I have tried all of the major distributors for a 90k 1/4w 1% resistor and no one has responded. Metal film or foil or wirewound, I don't care.

Does anyone know of a source for these resistors, I need 100 right away for a project.

 

Hello,

Take two 180K resistors parallel.

Bertus

 

Or take 22kΩ+68kΩ in series. They would need to be 0.5% tolerance to meet the 1% tolerance requirement. Ditto for the parallel case as well.

 

is 90.9K acceptable? That's only 1% off, within the tolerance of the resistor.
http://www.mouser.com/ProductDetail...=sGAEpiMZZMtlubZbdhIBIHemO/ugtEnd75FxrvxLIxc=

or, here's 90Kohm, but they're 40c each
http://www.mouser.com/ProductDetail...GAEpiMZZMtlubZbdhIBIMuXIz2kMfhR5%2bhbFivE/Yk=

 

I don't see the need for two 0.5% resistors to get a 1% tolerance combination. Combining resistors of a given tolerance in series or parallel gives a result of the same tolerance worst-case. In fact, the statistical distribution of the combined result tends to be tighter, as to obtain a worst-case result requires two devices to be at their tolerance limit.

It is true that the worst-case percentage tolerance error given by a large ratio potential divider can approach double that of the resistors it is made from. That can be relevant if you are making something like an attenuator.

 

Yes Adjuster - I see my error. Thanks.

 

293-9.1m-rc
9.1 meg, 5%
9 cents at Mouser.
90.9k//9.1M = 90.001k

 

if the resister is used for DCDC's feedback, you can change the both resistence same time.

 

90.9k is the standard value for 1%. Strantor's Vishay 0.1% at Mouser for $0.38 is probably it for you.

 

It is a good thing for designer and/or hobbyist to be aware of and take into consideration 'standard' values for components to save from having to undergo your type of frustration. What if there were no standard values that could be manufactured and readily available. And of course as has been pointed out in the above responses, it is also a good thing to be familiar with the myriad of ways to get intermediate values when it imperative to have one. And IMO a third good thing is to know how to recognize whether it is in fact imperative or not.

 

I don't see the need for two 0.5% resistors to get a 1% tolerance combination. Combining resistors of a given tolerance in series or parallel gives a result of the same tolerance worst-case. In fact, the statistical distribution of the combined result tends to be tighter, as to obtain a worst-case result requires two devices to be at their tolerance limit.

Not necessarily true. Read this thread. Combined resistors of 0.1% tolerance can give up to 0.2% error worst case!

 

An older technician I knew told me that they used to file away part of the element on large carbon composition resistors to adjust them up in value.

 

An older technician I knew told me that they used to file away part of the element on large carbon composition resistors to adjust them up in value.

Up in value, but down in wattage rating.
Pat - W9ZO

 

Not necessarily true. Read this thread. Combined resistors of 0.1% tolerance can give up to 0.2% error worst case!

Yes, but that is precisely the division ratio error issue that I referred to in the second part of my post. You can indeed get a divider ratio error percentage approaching the sum of the two resistance tolerance percentage errors, but that is not the same situation as building a single composite resistance.

 

Ok, I don't see how you can add the 2 percentages. practical example:
I want a 600Ω resistor ±1% but all I have are (2) 1.2KΩ ±1%
So, 600 + 1% (worst case) = 606Ω
and, 1,200Ω +1% (worst case) = 1212Ω
I parallel the (2) 1212Ω resistors and I get 606Ω - That's still 1%.

Actually I would think that you would have a better chance of hitting your mark by parallelling resistors because, unless both are at the max tolerance, one should pull the other closer to the correct value.


EDIT, ok I see that you are talking about 2 different resistance value. never mind

 

By the way, The principle of the 9.1 meg resistor in parallel is that it is about 100 times the needed value, so its 5% tolerance could only affect the final product by .05%

 

Up in value, but down in wattage rating.
Pat - W9ZO

From the land of thick film hybrids when designing resistors, which are trimmed to value by cutting into them there is no factor for the kerf.

 

I have a bad habit ?? of putting an X fixed resistor parallel across an X trimmer pot, and dialing in what I actually need.

It might look a little odd and cobbled, though it certainly works. Nailing the tolerance needed can be tricky, since the little trimmers are none too tight.

It helps to have a shoebox full of trimmers..............

 

We should be asking Badbaud why he needs a 90k resistor. There is frequently another solution if the big picture is seen.