IMR26500x5/Osram64633 and regulated Dvr

Looking for some help, please. :candle:

I am trying to run the following setup; OS64633 at 16.5 volts with Alan's Regulated incan driver and five of AW's IMR26500's. I had two bulbs start up and then flash after two or three seconds. I am not savvy with hotwire setups and am looking for a little advice. I measure the voltage at the socket without a bulb and it reads 13.31 volts not RMS (fluke 77).
Thanks!!

Hi, got your pm asking for advice on this. I also forwarded it to AlanB since he knows what settings he used.

First I had to find your order in AlanB's sales thread to see what you specified.

Next I refreshed my memory on the Osram 64633 bulb from my tests.

You told AlanB that you were using 5 x IMR26500 (AW's Lithium Manganese cells) which fully charged, should provide 21.5+V.

You specified a regulation maximum setting of 16.5V, which should be well within the safe range, and you specified a nice, safe soft-startup time delay of 500ms (half a sec).

Based on just those two parameters, this bulb should not be blowing, and I suspect it was not setup/calibrated correctly.

The good news is it appears that you ordered a programmer with your order. The bad news is you may need to now learn how to use it...or send it back to AlanB. I believe the latest version of Alan's config files are in the phdConfig-R0020.zip file at this download link.

I have looked at Alan's graphical Python setup program, and there is an entry for the 64633 bulb, but let's let him see your post and my initial comments before we go any further. Even the graphical python program he wrote may be too detailed for someone just starting out.

I know you can start by taking the various voltage readings with the batteries outside the light (just touch them end to end, and measure that battery voltage), then like you did before, put the probes in the empty bulb socket, and see what your readings are with the various click combinations you specified in your order...but a double click going to maximum voltage should be the most important reading with regards to the bulb flashing issue.

Hi, folks. Sorry to hear you are having problems, Frank.

I would expect we are talking about your regulator SN127. I have the config file here in case we need to look at that.

Lets see if I can do anything with the data you provided already. 13.31 volts average DC. Battery voltage is 18 nominal and 21 fully charged.

If we assume for a moment that the battery was 21 volts at the time you measured 13.31, and we multiply those values together and take the square root we get 16.7Vrms.

If we assume the battery had dropped a bit and was at 18V we get 15.5Vrms.

With fresh batteries they are likely to be closer to the higher value.

Let's calculate battery voltage assuming the output was 16.5Vrms and the measurement was 13.31Vavg. I get 20.45 volts at the battery.

So if the battery was at about 20.5V when you measured 13.31 V average then the RMS value would be about 16.5. Plausible.

It is possible the regulator is working ok. One question I would have is does the output stay constant from turn-on through the several second window where the bulbs failed. Unfortunately the load is lighter so it is not quite the same test with the meter as with the bulb, but easy to do an no risk to bulbs.

If I was doing this I would probably put another bulb in and ramp it part way up and see if things hold stable. If the regulator is varying it should do so at a voltage below maximum voltage.

Another thing I would do is to make a file for say 15.0 Volts and try that. Just to see if the regulator is acting stable. I could make the file and send it to you. You would have to load the file into the regulator.

Of course if you want to send it back I can run some tests, but I can't load it exactly as the bulb does.

I can say that I don't ship any regulators that are high in voltage by any significant amount. Generally they are just a tiny bit low. But my tests are lightly loaded and they don't last very long. There could be a problem that I don't see.

I also wonder if filaments need a bit of "conditioning" cycles and/or time before running at higher voltage. We definitely do that for high power tubes, but that is a bit different...

If it helps I just measured my AlanB reg./1 stage/16.5v./5xIMR/4D Mag/'633 at 13.32v with fresh cells and a really crappy multimeter.I'm still on my first bulb with about 20-30 minutes of runtime.

If we are both getting the the same reading at the socket and he's blowing bulbs and I'm not then it must be........

ANDREAS FERRARI said:

If it helps I just measured my AlanB reg./1 stage/16.5v./5xIMR/4D Mag/'633 at 13.32v with fresh cells and a really crappy multimeter.I'm still on my first bulb with about 20-30 minutes of runtime.

If we are both getting the the same reading at the socket and he's blowing bulbs and I'm not then it must be........

Click to expand...

Interesting measurement. Thanks for the info.

Lets see if I can post Frank's configuration file. If you put this file in your regulator software directory as file SN127.cfg then you'll be able to see the config, and anyone can make a new config with lower voltages for Frank. It won't need to be recalibrated if the only thing changing is the output voltage and programmer type.

Hi Alan,
I could send you the whole setup if you would like to test it. This includes the tube/head a couple of bulbs and the batteries/triton charger. Just send me the address via PM. I am willing to do anything you suggest on my own if you think that would be the way to go. I think if I could simulate the bulb it might be a little cheaper for me. Would a large resistor be the right way to emulate the bulb? Here is the bulb label: "OSRAM HLX XENOPHOT 64633 15V 150W GERMANY 1368"
Thanks,
Frank

Alan B said:

Lets see if I can post Frank's configuration file. If you put this file in your regulator software directory as file SN127.cfg then you'll be able to see the config, and anyone can make a new config with lower voltages for Frank. It won't need to be recalibrated if the only thing changing is the output voltage and programmer type.

Click to expand...

Wow, that's a really nice setup.

I would hate to ship that around and risk damage/loss etc.

That would be quite a resistor to simulate that bulb. Using a resistor load requires a real AC+DC RMS meter as well as the right value of high power resistor, so it is not easy to do or very cheap.

What I would do is lower the voltage output of the regulator and test with the batteries and bulb and make sure the regulator output is constant. It doesn't matter what the meter reads, but that it be fairly steady. Then I would work up in small voltage increments until either the bulb blows or the voltage and output are adequate for the purpose. If the bulb blows then back off a half volt or so and call it good.

You can download the config file I sent in the earlier posting (or I can send it separately) and work with that. Don't change the calibration coefficient or voltage range, just the max output voltage. Those bulbs may be happier on a slightly lower voltage.

Or you can send the regulator to me and I'll set it down a half volt or so and it will probably be ok.

A 1.5 ohm, 150 watt resistor would be HUGE, and way more expensive than a bulb. Best way of making one would be to rewind the resistance wire of an old bar radiator in several parallel strands - oh just buy another bulb.

lctorana said:

A 1.5 ohm, 150 watt resistor would be HUGE, and way more expensive than a bulb. Best way of making one would be to rewind the resistance wire of an old bar radiator in several parallel strands - oh just buy another bulb.

Click to expand...

OK, now what do I wrap around this?

LuxLuthor said:

OK, now what do I wrap around this?

Click to expand...

Oh, keep up. I said "bar radiator" not "oil-filled column heater".

Measure it and see where you can tap it for 1.5 ohms. Might work.

At 150 watts? Think man!

The ONLY feasible load of these parameters is a light bulb.

lctorana said:

Oh, keep up. I said "bar radiator" not "oil-filled column heater".

Click to expand...

That's all bar metal and hot water circulating steam, baby!

LuxLuthor said:

That's all bar metal and hot water circulating steam, baby!

Click to expand...

Oh, I see. Heating from the Dark Ages. I once read about those in books.

lctorana said:

Oh, I see. Heating from the Dark Ages. I once read about those in books.........

Click to expand...

My friends use to call me Red.

I was the only person they knew who ever red a book!

And I thought that Lux heated his house with maglites?

Cheers,
Nova

One other setting I would change is to slow down the ramp further. The 500 millisecond value is approximate, I think there is also a setting for "slow" which may help a bit as it may be closer to 3/4 of a second. Depending on the roundoff this may be twice as long as the 500 value gives.

ANDREAS FERRARI said:

If it helps I just measured my AlanB reg./1 stage/16.5v./5xIMR/4D Mag/'633 at 13.32v with fresh cells and a really crappy multimeter.I'm still on my first bulb with about 20-30 minutes of runtime.

If we are both getting the the same reading at the socket and he's blowing bulbs and I'm not then it must be........

Click to expand...

Drywolf-have you figured out the problem?

I charged up my batteries tonight and did some testing.

Battery stack=21.1v
x
Socket voltage=12.79v

=16.42v

Does that sound right????

We seem to be getting the same voltage at the socket.Perhaps it's a problem with your bulbs????

No, I had an issue with my computer, I won't get back to my light until Friday. I'll keep everyone informed of my progress. I am wondering if you might be right and my bulbs are not the same? I'm planning on going down to 10volts and then work my way back to 16.5, a volt at a time. I bought my bulbs from replacement bulb dot comOsram BRJ 64633 15V 150W
Item# Osram BRJ 64633 15V 150W$3.99 USD

Thanks,
Frank