Destructive Incan Bulb Tests - Updated 8/27/2010 (Newer Info Added)

[Action]

Re: Destructive Incan Bulb Tests

Also Que? Pivot table? :thinking:

An Excel function. Search for help on pivot table and play with it a bit. Its quite a powerful feature for displaying data.

 

[DM51]

Re: Destructive Incan Bulb Tests

Ohhh, Excel. Right, yeah. Excel, huh? Sure thing. Easy, lol. Yeah, good old Excel. Everyone can work Excel, right?

Uhhh… hey guys, could someone tell me where the Excel for dummies thread is? lol

 

[Action]

Re: Destructive Incan Bulb Tests

I believe that all this data is already in excel. If I could get it, I'd be happy to combine it and send it back to someone to post for FTP...

 

[LuxLuthor]

Re: Destructive Incan Bulb Tests

Action, I put each bulb's data on its own Excel page mainly because this first phase involved collecting data points for each bulb. I'm doing the run times now.

Ultimately, we want to construct a hotrater file that has realistic formulas for each tested bulb. It didn't make sense to release the premature data points yet...and last I heard, AW's regulated D driver was "months" away.

I'm checking the Lux readings a 3rd/4th time when setting bulbs at my hopeful "sweet spot" voltages, and running them until dead. The run time information is just as important as the maximum voltage flash point. I'm doing the Osram 64430 at 10V now....but what if it died after 2 hours, despite the WA bulb formula derived Hotrater predicted 9 hours?

Of note with this 3rd O-64430 it gives Lux of 198 at 10V which is a little less than my last two bulbs tested in step up manner. Each 1-2 hour run time segment for life, I'm taking fresh Lux readings to see how wide the spread is.

 

[Action]

Re: Destructive Incan Bulb Tests

I think one of the valuable things about this data right now is the ability to choose the particular voltage output inherent in the battery design and quantity and then make truly informed choices about which bulb to use. Its not so easy to quickly do this with the choices spread among multiple spreadsheets (please do not take this as critical, just a statement, ease of data collection is important, consolidation after the fact is often easy).

 

[mzzj]

Re: Destructive Incan Bulb Tests

I wish they made Emoli/A123 in smaller sizes. I just can't deal with a 5C/5D or 6D for a light. Thanks for all the nice comments!

How about 2 or 3 A123 cell lamp with boost converter to drive 35/50W IRC?

I am pretty sure that I could fit that ~67W converter on 32mm x20mm round case. (1/3 of D-size cell or so) 120W could be more tricky, but nothing unheard of. Resulting converter size would be about 16cm3 (one cubic inc) State of the art dc-dc converters operate with power densities up to 1000W/in^3 so 70W/in3 or even 120W/in3 is hardly even any hi-tech.

 

[LuxLuthor]

Re: Destructive Incan Bulb Tests

I think one of the valuable things about this data right now is the ability to choose the particular voltage output inherent in the battery design and quantity and then make truly informed choices about which bulb to use. Its not so easy to quickly do this with the choices spread among multiple spreadsheets (please do not take this as critical, just a statement, ease of data collection is important, consolidation after the fact is often easy).

I agree with you about this difficulty using all the multiple spreadsheets for each bulb. I wanted to get some of the run times done though because it seemed that just using what I have so far isn't as useful until you know how long the bulb will last. I also wanted to see how the data came out during these tests...and as you might expect it is pretty interesting....even with only 1 run time done so far.

 

[LuxLuthor]

Re: Destructive Incan Bulb Tests

How about 2 or 3 A123 cell lamp with boost converter to drive 35/50W IRC?

I didn't want to get into these kind of questions in this thread, since it is mainly about getting objective data on bulbs, and how to improve/interperet information we have never had before.

What anyone wants to do with various solutions & possibilities like you suggest is then a whole other (& very useful/interesting) series of conversations.

 

[Timaxe]

Re: Destructive Incan Bulb Tests

Lux, I'm wondering if you can do a simple Voltage/Current bit to the collected data to derive the Resistance of the bulb.

Ideally this could give us some insight into the temperature of the bulb, since under most circumstances the resistance of materials increases as its temperature goes up (this behavior explains why a 'cold' bulb may need multiple clicks on some LiIon cells to warm up and stay on). I don't know how this holds at the extremes of temperatures, but this resistance data could be a useful value to look at.

Because of how the data is sampled there will be a lot of noise in this value, but it might be useful on a per-bulb basis to see how it is performing. Thanks & good work collecting this data!

 

[LuxLuthor]

Re: Destructive Incan Bulb Tests

Lux, I'm wondering if you can do a simple Voltage/Current bit to the collected data to derive the Resistance of the bulb.

Ideally this could give us some insight into the temperature of the bulb, since under most circumstances the resistance of materials increases as its temperature goes up (this behavior explains why a 'cold' bulb may need multiple clicks on some LiIon cells to warm up and stay on). I don't know how this holds at the extremes of temperatures, but this resistance data could be a useful value to look at.

Because of how the data is sampled there will be a lot of noise in this value, but it might be useful on a per-bulb basis to see how it is performing. Thanks & good work collecting this data!

I can add another column that posts result of V/I once I get more of these run times. Thanks for suggestion!

 

[hoffner5]

Re: Destructive Incan Bulb Tests

How about doing the ROP bulbs?

 

[LuxLuthor]

Re: Destructive Incan Bulb Tests

I would have considered it...despite the expense, but my testing bulb holder is a KIU bipin as in pix. I'm not doing any potted bulbs.

 

[deeuubee]

Re: Destructive Incan Bulb Tests

I'm not familiar with incan bulbs yet. Are any of the ones you tested MR-16? I'm looking at those for the Kui sockets when they come in mainly because I have a ton of them. If not, would you consider doing some if I send them to you?

 

[LuxLuthor]

Re: Destructive Incan Bulb Tests

I'm not familiar with incan bulbs yet. Are any of the ones you tested MR-16? I'm looking at those for the Kui sockets when they come in mainly because I have a ton of them. If not, would you consider doing some if I send them to you?

I do appreciate your generous offer, but if the MR-16 look like this page image, I dont think it will work, as I am testing clear glass envelope bulbs that show up as a "point light source" from the side like this view of "dead soldiers" below. I don't think the MR-16 would show up from a side view like my setup in post #1. If I'm wrong about your bulb, by all means I would enjoy adding it to the list.

 

[deeuubee]

Re: Destructive Incan Bulb Tests

Yes, those are the bulbs. I was thinking about trying a small variable pot to adjust the voltage to the bulb in order to have adjustable output. Knowing at what voltage to stop before they go POOF! would have been great. I could have put a stop or maybe found a pot in the right range before hand. I'll give it a go until the AW regulated drivers come out.

Thanks for all the great info. I search for your posts whenever I need straight answers.

 

[LuxLuthor]

Re: Destructive Incan Bulb Tests

Yes, those are the bulbs. I was thinking about trying a small variable pot to adjust the voltage to the bulb in order to have adjustable output. Knowing at what voltage to stop before they go POOF! would have been great. I could have put a stop or maybe found a pot in the right range before hand. I'll give it a go until the AW regulated drivers come out.

Thanks for all the great info. I search for your posts whenever I need straight answers.

Oh well if you want me to just test the amps and volts in a stepwise manner (but not lumens), I would be happy to do that for you....and can give you general observation of output. Send me a PM if this would be useful info.

I should be getting those O-64430 Tungsram clone replacement bulbs, and AW's 6V 30W bulb arriving next week. Meanwhile, gathering run times.

 

[SuRgE]

Re: Destructive Incan Bulb Tests

Just what I needed since I'm a newbie to M*g mods and keep flashing bulbs!

Great job Lux:twothumbs

 

[LuxLuthor]

Re: Destructive Incan Bulb Tests

Will be adding AW's chinese bulb results, and finally got my replacement shipment of Tungsram 6v from Svetila today. Run times continue to be surprising with variable decrease of lumens with different bulbs, and not what AWR's hotrater was predicting.

 

[Northern Lights]

Re: Destructive Incan Bulb Tests

Will be adding AW's chinese bulb results, and finally got my replacement shipment of Tungsram 6v from Svetila today. Run times continue to be surprising with variable decrease of lumens with different bulbs, and not what AWR's hotrater was predicting.

Yes, Please! LL, something I have noticed in certain bench runs or mock ups on my existing hot wires when experimenting with some of these bulbs was a qualitative, very subjective and noticeable decrease in expected brightness. I never mentioned much about it; I assumed the variables in battery conditions or circuits problems were the cause and not the bulb. So it does not surprise me you are finding actual light outputs to be lower and not predictable by our current collection of mathematical models.

 

[LuxLuthor]

Re: Destructive Incan Bulb Tests

Yes, there is that....but also a surprising drop in lumens with overdriven bulb as run time progresses that was never even brought up. I suspect that running a bulb at its default specification voltage/current would not result in the lumen drop over bulb life...but when overdriving....suffice it to say there is a whole other issue to factor into arriving at a bulb's "sweet spot" voltage/current. There is a degree of OD'ing particular bulbs (using Hotrater 9-15 hour ideal bulb-life range), after which it does so much apparent damage to the filament that it becomes like a limp dishrag.