equations round evaluated values to 6 decimal places

Yes - in fact that's what I did - I'm just surprised that equations have this limitation - why wouldn't it use the same system tolerance as everything else? My foggy memory remembers running into this same issue last winter.

Thanks,

Zander

That70sTick wrote:

Now that you mention it, I o remember the same thing.

Are yousure the value is only evaluated to 6 places? Maybe it is evaluated further but only displayed to 6 places. Many calculators are like that.

If l> Yes - in fact that's what I did - I'm just surprised that equations

"Zander" wrote in news:1159730196.764873.258950 @h48g2000cwc.googlegroups.com:

Are you sure that this is not due to the conversion between radians and degrees? 6 places is a helluva lot for an angle.

Daisy.

"Zander" wrote in news: snipped-for-privacy@e3g2000cwe.googlegroups.com:

What I'm saying is that SW works in radians, and the equations box gives you some options when it comes to radians or degrees.

6.28~ radians = 360 degrees 1 radian = 57~ degrees 1x10e-8 radians = 5.7x10e-7 degrees

So theres 1.5 orders of magnitude difference in the conversion, which is close enough to look like 2 places difference.

So if it uses the system precision with radians to 8 places, its not hard to see how it would only use 6 when converted to degrees because the other significant digits are shifted to the left.

I don't know if this is the real explanation, but it makes sense.Try doing your angle dimensions in radians to 8 places and see if your equations are still rounded to 6 places. I'll bet that solves the problem.

Daisy.

To simplify, its like doing your finances in cents instead of dollars. When you do them in dollars, you have 2 decimal places, but when you do them in cents, you work in whole numbers. You have the same number of significant digits in both cases, but fewer decimal places when working in cents. You can't just make up the other two significant digits.

$100.00 is every bit as precise as 10000 cents.

Daisy.

Hi FlowerPot,

I should have explained more..., I'm simply computing a number with an equation and then using the number (to 8 decimal) places to control a sequence of drafts in the model. The equation editor has no knowledge of my devious plans. I'm simply performing some math that results in an answer the runs out to 8 decimal places. But it's rounding it to 6

- which introduces rounding errors in my part. Of course I don't actually need the accuracy for manufacturing itself but rather to eliminate rounding errors in my model.

Zander

FlowerPot wrote:

"Zander" wrote in news: snipped-for-privacy@m73g2000cwd.googlegroups.com:

I'm not sure how that explains anything or makes any difference.

Here's an interesting thing.

if you take a number, say 3.14159265 and divide it by 100000000, and only use 8 decimal places, the number becomes 0.00000000. If you then multiply it by 100000000, it becomes 3.14159265, which suggests it is keeping the significant digits even if they go below the 8 place limit.

I did notice some funny results though, when switching between radians and degrees. The last two digits of the degrees were incorrect, even without using equations, just using units conversion.

Can you be explicit about the equations you are using and the numbers you are getting? Posting a model would be great.

Daisy

Well, I thought it explained everything but here goes:

I'm evaluating a cosine to a value of .17226581 (when rounded to 8 decimal places). That's if I use a calculator. I have another part where this slope is duplicated and I simply want the end diameters to be equal. If I use an equation to evaluate the same number I get .172266 which is the same answer rounded to 6 decimal places. All I was saying was that the equation editor should at least respect solidworks own internal system tolerance of 8 decimal places - seems logical.

In the other part this slope is a driven value so I'm simply trying to maintain consistancy.

Thanks,

Zander

FlowerPot wrote: