Precision resistors


Thread Starter

Mark B.

Does anybody know where I can buy 250 ohm .1% (or better) 1/2 watt resistors to drop a 4-20 mADC signal to 1-5 VDC? I really don't want to go to one of the instrument suppliers because they charge $10.00 apiece or more. I need quite a few and want someone who specializes in these types of resistors so they are more affordable.


Eng. Sergio Toledano

This is a rare solution but work's. You can buy a lot of 2% or 5% 250 ohm metal film resistors, then with a good's precision multimeter test and selects resistors inside 247.5-252.5 ohm range (1%). According to my own experience, in a 100 resistors stock's, at least 15 to 25 fall into the 1% tolerance.
best regards


Jocko Harmet

Look for Precision resistors on or or or - There are many supplier of this type of components on the internet Note: a standard resitor will be 5% tolerance, this will usually work for your application, but the precision resistor (1%) will assure success.

Jocko Harmet
[email protected]


Sergio Toledano

Sorry, you means 0.1% or 1%, in the event of first case the price is right, and forget my last reply.
best regards
Newark, Digikey, or some other electronics supply house might be reasonable sources. But you wouldn't need anything close to 1/2 watt for 20 mA, since that's nominally only 0.1 W
(calculated as P = I^2R).

Ken Irving
These are readily available in India at very attractive price. If you would like to source the same from India, please let me know Qty
required so that same can be arranged.

Kind regards


Curt Wuollet
Or any other component house. you'll be amazed at what they really cost. You might as well buy 100 so the shipping doesn't predominate.



Free Tools!
Machine Automation Tools (LinuxPLC) Free, Truly Open & Publicly Owned
Industrial Automation Software For Linux. Day Job:
Heartland Engineering, Automation & ATE for Automotive Rebuilders.
Consultancy: Wide Open Technologies: Moving Business & Automation to Linux.

Johan Bengtsson

What I can see that would not be that much above normal price for that kind of precision.

I checked at elfa and they charge 50SEK (for one, falling to 19SEK per unit for 100 or more) (the resistors they have there don't meet your requirement however, they are only 1/3W). They are made by "General Resistance" and a google search ended in this website: "
I have no idea if this might help you or not but it might be worth a try....

/Johan Bengtsson
P&L, Innovation in training
Box 252, S-281 23 H{ssleholm SWEDEN
Tel: +46 451 49 460, Fax: +46 451 89 833
E-mail: [email protected]
Look up the manufacturer's sites and find their local distributor.

People such as Vishay, Meggitt Electronic Components, Welwyn and Phycomp all make resistors of the general type you're looking for but you'd need to know a bit more information to be able to suggest a specific manufacturer's part number.

Also, what do you define as "quite a few"? 10, 100, 1000, 10000? Makes a big difference, but when you're talking to resistor manufacturers you need to be talking BIG numbers before they consider it to be "quite a few".

The high power preceision resistors tend to be quite pricey even in large volumes. Can you get away with "select on test"? 0.1% is not very high

Sorry can't give you any URLs at the moment but you'll find all those I mention above with Google quick enough.

Geoff Moore
Straight Forward Solutions Ltd
Maynooth Road, Prosperous,
Naas, Co.Kildare, Ireland
Phone : +353 (0)45 892739
Fax : +353 (0)45 893880
Mobile : +353 (0)86 8179683
email : [email protected]

Either Digi-Key ( or TTI ( will be able to get you just what you need at a good price.

I don't know how much these particular parts cost; but if .1% resistors are too expensive, you should be able to buy 1% resistors, set up a test jig, and get at least a 60-80% yield by matching them yourself. If you need (say) 10,000 of them, you should be able to make a go/no-go setup that is easy to run so someone else can help you.


Willy Smith
Numatics, Inc.
Costa Rica
Unless "quite a few" translates to thousands, one thing you might consider is looking up on one of the electronic surplus sites. This can be a hit or miss affair, but might yield what you need. This web site lists a number of potential sources.
Another thought - why not parallel a 270 ohm, 1% resistor with a 5K, 25 turn Cermet trim pot? This would give you a 250 ohm load (fixed resistor is
precisely 270 ohms) when the trim pot is adjusted to 3375 ohms, or about 68% of pot adjustment. At 270 ohms +1% (272.7 ohms) the pot would need to be adjusted to 3004 ohms to derive 250 ohms total, and at -1% tolerance (267.3 ohms) it would be set to 3862 ohms.

A 25 turn 5K pot gives about 200 ohms per turn, and the usable adjustment range is about 850 ohms (3862 ohms to 3004 ohms) to trim the +/- 1% 270 ohms resistor tolerance, so about 4 adjustment turns. If we figure the trim pot has a
resolution of about 1/20th of a turn (10 ohms), then we should be able to adjust to within 0.03% of the desired value.

This approach is probably less expensive than the cheapest 0.1% resistor (somewhere around $3 per resistor/trimmer pair). It gives the benefit of
adjustment (and the con of more parts; more potential failures).


If you need have a 250 ohm resistor quickly and low cost you can do it.

You need one resistor (carbon material) more low than 250 ohm, multimeter and a sharp knife.

Only you need eliminate material with the knife up to read in the multimeter 250 ohm. Precision will depend of multimeter accuracy and you steady hand.

Jorge Diaz
---------- Forwarded message ----------
From: "Erich Mertz" <[email protected]>
To: <[email protected]>
Subject: Re: PROC: Precision resistors

Deviating a little in one direction or another on the 250 ohm load resistor isn't too important since one can always adjust the zero and span of the transmitter to produce 1 to 5 volts across a selected load resistor. However a more important consideration is the temperature coefficient of resistance of the resistor. For example, lets say the 1 to 5 volts represents a temperature span of 0 to 400 deg C. If the load resistor is copper at 250 ohms at 25 deg C. It has a TC of 3.9x10^(-3) ohm/deg C. If the self heating effect or the ambient rises to 30 deg C. The resistor is now at 254.875 ohms. The 0 deg process temperature will now be read as 1.012 volts instead of 1 volt. Since you are expecting a 400 deg rise over 4 volts or 100 deg per volt, you have just introduced an error of 1.2 deg C.

Erich Mertz
[email protected]

Do you mean to buy 270 ohms resistors, and calibrate your transmitters to an electrical output range of 3.70 to 18.518 mA ???

Sorry, but if that is what you meant, I am SHOCKED. The 4 to 20 mA standard is so embedded into my mind, that I can't even see all the implications on: instrument calibration, bench and field calibrators, changing consolidated years of training/procedures taught to field/panel
maintenance personnel, how to proceed when hiring third party extra labor to help during plant maintenance, etc.

Filling the data-sheet to buy a new transmitters could be a challenge. Usually there is a check-box saying 4-20, and I don't know if the software that generates the data sheet will allow you to change that, except maybe for 0-20 or 10-50 mA or any other older "standard".

I'd rather destandardize the "right side of the street" to drive within a little city. Then watch what happens when people from abroad enter this
'English Way' of driving area, or your own citizens, needing to change sides in the highways, as they move away from the city.... Old drivers, or sleepy drivers drifting to the right, instead of remaining in the left side of roads, ... Accidents are sure prone to multiply themselves in such environment.

Sorry, but, IMHO, there are standards so consolidated by use, you should not play around with them.

Vitor Finkel [email protected]
P.O. Box 16061 Tel (+55) 21 2285-5641
22221.971 Rio de Janeiro Brazil Fax (+55) 21 2205-3339
Off the shelf 1% metal film resistors should be entirely accurate enough for this application and the tempco is speced, if not zero. Very, very, little of the analog stuff is capable of maintaining .1% accuracy over coffee break let alone long term. Use the 1%, tweak it and call it good. If that is the largest error term, I'll buy the coffee.


That's a very good point about the resistor temperature coefficient. I was figuring that's why Mark was specifying 1/2 watt resistors (to limit the self-heating effect), but of course that doesn't address ambient temperature variations.

It put me in mind of older tube type (and, I would expect, newer design as
well) frequency generators, which placed their timebase crystals in an "oven" to maintain them at a constant temperature a little bit above ambient, and thus limit frequency variations caused by temperature effects. This prompted a question which someone on the list might be able to answer.

Has anybody seen Peltier thermocouple arrays set up to create a stable temperature environment for this purpose?

What I envision is a closed enclosure where temperature sensitive components are located. A fan would be inside the enclosure to minimize local hot and cold spots. One or two of the enclosure walls would be Peltier thermocouple arrays with the 'cold' side in. A proportional controller would monitor enclosure ambient temperature, and use PWM to proportion power (and flip polarity as necessary) to maintain temperature at setpoint. Some safeguard against cooling below dewpoint would need to be incorporated (or water in the air inside the enclosure would condense ... unless it was hermetically sealed, and filled with a dried gas).

I'm not necessary suggesting this - the additional cost is probably prohibitive in most applications - just wondering if this approach has ever been used.


Dr. R. Shashidhar

I came across your mail by chance. I am looking for 1% MFRs and 2% polypropylene box capacitors of various values. could you suggest where I
can buy them in India through mail order (I require about 100 peices in all)

[email protected]