DIY low value resistors for current regulation

Hi all,
The deeper I get into bench testing the more parts I find I need. When working with the common DDL LM317 driver I was trying to hit a target ma value. Naturally I didn't have the exact resistor and don't need enough parts to justify and order for 1 special resistor. A solution is at hand! It does take a little practice but is very inexpensive. Nichrome wire! If this has been posted I apologize. I searched , didn't find it and wanted to share.
I needed a resistor less than 3ohms but greater than 2.5ohms to achieve an output current of 450ma. Here is a nice calculator to give you the resistance needed just by inputting the ma desired when using the LM317.
LM317 Current Calculator - Electric Circuit
Nichrome wire can be purchased in 10' lengths for $2-$3 free shipping and 30' for less than $5-$6 from ebay.

I needed a 2.8ohm resistor. Like I have that laying around..........:undecided:
I took an old high value resistor (470K) from some ancient parts I had. It looks like the phenolic type with a carbon core. I used an emery board to remove all markings. Then I soldered the nichrome wire to one side. Next slide your DMM along the length till you have more ohms than you need. I stopped at 3.5ohms and cut the length of 36ga nichrome wire(27ohms per foot) or 2.25ohms per inch. Next was to wrap it around the old resistor and measure the value. I did clean the resistor leads right against the body for a good reading. When you have it close, but higher, 3.1ohms, apply just a tiny bit of flux with a toothpick and solder it to the resistor lead. Flux is needed to solder to nichrome wire. Rosin core just does not work well. The solder will creep up the wire some and hopefully you should be near 2.9ohms. To fine tune it all you need to do is reheat the solder and drag it up the wire just a very little distance and remeasure. If its to low after the the 1st soldering, it can be undone, or marked as its value with a folded piece of tape on the lead for later use.
Next I checked the amount of heat produced. I ran a LPC-815 for 3-5 min constantly and touched the bare wire. It was warm, almost hot holding my finger on it for 15 seconds while running. Many will state that nichrome wire increases resistance with heat. This is true, but if the right gauge is used this presents no problem. My 2.8's current changed 3ma from a cold state to full running with an actual current of 453ma. Works for me, is cheap and no waiting to finish a build! I can and have put heat shrink on them or Liquid Tape if fear of a possible short exists. 34ga is .005(0.13mm) dia. A better choice for less heat would be 34ga at 450ma as it is producing just over 1/2watt of heat.

In the picture the wire is new and is away from the form just a bit as it is slightly springy. This is easily solved. After the wire is soldered to the 1st side, it can be connected to a battery very quickly for a second till red hot.
This removes the springiness, forms an oxide coating on the wire that does not conduct well at all so the wire can be wrapped with the coils touching if needed! A little gap between winds is preferred. All that is needed is to clean the end to be soldered.

Hope you enjoyed the post.

Interesting. I wish I knew more about electronics to appreciate this.

Tech_Junkie said:

Interesting. I wish I knew more about electronics to appreciate this.

Click to expand...

Hey T_J...

Basically.. a resistor resists the Flow of electrons...
The amount it resists is noted as Ohms...

A 1 Ohm resistor.. resists less electrons than a 1000 Ohm
resistor...

A standard small wire may have a resistance of 1 Ohm per 10-20 feet
where a Nichrome wire as used above has a resistance of 27 Ohms
per foot... (different size nichrome wires have other values per foot)

So if you take a DMM and set it to read resistance and put the 1st
probe on one end of that 1 foot long nichrome wire and the 2nd probe
at the other end.. you would have a reading of 27 Ohms on the meter...
Now place the 2nd probe 1/2 way along the Nichrome wire (at 6") and the
resistance of the wire is now 13.5 Ohms...

You can basically make your own resistor values if you need a non standard
one...
The large value resistor that Vaporizor wrapped the Nichrome wire around
is only a form... Its High resistance will make almost no difference to the
nichrome wire's resistence..

Jerry

Ah...I get it. So a length of nichrome wire is used to tweak the resistance -/+ on top of the original resistor value.

Hehe.. that sure is a labour intensive approach to a problem, but its possible obviously.
But why not just take resistors in paralel and series from standard values to achieve the resistance you require? You could do 1.0 and 1.8 ohms in series to get to 2.8 ohms, using only 2 standard E12 series resistors.

One alternative i've seen people use to get exact values is a filing down resistors. You start with a resistor that is a bit lower than what you need, and use a file to shave down the material to the exact value. In this case you'd take a 2.2 or 2.7 ohms resistor and file away until it reads 2.80 ohms. Obviously this is rather crude, but people have been doing this for ages on pth resistors.

For the example indicated the 470k resistor is just a former as mentioned - its contribution to the actual resistance is neglible. The approach is more useful for making RF inductors, since such resistors provide a nice a rigid former with wires attached.

Yeah... that is the way we do it in the shop...by putting
combinations of resistors in parallel or series as required..
but you do need a variety of values on hand..

The filing of notches in resistors only works on Carbon resistors
and could possibly change their wattage... depending how deep
you need to file...


Jerry

Thats a valid point alright - filing down resistors creates a local weak spot that reduces the maximum dissipation. In most examples i've seen the goal was to create exact resistors that didnt see much thermal load at all, more attepts to get a LR circuit to exact timing and such.

It does work on metal film resistors too though, but its much more tricky to do.

Using series/paralel configurations is much easier though. If you have a complete E6 series on hand you can create any value down to about 1 percent using no more than 3 stock value resistors. Many sets contain complete E12 series, which would allow you to make 1% accurate resistors with only 2 components.

Obviously the rated tolerance on E12 series is 5%, but i've found the real components to be much more accurate than that. I know someone who tried to figure it out, using 5% tolerance 100k resistors from numerous supliers. It turned out that >>90% of all resistors sample were within 1% or 100k.

That was the main idea, to have something you need till you can have enough items to make an order. I don't have drawers of low ohm resistors laying around to combine and I certainly didn't want to place an order for a few resistors with the shipping charges I've seen.
General assrt bags rarely have less than a 10 ohm.
Now I did have various cards of $2 10-30' lengths of different gauges of nichrome wire. :thinking:
So, this is my solution to keep building and testing until I make an order. Then I can swap it out if I choose. I wouldn't suggest this as a substitute for a "real" resistor, but it will get get you buy in a pinch. This one stays cool enough to leave in at 450ma's. 34ga would be a few more wraps and a lot cooler.
If it was just for test purposes....you could just clip lead both ends of the wire, read with a meter to the value you wanted to try, and test with it that way.
Crude? Yea kinda, but power resistors were made that way for yrs and sealed in ceramic. Still are. Do we have to do it that way? Well, I'd rather have my diode/driver testing going on rather than it all just sitting there. I have a good idea of what low values & wattage's I want to order now and never waited a day to play.

I have been looking into making some coils for a boost drive. Same situation...no drawer full of them either. I do have many ferrite cores, old coils laying around.

Come on......I did call it DIY an it does give you a warm fuzzy making your own.
Maybe it should have been The DIY Precision Resistor Survivalist Guide? :crackup:

Very good way to make the resistor, many commerical resistors are made this way buy winding some nicrome on an existing resistor, but at the same time you are making an inductor and this COULD be bad news for a boost driver :undecided:
I found out a long time ago, If a job is worth doing it's worth doing right ! and if you are going to do it right you have to have the right tools, if you are going to be setting currents on a DDL you need to have many different values of resistors and when working with boost drivers it's even more important not to use inductive type resistors,.
That is why I keep on hand a complment of many different values that can be parrelled to make any resistance I need

The inductor/choke coil would be wound over ferrite like on the boost drive I saw a pic of. That would be its specific purpose. I was told it was a ferrite core. If its a plastic form..all the better to work the dia with. I have spools of enameled wire in the 40ga's up into the 20ga's.

This was mainly for just getting values I needed to see what I really wanted so I decided to post it since I searched and didn't find it. Looks like a good assortment of anything less than 5 ohms is what I'll be buying. I'm loaded on above 10ohms for lower power builds.
It works fine on DDL types. An carbon core shouldn't be much of an inductor at all being non magnetic. It may have some "moderation effect" for adsorption effecting the value as an inductor, but can't "ring" it back AFAIK. Being a metal coil it will have EMF. I can build a DDL "dead bug" style. Not a boost.

On the DDL drivers(LM1117/LM317) if it's gonna push 450ma or greater it will require a 1/2watt resistor min from my tests and calculations. I figure you and a lot of ppl on here know that. Maybe it'll be read by a noob and they can see something a little diff. I like to do the math, build, see the discrepancies that occur. :undecided:

I've read so many intricacies on here since joining abt boost drives, part placement, tracing routes and such. I'm still reading and learning. :thanks:

Appreciate the input Flamin!
Speaking of input ....it's the 15th! Donation time! :wave:

Edit:

Tech_Junkie said:

Ah...I get it. So a length of nichrome wire is used to tweak the resistance -/+ on top of the original resistor value.

Click to expand...

TJ, yes you can depending on the value of the resistor and the value of the wire. You just think of the wire as another resistor in parallel. I chose a 470K so it would basically "not be seen". Electricity follows the path of least resistance. Parallel resistors do a sharing thing.

way to go! BTW, any wire has a resistance per length ratio. What makes nichrome special is that its ratio of resistance to length basically doesn't change with temperature.

Thanks.
Actually when coiled and heated to high temps, it does increase in resistance a bit. For our power loads it's not significant to really mention. Here is a quote from the wire gauge table.
"Note: For *Coiled* elements divide *straight* wire amps by 2 for approximation purposes."
Here is a table of nichrome wire and its ohms per foot. The difference in the "A" & "C" is "A" is Nichrome 80( 80% nickle, 20% chrome). Nichrome 60 ("C") is higher because it is 80% nickle, chrome and a little iron.