using a Jimmy M PHD, and the plan is on normal use/ startup to press from off once for a med or slow soft start to medium level, or level 4, then click to high (5) with a slow ramp, then click again for low (3) then click low low (2), then click low low low (1).....but I have toyed with the idea of having the option to be switched on straight to high with a fast ramp when I want full power instantly ( I dont like a waiting for the soft ramp to start up ( but realise for general use this is the way to go) but somtimes ( most of the time, but I will restrain myself) I want to go straight to high like a DD hotwire....So when I hit the switch it starts going up on a slow ramp to med, but before thats complete click again and jump straight to high ( lvl 5) with minimal soft start)
It might be nice, but unsure on the coding) if I can press and hold at anytime for full instant to get an almost instantanious big burst of light.


:thumbsup:

 

[Justin Case]

Visually, I doubt you will see any functional difference between fast soft start vs medium soft start with the JimmyM regulator. I've set my regulators to both and I don't notice any difference. Thus, I've gone with medium as the default.

Re: the mutlimode UI. It's a sequential access UI, cycling through with button presses. You can program the order of your levels (e.g., L, M, H; H, M, L; or even something mixed like H, L, M). You'd have to add code if you want a fancier UI.

 

[JimmyM]

The slow start can take as much as a full second from off to regulated voltage. Fast happens in about 100-200 mS.
The medium start is a nice compromise. The slow start is really just for creeping up to the regulation point if you're seriously overdriving the bulb.
If you want to go M->H->L->LL->LLL, I can do a medium start when turning on initially. If you double click (which is simply doing the second click before the regulator is finished ramping) it will disengage the soft start and ramp very fast to High. I still haven't worked out the details, but it's possible.

The way Alan's, Will's, and my regulation is coded right now is to calculate 2 integer values and compare them. If one is too high, it reduces the PWM duty by one step. If it's too low, it increases duty by one step. The step size is varied to effect a ramp in voltage that varies depending on several things. It does this calculation and adjustment ~250 times per second. So it's a matter of determining when and how much to change this step size such that it ramps fast enough without overshooting the regulation point.

 

[JimmyM]

I see that Lux added the 4509Q to his destructive bulb testing. 18.1V Impressive.

EDIT: 18.8 volts. Even MORE impressed.

 

[LuxLuthor]

I see that Lux added the 4509Q to his destructive bulb testing. 18.1V Impressive.

18.8 V