# Foil resistors

Thread Starter

#### cmartinez

Joined Jan 17, 2007
6,644
I'm trying to choose a resistor that would be the best choice for an instrumentation amplifier gain resistor. And after doing a little bit of googling, I found that the best type of resistor (at least concerning its temperature coefficient) is of the foil type.
But the thing is, those suckers are listed at Digikey at more than $10.00 dlls a piece! They're almost twice as expensive as the amplifier chip itself! I plan to use this circuit as a load cell amplifier. Question, is using this type of resistor overkill? Are there any other more reasonable alternatives out there? #### #12 Joined Nov 30, 2010 18,167 Backwards working, you are. Instead of trying to find the best resistor on the planet, do the math. Find your error budget. Have you done that? Lots of resistors are in parts per million per centigrade. How tight do you need it? Thread Starter #### cmartinez Joined Jan 17, 2007 6,644 Backwards working, you are. Instead of trying to find the best resistor on the planet, do the math. Find your error budget. Have you done that? Lots of resistors are in parts per million per centigrade. How tight do you need it? Good point, master Jedi... guess I have to first study all the alternatives and see how they affect the end result... Thing is, I'm feeling a bit lazy, since it's almost 10 PM down here, so I thought I'd just ask... but on the other hand, no one knows my intended application better than myself. A bit of work doing some research is unavoidable, I guess. Thanks for your input. Gonna hit the sac... nighty night. #### #12 Joined Nov 30, 2010 18,167 It's eleven PM here. I had to stop work because the parts store closed right after they sold me the wrong brake shoes. On the other hand, I got the rear air conditioner working today! 25 F degrees colder coming out of the vents than going into them. I have dual air in Florida! Still have to make an adapter to hold the ceiling pod up. It contains the switches for the rear air and the interior lights. Darn nephew took the old one out and threw it away because one prong was broken. Now I have to make one for all 4 fastening points. Thread Starter #### cmartinez Joined Jan 17, 2007 6,644 It's eleven PM here. I had to stop work because the parts store closed right after they sold me the wrong brake shoes. On the other hand, I got the rear air conditioner working today! 25 F degrees colder coming out of the vents than going into them. I have dual air in Florida! Still have to make an adapter to hold the ceiling pod up. It contains the switches for the rear air and the interior lights. Darn nephew took the old one out and threw it away because one prong was broken. Now I have to make one for all 4 fastening points. Ahh..... there's your chance to design and build the touch-actuated, automotive mosfet relay board you've always dreamed of! Thread Starter #### cmartinez Joined Jan 17, 2007 6,644 Ahh..... there's your chance to design and build the touch-actuated, automotive mosfet relay board you've always dreamed of! EDIT: that... or make your nephew pay for the bloody part he threw away without consulting you first! #### #12 Joined Nov 30, 2010 18,167 Ahh..... there's your chance to design and build the touch-actuated, automotive mosfet relay board you've always dreamed of! Mostly, I am eliminating microcomputer controlled devices. Most of the problems with this car are wrong (or absence of) outputs from microcontrollers. Lights stuck on? Cut the wire and turn the light into a "manual only" control. Flappers don't work in the air controllers? Unplug the motors, jerk the flapper into the "air conditioning only" position and put a screw in the bearing so it can never move again. Dealer wants$130 for an ignition key with a microchip in it? Teach the car to use a $3 key. #### Alec_t Joined Sep 17, 2013 10,598 Darn nephew took the old one out and threw it away because one prong was broken. That's typical of the younger generation . Old geezers like me can't bear to throw things out if there's any prospect of fixing them. #### #12 Joined Nov 30, 2010 18,167 While there is very little chance of repairing a broken plastic pin intended for an insertion clip, completely replacing it is within my abilities. Replacing 4 of them is just 4 times as much work. I could wish Our Father who art in Detroit would use fasteners that can function more than once, but that's just a fantasy. I have already done the measurements and found the right materials to make new attachment prongs, but it's going to take all day to make 4 of them. I consider myself lucky that the background for this rig contains plenty of sheet metal structure I can put screws in. Hmmm...That's an idea. I could just put screws through the front of the instrument pod, except it would look like it was designed in Kentucky. Low class craftsmanship is not what I want on display. Why am I not installing the brake shoes? It's raining today. Weather RADAR says the sky will clear any minute and the streets will be dry in an hour. I just hate driving among the winter tourists when the streets are slippery! #### ronv Joined Nov 12, 2008 3,770 Good point, master Jedi... guess I have to first study all the alternatives and see how they affect the end result... Thing is, I'm feeling a bit lazy, since it's almost 10 PM down here, so I thought I'd just ask... but on the other hand, no one knows my intended application better than myself. A bit of work doing some research is unavoidable, I guess. Thanks for your input. Gonna hit the sac... nighty night. Maybe this will help. Thread Starter #### cmartinez Joined Jan 17, 2007 6,644 Thread Starter #### cmartinez Joined Jan 17, 2007 6,644 Alright... just for the sake of completing this thread, I did some calculations using Ron's table, plus a couple of other types of (really cheesy) resistors, and here's what I found. This is the theoretical deviation (in Kg) if a 100Kg load cell were to be used (with an applied force of 100Kg) connected to the instrumentation amplifier that I originally chose, with a temperature change of +30°C. Again, this is the deviation when only the RG's TCR is considered, and it doesn't take into account the load cell's TCR itself or other factors, such as wiring, or drift in excitation voltage. My conclusion is that either a wirewound or high precision thin film resistor are an acceptable alternative to the luxurious bulk metal foil type. A 15g variation on a 100Kg measurement is not that bad. Last edited: #### ronv Joined Nov 12, 2008 3,770 Alright... just for the sake of completing this thread, I did some calculation's using Ron's table, plus a couple of other types of (really cheesy) resistors, and here's what I found. This is the theoretical deviation (in Kg) if a 100Kg load cell were to be used (with an applied force of 100Kg) connected to the instrumentation amplifier that I originally chose, with a temperature change of +30°C. Again, this is the deviation when only the RG's TCR is considered, and it doesn't take into account the load cell's TCR itself or other factors, such as wiring, or drift in excitation voltage. My conclusion is that either a wirewound or high precision thin film resistor are an acceptable alternative to the luxurious bulk metal foil type. A 15g variation on a 100Kg measurement is not that bad. Depends on what your weighing. Your not selling drugs again are you? #### #12 Joined Nov 30, 2010 18,167 Depends on what your weighing. Your not selling drugs again are you? Either that or he's sending something to the International Space Station. Only NASA and the DEA would care about that kind of precision. #### WBahn Joined Mar 31, 2012 24,974 Good point, master Jedi... guess I have to first study all the alternatives and see how they affect the end result... Thing is, I'm feeling a bit lazy, since it's almost 10 PM down here, so I thought I'd just ask... but on the other hand, no one knows my intended application better than myself. A bit of work doing some research is unavoidable, I guess. Thanks for your input. Gonna hit the sac... nighty night. One thing you want to consider is what is more important -- that the resistor values be absolutely accurate and don't change value, or that if they do change value that they are well enough matched that they change by very close to the same amount? In practice it will be a combination of the two, but which dominates will vary. Thread Starter #### cmartinez Joined Jan 17, 2007 6,644 Depends on what your weighing. Your not selling drugs again are you? Either that or he's sending something to the International Space Station. Only NASA and the DEA would care about that kind of precision. Gee... thanks for your support, guys... I just love to watch the expression of annoyance on that horse's face Last edited: #### #12 Joined Nov 30, 2010 18,167 After I went to bed, I remembered...the list should have been, "NASA, the DEA, and WBahn would care about that kind of precision." I wake up today and find a reply from WBahn. But seriously, I believe @cmartinez is capable of doing the math for both initial accuracy and drift errors. I just needed to remind him to calculate an error budget. ps, got the brakes done in a quick hour, then found out this car does not have a transmission dip stick or fill tube. It was designed to run until it got a fluid leak, burned out the clutches, and then you buy a new transmission. Guess who is going to change that. I am not only going to fix the drip at the drain plug, I am going to devise a way to check the fluid level and add fluid if necessary. Just, not today. I've been rolling under that car for 3 days in a row and I'm deeply tired. That's the kind of day when you hurt yourself so, today is cancelled for physical work. #### KL7AJ Joined Nov 4, 2008 2,224 I'm trying to choose a resistor that would be the best choice for an instrumentation amplifier gain resistor. And after doing a little bit of googling, I found that the best type of resistor (at least concerning its temperature coefficient) is of the foil type. But the thing is, those suckers are listed at Digikey at more than$10.00 dlls a piece! They're almost twice as expensive as the amplifier chip itself!

I plan to use this circuit as a load cell amplifier.

Question, is using this type of resistor overkill? Are there any other more reasonable alternatives out there?
Well, you get what you pay for.
Instrumentation design is a whole 'nother tomato than any other kind of electronics. That is because an instrument has to be an order of magnitude more accurate than anything you're measuring with it!
It's TOUGH design work...but very rewarding. And if you know how to do it, you have pretty good job security!
Eric