whats the deal with these values?

[Eddododo]

http://www.rane.com/pdf/old/pe15sch.pdf

an interesting parametric circuit from Rane

There are a lot of uncommon values... .39uF,   90.9K resistors and plenty more.

from a glance at the layout, there dont appear to be parallel / series combinations of standard values..

[Eddododo]

followup:
I understand the approach of using exacting math to fine-tune nice gear (im assuming thats whats going on here) but why would you so finely tune something as varying as a parametric eq?

[Seljer]

followup:
I understand the approach of using exacting math to fine-tune nice gear (im assuming thats whats going on here) but why would you so finely tune something as varying as a parametric eq?

and use them on the current-limiting resistors for the LEDs as well 

[R.G.]

followup:
I understand the approach of using exacting math to fine-tune nice gear (im assuming thats whats going on here) but why would you so finely tune something as varying as a parametric eq?

I didn't read the document, but in many filter circuits, there is a high _sensitivity_ to the ratios of the values of the components. For instance, for a given filter response, two caps may have to be in a 2:1 ratio *exactly* to get results as expected. Ditto for resistors; some circuits need quads of resistors matched to 0.1% or better for their objectives, or two pairs of exactly matched values the pairs of which are in some exact ratio. Something specifying capacitors  with 0.5% tolerance is more manufacturable than demanding 5% or 1% resistors hand-matched to an exact ratio.

So - to get the right shape, they may have needed more-exact values, or value ratios, so they may have  specified low-tolerance parts instead of selection.

Or, their lab may just have a whole wall full of high-precision caps and resistors.

[Seljer]

Is there any merit to the idea that theres a smaller chance of somebody selling not-up-to-spec/counterfeit 90.9k low-tolerance resistors than a 100k low-tolerance resistors?

or at least a smaller chance of whoever is soldering it together to not-mix up resistors of different tolerances? Kind of like the Van Halen no brown M&Ms in the tour rider thing, which was there to make sure whoever was hosting was paying attention to all the details.

[amptramp]

There is a standard set of 1% resistor values:

10.0 14.7 21.5 31.6 46.4 68.1
10.2 15.0 22.1 32.4 47.5 69.8
10.5 15.4 22.6 33.2 48.7 71.5
10.7 15.8 23.2 34.0 49.9 73.2
11.0 16.2 23.7 34.8 51.1 75.0
11.3 16.5 24.3 35.7 52.3 76.8
11 5 16.9 24.9 36.5 53.6 78.7
11.8 17.4 25.5 37.4 54.9 80.6
12.1 17.8 26.1 38.3 56.2 82.5
12.4 18.2 26.7 39.2 57.6 84.5
12.7 18.7 27.4 40.2 59.0 86.6
13.0 19.1 28.0 41.2 60.4 88.7
13.3 19.6 28.7 42.2 61.9 90.9
13.7 20.0 29.4 43.2 63.4 93.1
14.0 20.5 30.1 44.2 64.9 95.3
14.3 21.0 30.9 45.2 66.5 97.6

As you can see, 90.9 is one of those values and it can be multiplied by powers of ten to provide the resistance value you need.

Capacitors tend to follow the 10% resistor values which are:

10
12
15
18
22
27
33
39
47
56
68
82

[R.G.]

google "eia 24"

You'll find the preferred sequence values for component values. Resistors have quite fine resolutions available. Capacitors do not.  They tend to use the lower EIA sequences.

[Eddododo]

I suppose these exacting demands of precision are why you dont find a lot of fully parametric eq's on this [and other] diy sites?

thanks for the info guys, as always

[PRR]

> 90.9K resistors

90.9K is another way to say "One Percent!!"

The assemblers can't find any "90.9" in the 10% and 5% bins, they are forced to go for 1% parts.

Why? In a many-stage path in a Stereo signal, "small" tolerances *can* add-up to significant difference between channels. Not even looking at the bandpass blocks, I see four gain-stages from In to Out. Suppose two gain-set resistors were just 1% off but in opposite directions. Gain error may be 2%. Now suppose four stages all had a 2% error the same way. Error is 8% (1.0824). That's almost 0.7dB off. Now suppose we really got "lucky" and built a second unit with all the errors the other way. Gain-difference between left and right is 1.4dB. In a good monitoring studio, that's a very significant gain-shift and thus image-shift. Enought to make the user annoyed with the builder. (Fortunately the odds of 16 resistors ALL falling to a worst-difference value is lower than winning the PowerBall.)

Four worst-case stages with 10% parts could be almost 7dB off either way. (So even home stereos use 5% parts and give you a Balance knob.)

That does not directly explain the 1% on the LED-setting resistor.

> not-up-to-spec/counterfeit

Bah, bah.

The plan is old. RANE was probably buying resistors directly traceable to Bourns or A-B or other no-doubt-good resistor-shop.

Companies somewhat larger than RANE have a whole Qualification department. They check part samples with the specs and the designer, disqualify sources that don't work good (whether bad parts or just not happy in the circuit), then re-test each batch delivered. RANE is a little smaller and Dennis (or his production cohort) may have just been all over any new box of parts or any report of trouble at check-out.

And the 1% LED part suggests that at some point he threw-out ALL the >1% parts and mandated 1% best-brand everywhere.

> why you dont find a lot of fully parametric eq's on ...diy sites?

Nah, you can build a para-EQ with 10% parts. Not really a big problem until you try to build a second "identical" path.

Mostly it is a BIG job to do RIGHT. The boost/cut pots should be special taper. You end up with so many knobs that you wish for coaxial pots/switched (not readily available). R.G.'s simplified plan allows Q to vary with F, which looks bad in the ad. And if you really ned a good para-EQ, they turn-up in studio surplus often enough.

[Eddododo]

I love this site