Current limiting resistor figures- Mtbr.com

# Thread: Current limiting resistor figures

1. ## Current limiting resistor figures

I was experimenting on the work bench today with a SSC P7, trying to dial in the resistor values needed to make a simple driver board. The dual 1400ma stack I make is too difficult for the average user and I need something easier to assemble.

I started with a freshly charged 4.2 volt - 4 x 18650 protected cell battery pack.
I set up two multimeters for measurements.
One is in line measuring the current flow on the way to the P7.
The other one is measuring the voltage drop across the test resistor.

I had a bag of .68 ohm Vishay resistors and so I started soldering them together in parallel to reduce the resistance and then take a new measurement. I also used a couple 1 ohm resistors to get 1 and .5 ohms.

The spreadsheet below shows the resistor value in ohms
The voltage drop across the resistor in volts
The current flowing in the circuit in amps
The wattage column was just a calculation of Voltage x current

Note that when I did a quick direct drive/no resistor check, I read 3.4 amps.

<TABLE dir=ltr border=1 cellSpacing=0 borderColor=#000000 cellPadding=2 width=281><TBODY><TR><TD height=16 colSpan=4>
Battery pack = 4 x 18650 = 4.2 volts DC
</TD></TR><TR><TD height=16 width="26%">
Resistor
</TD><TD height=16 width="24%">
Voltage
</TD><TD height=16 width="26%">
Current
</TD><TD height=16 width="25%">
Wattage
</TD></TR><TR><TD height=16 width="26%">
1.0000
</TD><TD height=16 width="24%">
0.890
</TD><TD height=16 width="26%">
0.88
</TD><TD height=16 width="25%">
0.7796
</TD></TR><TR><TD height=16 width="26%">
0.7000
</TD><TD height=16 width="24%">
0.820
</TD><TD height=16 width="26%">
1.00
</TD><TD height=16 width="25%">
0.8200
</TD></TR><TR><TD height=16 width="26%">
0.6800
</TD><TD height=16 width="24%">
0.775
</TD><TD height=16 width="26%">
1.14
</TD><TD height=16 width="25%">
0.8835
</TD></TR><TR><TD height=16 width="26%">
0.5000
</TD><TD height=16 width="24%">
0.685
</TD><TD height=16 width="26%">
1.30
</TD><TD height=16 width="25%">
0.8905
</TD></TR><TR><TD height=16 width="26%">
0.3400
</TD><TD height=16 width="24%">
0.575
</TD><TD height=16 width="26%">
1.71
</TD><TD height=16 width="25%">
0.9833
</TD></TR><TR><TD height=16 width="26%">
0.2267
</TD><TD height=16 width="24%">
0.466
</TD><TD height=16 width="26%">
2.06
</TD><TD height=16 width="25%">
0.9600
</TD></TR><TR><TD height=16 width="26%">
0.1700
</TD><TD height=16 width="24%">
0.391
</TD><TD height=16 width="26%">
2.31
</TD><TD height=16 width="25%">
0.9032
</TD></TR><TR><TD height=16 width="26%">
0.1360
</TD><TD height=16 width="24%">
0.337
</TD><TD height=16 width="26%">
2.48
</TD><TD height=16 width="25%">
0.8358
</TD></TR><TR><TD height=16 width="26%">
0.1145
</TD><TD height=16 width="24%">
0.293
</TD><TD height=16 width="26%">
2.62
</TD><TD height=16 width="25%">
0.7677
</TD></TR><TR><TD height=16 width="26%">
0.0971
</TD><TD height=16 width="24%">
0.265
</TD><TD height=16 width="26%">
2.73
</TD><TD height=16 width="25%">
0.7235
</TD></TR><TR><TD height=16 width="26%">
</TD><TD height=16 width="24%">
</TD><TD height=16 width="26%">
</TD><TD height=16 width="25%">
</TD></TR><TR><TD height=16 width="26%">
0.0000
</TD><TD height=16 width="24%">
0.000
</TD><TD height=16 width="26%">
3.40
</TD><TD height=16 width="25%">
0.0000
</TD></TR></TBODY></TABLE>

Here's the diagram of how I set up the cicuit.

2. Your measurements are very accurate I checked some of them: I=V/R. Look:
0.775/0.68 = 1.13970588235 which is 1.14 rounded. 0391/0.17 = 2.3. And so on. You really don't need to use two meters if you know your resistors values quite precise.

3. One meter was always measuring current and one meter was always measuring voltage as I added resistors.

The current measuring meter has a special tool plugged into it

It was way faster to use two meters at the same time.

You never know the exact value of resistors because of the tolerance variation.
I like to measure things to get a real world figure.

4. In DD mode, current would drop fairly quickly (a few minutes) to 3.0A. Based on visual observations of my DX P7 lights, I estimate they stay at max brightness for only a couple minutes before you begin to observe the I/V drop - then another ~88 minutes till they're so dim you know it's time to shutdown to protect the 18650s.

So, I wonder if a 0.1 ohm resistor is necessary. All the wiring involved in a bike light will drop the voltage a little bit anyway.

I know 2.8A is recommended max, but really, who cares if you get only 500 hrs from a P7 emitter, rather than the XX,xxx hours they claim?

500 hrs is a lot of riding - about 5,000 miles.

The emitter will be obsolete long before then, probably.

5. Good points, Yeah, I agree.

In my test rig, I had very short wires, no switches, minimal solder joints, etc.

Adding longer battery wires, connectors, more solder joints and a switch can take the place of any current limiting resistor for hi mode.

I thought that about .07 ohms might be ok, but I was more interested in a low power mode of about 1 to 1.4 amps for the uphill climbs here that can last an hour

I also am not too worried about the lifespan of my P7's.
I buy them in bulk and they are cheap.
I wouldn't care if I got one or two seasons out of one.

The replacement bulb on my HID is \$150, which would really suck to have to replace that.

P7's are so cheap, I don't care.

6. I had time today to build a simple driver board consisting of two 1.3ohm flame proof Visay/Dale resistors, in parallel.

This gave me .65 ohms and worked out really nice as a low setting on a SSC P7.

I then cut some small round circuit boards on my CNC that make assembling these driver boards very easy with a clicky switch.

I tested a fully assembled Marwi SSC P7 conversion with the new driver board. The driver board has a DX clicky switch with 3 positions
Off - High - Low

High is direct drive, but there is just enough resitance in the whole circuit to keep the current from running away.

The current reading on Low came out to 1.13 amps through the resistors.
I hooked up a fresh 4x18650 4.2volt batterty pack and my current reading tool and let the Marwi light run for 3 hours on the work bench.

After a couple hours, the current dropped to about 1 amp on low.
After 3 hours, the current was just under an amp.

Then switched the light to High and I still had a couple amps of current running.

This is going to be a whole lot easier to build than my Hi-Lo dual 1400ma board driver. It was so easy, that I assembled two of the Marwi SSc P7 conversions in just over and hour, not including battery packs, just the complete lights with cord and Anderson Powerpole connectors.

I'll be documenting the build info on my web site.
I'll also have some of the Marwi SSC P7 conversions for sale in the next couple days for anyone that wants a completed light.

7. Can you run the same test with your more expensive conversion.
Just wondering if the 2800 driver can still produce 2800 mahs after a similar time?

8. It does about the same thing. You only gets the full 2.8 amps current from the Hi-Lo driver boards until the battery voltage drops below a certain point.

Not a problem though, it's hard to see the difference in light levels until the current drops way down. It's so gradual, your eyes don't see it.

9. EL...I couldn't help but wonder: What do you think the Lumen output is with your new set up on low? Also, while I'm at it I thought I'd throw a couple facts in that might effect a setup using direct drive. SSC list the M-bin emitter as having a voltage drop between 4.25volts and 4.50 volts but the D-bin is nearly half the drop. Also, different 18650's also have different internal voltage drops as well. Especially if you were using the new red 3000mah Trustfires. Anyway, after all is said and done, which drive type do you think is better? Is one more reliable than the other or more efficient?

10. Hmm, I don't know, but it is way more than enough for the long uphill climbs we do here.
Should be bright enough for many folks to use as the main light for in town riding without cars flashing you with their brights.

It is actually still a nice amount of light on low, plus it extends the battery life.

If someone wanted to, they could go down to a super small and light weight 2x18650 pack and easily get a couple hours of useable light.

If I had to do a wild arse guess on the low light output, based on the exagerated 900 lumen, full output SSC figures, I would say it's about 60% of full output at 1/2 the current draw?

Just a WAG.

Off to the shop to cut some boards on the CNC, take some pics, write up a tutorial, slap together a Marwi or two.
Need another pot of coffee!

see ya

11. Isn't that why other lights use higher voltage packs and a more suitable current limiting driver, therefore running at full current for a longer duration without dipping off until near the end of the battery life?

12. It is a trade off, both systems have their merits

High voltage packs reaquire some sort of electronics that burn off the excess voltage in order to keep the current steady. Just like a solid state voltage regulator, the excess voltage is burned off as heat. You can't get something for nothing, so you end up having to put batteries in series to get higher voltages but less current capacity for the same number of batteries that a parallel pack would give.

Coupling a 3.6 volt bulb to a 3.6 volt battery pack is the most efficient/least wasteful method. However, as the voltage drops on the battery pack, so does the light output. It has never been a problem for me, you still get tons of useable lightas the battery voltage drops. It is not a bad way to go.

It takes more batteries and weight to run higher voltage packs, you need driver boards for the specified voltage and you get less current storage for the same number of batteries.

Some folks like the high voltage packs with driver boards made for the task.
Personally, I like the simpler approach that a 3.6 volt LED and a 3.6 volt battery pack offers.
Both system work, whatever works for you is the way to go.

13. It would be interesting to see how your light design compares to a DX bike light after an hour or so on full.
Once I have have built my marwi conversion maybe I will run a few tests myself.

If your simply driver is giving comparable run times in high mode then is seems like a great option to go for.

BTW the parts turned up today, do I need an o ring between the glass and the bezel as never ordered one.
Also should the machined part that holds the led be a tight fit to help heat disipation, it seems to rattle slightly.

14. My design is brighter than the DX on high.
The DX light runs lower current than the max.

I base my designs on my real world riding needs, which may be different for other riders, like road riders or commuters.

This is the way most of my Mountain bike riding with lights goes

We start a ride and climb for up to an hour using a low mode. You don't need very much light on a slow climb.
We get to the top of the mountain and then turn on every light we can get our hands on.

After the descent, we start another climb. Rinse and repeat, rinse and repeat.....

I only need the high beams for the descents, which are over way sooner than we all like them to be.

It's not hard to beat the high output mode on the DX light.

I get 3+ hours of useable light from a 4x18650 pack, which is good enough for my personal riding situations.
It's easy to add batteries to make your battery packs any size that works for your particular situation.

15. A ride a lot of XC twisty singletrack and not that many mountains due to where I like.
I want max light at all times so don't mind carrying an extra battery pack in the camel back so maybe a different battery/driver design would suit my riding.

Can you answer my other questions?

16. Originally Posted by El34
High voltage packs reaquire some sort of electronics that burn off the excess voltage in order to keep the current steady. Just like a solid state voltage regulator, the excess voltage is burned off as heat.
This is totally incorrect. A buck converter is far more efficient than a current limiting resistor, it does not just burn off the exess as heat. It steps down the voltage, and reduces the current drawn from the battery. They typically have less than 20% power loss. Conversely, the current limiting resistor loses 100% of the power to heat.

I won't argue that your way isn't cheaper, but a converter will make a lot more light from a given amount of energy in your batteries.

17. Ooops, sorry, forgot.

The Marwi light should already have had an o-ring inside the bezel.
The glass goes up against the o-ring on one side and up against the reflector on the other.

The machined heat sink slides into the Marwi body as a firm fit, it should not rattle around.

Is that what you were asking?

I just finsihed setting up a new assembly page, have a look here.
http://www.el34world.com/Misc/bike/BikesLights40.htm

18. This is totally incorrect. A buck converter is far more efficient than a current limiting resistor.
Sorry, wasn't talking about the resistor mode, I was talking about direct battery to LED drive compared to other systems that drop voltages, like voltage regulators. Never used a buck, haven't had the need for one in my particular situation.

Just so you understand what I am trying to achive here. My whole approach is to make light building for the average person easy using the cheap Marwi carcasses.
Any parts used have to fit inside the body of the Marwi and it has to be easy to assemble.
The simpler the setup the better for this situation.

Hence the new simple resistor board driver and not the dual 1400ma 7135 based board mod that I have on my web site, which is a huge amount of labor to build.

My whole situation involves making it easy for my customers to build things by providing the parts and information whether it is guitar tube amp (my main business) or simple P7 bike lights.
Since about 1993 I have been known on the web as providing tons of useful free info on my web sites for my customers.

I deal with avearge, not much electronics experience type folks all day long and so I have to keep it simple for them.
The diy'ers here are much more knowledgable than the general public and so I test my ideas here first.

I rather enjoy doing tutorials and spreading free info. It's fun for me.

19. Yes thanks for the answer

I am missing an o ring I purchased the silver marwi.

There is a very small amount of movement between the marwi housing and the machined component when slid in, I guess this is due to wider tolerances is the marwi housing which are not critical to a halogen light.

I am concerned you will not get a great thermal transfer between the machined led housing and the marwi housing. I was hoping to introduce some thermal paste between both but I think the layer may be too big.

20. Sorry about that. Email me with your name I have on file here and I will throw an o-ring in an envelope for you. I guess I am going to have to look inside every marwi body to make sure there is an o-ring.

Must be that some of the Marwi bodies are looser than others.
The black ones I have used are pretty snug.

The silver ones were some sort of prototype that never saw production.
Also, the black ones are anodized and probably have a thicker inner diameter.
I would say to use a bit of thermal paste if you have to.

If it is too loose, you may be able to shim it with some very thing tin?
Just throwing out ideas.

I just checked a silver one here and it fits pretty good, not sure what the deal is?

21. I will try some tin foil with paste and hope for the best.

22. Originally Posted by Mr AB
I will try some tin foil with paste and hope for the best.
I would want a new light body that the heatsink fits correctly... using tinfoil and paste to make it work seems kind of hokey...

23. Originally Posted by El34
Hmm, I don't know, but it is way more than enough for the long uphill climbs we do here.
Should be bright enough for many folks to use as the main light for in town riding without cars flashing you with their brights.

It is actually still a nice amount of light on low, plus it extends the battery life.

If someone wanted to, they could go down to a super small and light weight 2x18650 pack and easily get a couple hours of useable light.

If I had to do a wild arse guess on the low light output, based on the exagerated 900 lumen, full output SSC figures, I would say it's about 60% of full output at 1/2 the current draw?....
Okay, I just did some more looking at the P-7 charts. Based on those and the fact that you mentioned the low output current was near 1000ma, I'm going to guess the output is somewhere around 300-250 realized lumens ( if your using the M-bin ). I see that as making a real nice all-around usable light setup. Heck, right now I've been using my MTE P-7 5-mode on medium which gives out about 175lm and I've been thrilled. If your conversion kit is price friendly enough I might just have to give it a try. When I get a chance I'll be checking out your web site again.

24. I would want a new light body that the heatsink fits correctly... using tinfoil and paste to make it work seems kind of hokey
Every Marwi light body I have here that I checked fits real nice.
He has one of the silver prototype Marwi's.
Who knows why that one light body is different.
The black ones are anodized and they all have been spot on.

I have another 100 black bodies on the way to me.
I'll make sure that all of those are good to go when I receive them.

I added a handy parts list page here for the Marwi conversions
http://www.el34world.com/Misc/bike/PartsList1.htm

25. Hello should there also be a seal between the plastic housing and the metal housing?
It appears water can in at the top between the metal and the plastic.Is the silver prototype missing some pieces. Is there any way you can get me a black housing sent out instead?

26. Alan,
If you have the need to waterproof a light, use clear RTV silicon on any areas that you think needs it.
Clear silicon is very flexible and can stand the heat.

Email me if you need instructions on how to do this.

27. Can you replace the silver housing for a black?

28. Allan, please email me with questions.

29. Why not make a new driver board using direct drive for full mode and 1 driver board for low mode. At least you then get a constant current low mode that is more efficiency than using resistors.

30. two reasons

This is supposed to be a simple driver so people can actually build it.
Where do you get a 2.8amp simple driver board?

If want complicated, build the dual 1.4amp board on my web site.

31. What I mean is,

I can't see the point of the twin driver as you can run direct drive without any driver efficiency loss hence making the more expensive option overkill.

I can see how the single driver is still excellent for the low mode.

32. At the time, I thought using both boards in parallel to get 2.8amps on high was a good idea and maybe it is for some folks.

Just use one board if you don't want the 2.8amps on high.

You can use any driver you want, after all, this is a DIY forum.
Someone else here will probably come up with other drivers to use in the Marwi's.

You can use anything that will fit inside, you don't have to use my driver boards.

33. It was just another idea!

34. You can work out your own driver that fits your needs and then post the info here.
Maybe someone else would like your driver idea.

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•