I am using the "Mouthpiece Facing Curve Analysis" (credited to Mojo & A. Donaldson) in Neo Office, and I want to start using the solver to calculate best fit curves. I believe it works very similar to Excel, so I hope that will not be a complicating issue.

What I'm having difficulty getting my head around is how the solver works. My math skills haven't been used in 20 years, and I've never used a spreadsheet solver before.

Is there anyone willing to walk me through it?

This is where I'm at:

1) I set the Error Sum cell as the "Target Cell" and check on "minimum."

2) Where it says "By changing cells" I think I need to highlight cell "Elliptical Radius."

When I try this, I get "Err:502" in many of the "Calculated Curve Fit" and "Diff" columns, which I believe means I'm getting multiple possible solutions.

3) Then there are the "Limiting Conditions." I'm guessing I use these limits to get only one solution?

Thanks in advance for any help!

I am not familiar with Neo Office.  
 
There are 3 paramaters that need to be solved to get an ellipse of best fit for a set of data.  It depends on how the sheet is set up as to which 3 parameters these are.
 
I went looking for the spreadsheet you are trying to use but I could not find one with an error sum cell in it.  What is the file name and where is it located?



________________________________
From: Mike <mike@mikewilkens.com>
To: MouthpieceWork@yahoogroups.com 
Sent: Friday, April 26, 2013 8:44 PM
Subject: [MouthpieceWork] Solver use in Spreadsheet Application

  
I am using the "Mouthpiece Facing Curve Analysis" (credited to Mojo & A. Donaldson) in Neo Office, and I want to start using the solver to calculate best fit curves. I believe it works very similar to Excel, so I hope that will not be a complicating issue.

What I'm having difficulty getting my head around is how the solver works. My math skills haven't been used in 20 years, and I've never used a spreadsheet solver before.

Is there anyone willing to walk me through it?

This is where I'm at:

1) I set the Error Sum cell as the "Target Cell" and check on "minimum."

2) Where it says "By changing cells" I think I need to highlight cell "Elliptical Radius."

When I try this, I get "Err:502" in many of the "Calculated Curve Fit" and "Diff" columns, which I believe means I'm getting multiple possible solutions.

3) Then there are the "Limiting Conditions." I'm guessing I use these limits to get only one solution?

Thanks in advance for any help!

It's the "Facing Curve Worksheet." I believe I got it from the Yahoo Workgroup Files, but I'm having trouble finding it again, so maybe I got it elsewhere.

I'll e-mail it to you so we can be sure to be on the same page.

So far I have just been playing with the Elliptical Radius and the Curve Start numbers to get the best match and thus minimize the error sum. I'm guessing this is what the solver does, (only better!)

When I try to do this I get an error message saying "The model is not linear."

Thanks!

--- In MouthpieceWork@yahoogroups.com, Keith Bradbury <kwbradbury@...> wrote:
>
> I am not familiar with Neo Office.  
>  
> There are 3 paramaters that need to be solved to get an ellipse of best fit for a set of data.  It depends on how the sheet is set up as to which 3 parameters these are.
>  
> I went looking for the spreadsheet you are trying to use but I could not find one with an error sum cell in it.  What is the file name and where is it located?
> 
> 
> 
> ________________________________
> From: Mike <mike@...>
> To: MouthpieceWork@yahoogroups.com 
> Sent: Friday, April 26, 2013 8:44 PM
> Subject: [MouthpieceWork] Solver use in Spreadsheet Application
> 
>   
> I am using the "Mouthpiece Facing Curve Analysis" (credited to Mojo & A. Donaldson) in Neo Office, and I want to start using the solver to calculate best fit curves. I believe it works very similar to Excel, so I hope that will not be a complicating issue.
> 
> What I'm having difficulty getting my head around is how the solver works. My math skills haven't been used in 20 years, and I've never used a spreadsheet solver before.
> 
> Is there anyone willing to walk me through it?
> 
> This is where I'm at:
> 
> 1) I set the Error Sum cell as the "Target Cell" and check on "minimum."
> 
> 2) Where it says "By changing cells" I think I need to highlight cell "Elliptical Radius."
> 
> When I try this, I get "Err:502" in many of the "Calculated Curve Fit" and "Diff" columns, which I believe means I'm getting multiple possible solutions.
> 
> 3) Then there are the "Limiting Conditions." I'm guessing I use these limits to get only one solution?
> 
> Thanks in advance for any help!
>

After a little more playing around and research I think I understand my problem and figured I would post it here in case it helps anyone else.

I downloaded a free trial of Excel and after adding on the solver feature it worked exactly like it is supposed to. While doing this I realized there are different solver settings in the Excel version, including the "GRG Non-linear" solver, which is the one to use for this type of equation.

I am trying to find a free spreadsheet program that can do this, but (as far as I know) google docs and NeoOffice don't have non-linear solver features.

OpenOffice (free open-source software) has an extension (add-on) called Solvit which apparently has a non-linear equation solver (some details at http://extensions.openoffice.org/en/node/4659 ) Although I haven't been able to get it to solve for facing curves.

Is there anyone here that has gotten OpenOffice Solvit to work solving facing curves? It would be great to have a free working solution for this.

In the meantime I'll be using the one month free demo version of MS Office.

Thanks in advance for any help!

--- In MouthpieceWork@yahoogroups.com, "Mike" <mike@...> wrote:
>
> It's the "Facing Curve Worksheet." I believe I got it from the Yahoo Workgroup Files, but I'm having trouble finding it again, so maybe I got it elsewhere.
> 
> I'll e-mail it to you so we can be sure to be on the same page.
> 
> So far I have just been playing with the Elliptical Radius and the Curve Start numbers to get the best match and thus minimize the error sum. I'm guessing this is what the solver does, (only better!)
> 
> When I try to do this I get an error message saying "The model is not linear."
> 
> Thanks!
> 
> --- In MouthpieceWork@yahoogroups.com, Keith Bradbury <kwbradbury@> wrote:
> >
> > I am not familiar with Neo Office.  
> >  
> > There are 3 paramaters that need to be solved to get an ellipse of best fit for a set of data.  It depends on how the sheet is set up as to which 3 parameters these are.
> >  
> > I went looking for the spreadsheet you are trying to use but I could not find one with an error sum cell in it.  What is the file name and where is it located?
> > 
> > 
> > 
> > ________________________________
> > From: Mike <mike@>
> > To: MouthpieceWork@yahoogroups.com 
> > Sent: Friday, April 26, 2013 8:44 PM
> > Subject: [MouthpieceWork] Solver use in Spreadsheet Application
> > 
> >   
> > I am using the "Mouthpiece Facing Curve Analysis" (credited to Mojo & A. Donaldson) in Neo Office, and I want to start using the solver to calculate best fit curves. I believe it works very similar to Excel, so I hope that will not be a complicating issue.
> > 
> > What I'm having difficulty getting my head around is how the solver works. My math skills haven't been used in 20 years, and I've never used a spreadsheet solver before.
> > 
> > Is there anyone willing to walk me through it?
> > 
> > This is where I'm at:
> > 
> > 1) I set the Error Sum cell as the "Target Cell" and check on "minimum."
> > 
> > 2) Where it says "By changing cells" I think I need to highlight cell "Elliptical Radius."
> > 
> > When I try this, I get "Err:502" in many of the "Calculated Curve Fit" and "Diff" columns, which I believe means I'm getting multiple possible solutions.
> > 
> > 3) Then there are the "Limiting Conditions." I'm guessing I use these limits to get only one solution?
> > 
> > Thanks in advance for any help!
> >
>

It is possible to set up a spreadsheet for iterative calculations.  I used to do it decades ago before built-in solvers were available.  It is a lot more involved and more things can go wrong.

I would suggest surfing eBay and look for some deals on MS Office.   I use a 2003 version still.  Some are there under $50.

On May 6, 2013, at 4:39 AM, "Mike" <mike@...> wrote:

> After a little more playing around and research I think I understand my problem and figured I would post it here in case it helps anyone else.
> 
> I downloaded a free trial of Excel and after adding on the solver feature it worked exactly like it is supposed to. While doing this I realized there are different solver settings in the Excel version, including the "GRG Non-linear" solver, which is the one to use for this type of equation.
> 
> I am trying to find a free spreadsheet program that can do this, but (as far as I know) google docs and NeoOffice don't have non-linear solver features.
> 
> OpenOffice (free open-source software) has an extension (add-on) called Solvit which apparently has a non-linear equation solver (some details at http://extensions.openoffice.org/en/node/4659 ) Although I haven't been able to get it to solve for facing curves.
> 
> Is there anyone here that has gotten OpenOffice Solvit to work solving facing curves? It would be great to have a free working solution for this.
> 
> In the meantime I'll be using the one month free demo version of MS Office.
> 
> Thanks in advance for any help!
> 
> --- In MouthpieceWork@yahoogroups.com, "Mike" <mike@...> wrote:
> >
> > It's the "Facing Curve Worksheet." I believe I got it from the Yahoo Workgroup Files, but I'm having trouble finding it again, so maybe I got it elsewhere.
> > 
> > I'll e-mail it to you so we can be sure to be on the same page.
> > 
> > So far I have just been playing with the Elliptical Radius and the Curve Start numbers to get the best match and thus minimize the error sum. I'm guessing this is what the solver does, (only better!)
> > 
> > When I try to do this I get an error message saying "The model is not linear."
> > 
> > Thanks!
> > 
> > --- In MouthpieceWork@yahoogroups.com, Keith Bradbury <kwbradbury@> wrote:
> > >
> > > I am not familiar with Neo Office.  
> > > Â 
> > > There are 3 paramaters that need to be solved to get an ellipse of best fit for a set of data.  It depends on how the sheet is set up as to which 3 parameters these are.
> > > Â 
> > > I went looking for the spreadsheet you are trying to use but I could not find one with an error sum cell in it.  What is the file name and where is it located?
> > > 
> > > 
> > > 
> > > ________________________________
> > > From: Mike <mike@>
> > > To: MouthpieceWork@yahoogroups.com 
> > > Sent: Friday, April 26, 2013 8:44 PM
> > > Subject: [MouthpieceWork] Solver use in Spreadsheet Application
> > > 
> > > Â  
> > > I am using the "Mouthpiece Facing Curve Analysis" (credited to Mojo & A. Donaldson) in Neo Office, and I want to start using the solver to calculate best fit curves. I believe it works very similar to Excel, so I hope that will not be a complicating issue.
> > > 
> > > What I'm having difficulty getting my head around is how the solver works. My math skills haven't been used in 20 years, and I've never used a spreadsheet solver before.
> > > 
> > > Is there anyone willing to walk me through it?
> > > 
> > > This is where I'm at:
> > > 
> > > 1) I set the Error Sum cell as the "Target Cell" and check on "minimum."
> > > 
> > > 2) Where it says "By changing cells" I think I need to highlight cell "Elliptical Radius."
> > > 
> > > When I try this, I get "Err:502" in many of the "Calculated Curve Fit" and "Diff" columns, which I believe means I'm getting multiple possible solutions.
> > > 
> > > 3) Then there are the "Limiting Conditions." I'm guessing I use these limits to get only one solution?
> > > 
> > > Thanks in advance for any help!
> > >
> >
> 
> 

This sounds pretty nice but i wonder if you dont get just as good results "eyeballing" the curve against the measurements curve in a chart. Then manyually varying the parameters of the elipse. Ive seen a case or 2 when a point is clearly an anomally and i wouldnt want the solver to include it in the optimization.You're still fitting an elipse to get a smooth curve but you're anomally points could have a significantly bigger error than the other points.

--- In MouthpieceWork@yahoogroups.com, Keith Bradbury <kwbradbury@...> wrote:
>
> It is possible to set up a spreadsheet for iterative calculations.  I used to do it decades ago before built-in solvers were available.  It is a lot more involved and more things can go wrong.
> 
> I would suggest surfing eBay and look for some deals on MS Office.   I use a 2003 version still.  Some are there under $50.
> 
> On May 6, 2013, at 4:39 AM, "Mike" <mike@...> wrote:
> 
> > After a little more playing around and research I think I understand my problem and figured I would post it here in case it helps anyone else.
> > 
> > I downloaded a free trial of Excel and after adding on the solver feature it worked exactly like it is supposed to. While doing this I realized there are different solver settings in the Excel version, including the "GRG Non-linear" solver, which is the one to use for this type of equation.
> > 
> > I am trying to find a free spreadsheet program that can do this, but (as far as I know) google docs and NeoOffice don't have non-linear solver features.
> > 
> > OpenOffice (free open-source software) has an extension (add-on) called Solvit which apparently has a non-linear equation solver (some details at http://extensions.openoffice.org/en/node/4659 ) Although I haven't been able to get it to solve for facing curves.
> > 
> > Is there anyone here that has gotten OpenOffice Solvit to work solving facing curves? It would be great to have a free working solution for this.
> > 
> > In the meantime I'll be using the one month free demo version of MS Office.
> > 
> > Thanks in advance for any help!
> > 
> > --- In MouthpieceWork@yahoogroups.com, "Mike" <mike@> wrote:
> > >
> > > It's the "Facing Curve Worksheet." I believe I got it from the Yahoo Workgroup Files, but I'm having trouble finding it again, so maybe I got it elsewhere.
> > > 
> > > I'll e-mail it to you so we can be sure to be on the same page.
> > > 
> > > So far I have just been playing with the Elliptical Radius and the Curve Start numbers to get the best match and thus minimize the error sum. I'm guessing this is what the solver does, (only better!)
> > > 
> > > When I try to do this I get an error message saying "The model is not linear."
> > > 
> > > Thanks!
> > > 
> > > --- In MouthpieceWork@yahoogroups.com, Keith Bradbury <kwbradbury@> wrote:
> > > >
> > > > I am not familiar with Neo Office.  
> > > > Â 
> > > > There are 3 paramaters that need to be solved to get an ellipse of best fit for a set of data.  It depends on how the sheet is set up as to which 3 parameters these are.
> > > > Â 
> > > > I went looking for the spreadsheet you are trying to use but I could not find one with an error sum cell in it.  What is the file name and where is it located?
> > > > 
> > > > 
> > > > 
> > > > ________________________________
> > > > From: Mike <mike@>
> > > > To: MouthpieceWork@yahoogroups.com 
> > > > Sent: Friday, April 26, 2013 8:44 PM
> > > > Subject: [MouthpieceWork] Solver use in Spreadsheet Application
> > > > 
> > > > Â  
> > > > I am using the "Mouthpiece Facing Curve Analysis" (credited to Mojo & A. Donaldson) in Neo Office, and I want to start using the solver to calculate best fit curves. I believe it works very similar to Excel, so I hope that will not be a complicating issue.
> > > > 
> > > > What I'm having difficulty getting my head around is how the solver works. My math skills haven't been used in 20 years, and I've never used a spreadsheet solver before.
> > > > 
> > > > Is there anyone willing to walk me through it?
> > > > 
> > > > This is where I'm at:
> > > > 
> > > > 1) I set the Error Sum cell as the "Target Cell" and check on "minimum."
> > > > 
> > > > 2) Where it says "By changing cells" I think I need to highlight cell "Elliptical Radius."
> > > > 
> > > > When I try this, I get "Err:502" in many of the "Calculated Curve Fit" and "Diff" columns, which I believe means I'm getting multiple possible solutions.
> > > > 
> > > > 3) Then there are the "Limiting Conditions." I'm guessing I use these limits to get only one solution?
> > > > 
> > > > Thanks in advance for any help!
> > > >
> > >
> > 
> >
>

It is easy to throw out a point you do not want to include in your fit.  Just delete the diff^2 calculation in the cell for that point.
 
The fit is a way to study a curve.  But how you determine a target curve can be very different.  
 

________________________________
From: megaohmer <megaohmer@...>
To: MouthpieceWork@yahoogroups.com 
Sent: Friday, May 10, 2013 4:06 PM
Subject: [MouthpieceWork] Re: Solver use in Spreadsheet Application

  
This sounds pretty nice but i wonder if you dont get just as good results "eyeballing" the curve against the measurements curve in a chart. Then manyually varying the parameters of the elipse. Ive seen a case or 2 when a point is clearly an anomally and i wouldnt want the solver to include it in the optimization.You're still fitting an elipse to get a smooth curve but you're anomally points could have a significantly bigger error than the other points.

--- In mailto:MouthpieceWork%40yahoogroups.com, Keith Bradbury <kwbradbury@...> wrote:
>
> It is possible to set up a spreadsheet for iterative calculations. I used to do it decades ago before built-in solvers were available. It is a lot more involved and more things can go wrong.
> 
> I would suggest surfing eBay and look for some deals on MS Office. I use a 2003 version still. Some are there under $50.
> 
> On May 6, 2013, at 4:39 AM, "Mike" <mike@...> wrote:
> 
> > After a little more playing around and research I think I understand my problem and figured I would post it here in case it helps anyone else.
> > 
> > I downloaded a free trial of Excel and after adding on the solver feature it worked exactly like it is supposed to. While doing this I realized there are different solver settings in the Excel version, including the "GRG Non-linear" solver, which is the one to use for this type of equation.
> > 
> > I am trying to find a free spreadsheet program that can do this, but (as far as I know) google docs and NeoOffice don't have non-linear solver features.
> > 
> > OpenOffice (free open-source software) has an extension (add-on) called Solvit which apparently has a non-linear equation solver (some details at http://extensions.openoffice.org/en/node/4659 ) Although I haven't been able to get it to solve for facing curves.
> > 
> > Is there anyone here that has gotten OpenOffice Solvit to work solving facing curves? It would be great to have a free working solution for this.
> > 
> > In the meantime I'll be using the one month free demo version of MS Office.
> > 
> > Thanks in advance for any help!
> > 
> > --- In mailto:MouthpieceWork%40yahoogroups.com, "Mike" <mike@> wrote:
> > >
> > > It's the "Facing Curve Worksheet." I believe I got it from the Yahoo Workgroup Files, but I'm having trouble finding it again, so maybe I got it elsewhere.
> > > 
> > > I'll e-mail it to you so we can be sure to be on the same page.
> > > 
> > > So far I have just been playing with the Elliptical Radius and the Curve Start numbers to get the best match and thus minimize the error sum. I'm guessing this is what the solver does, (only better!)
> > > 
> > > When I try to do this I get an error message saying "The model is not linear."
> > > 
> > > Thanks!
> > > 
> > > --- In mailto:MouthpieceWork%40yahoogroups.com, Keith Bradbury <kwbradbury@> wrote:
> > > >
> > > > I am not familiar with Neo Office. 
> > > > Â 
> > > > There are 3 paramaters that need to be solved to get an ellipse of best fit for a set of data. It depends on how the sheet is set up as to which 3 parameters these are.
> > > > Â 
> > > > I went looking for the spreadsheet you are trying to use but I could not find one with an error sum cell in it. What is the file name and where is it located?
> > > > 
> > > > 
> > > > 
> > > > ________________________________
> > > > From: Mike <mike@>
> > > > To: mailto:MouthpieceWork%40yahoogroups.com 
> > > > Sent: Friday, April 26, 2013 8:44 PM
> > > > Subject: [MouthpieceWork] Solver use in Spreadsheet Application
> > > > 
> > > > Â 
> > > > I am using the "Mouthpiece Facing Curve Analysis" (credited to Mojo & A. Donaldson) in Neo Office, and I want to start using the solver to calculate best fit curves. I believe it works very similar to Excel, so I hope that will not be a complicating issue.
> > > > 
> > > > What I'm having difficulty getting my head around is how the solver works. My math skills haven't been used in 20 years, and I've never used a spreadsheet solver before.
> > > > 
> > > > Is there anyone willing to walk me through it?
> > > > 
> > > > This is where I'm at:
> > > > 
> > > > 1) I set the Error Sum cell as the "Target Cell" and check on "minimum."
> > > > 
> > > > 2) Where it says "By changing cells" I think I need to highlight cell "Elliptical Radius."
> > > > 
> > > > When I try this, I get "Err:502" in many of the "Calculated Curve Fit" and "Diff" columns, which I believe means I'm getting multiple possible solutions.
> > > > 
> > > > 3) Then there are the "Limiting Conditions." I'm guessing I use these limits to get only one solution?
> > > > 
> > > > Thanks in advance for any help!
> > > >
> > >
> > 
> >
>

When you get it all figured out can you post a DIY guide on this tool?

I have used Excel back to Lotus 123 and even Open office for years and still have trouble with the mouthpiece spreadsheets and the solver.
Thanks!

> > On May 6, 2013, at 4:39 AM, "Mike" <mike@> wrote:
> > 
> > > After a little more playing around and research I think I understand my problem and figured I would post it here in case it helps anyone else.
> > > 
> > > I downloaded a free trial of Excel and after adding on the solver feature it worked exactly like it is supposed to. While doing this I realized there are different solver settings in the Excel version, including the "GRG Non-linear" solver, which is the one to use for this type of equation.
> > > 
> > > I am trying to find a free spreadsheet program that can do this, but (as far as I know) google docs and NeoOffice don't have non-linear solver features.
> > > 
> > > OpenOffice (free open-source software) has an extension (add-on) called Solvit which apparently has a non-linear equation solver (some details at http://extensions.openoffice.org/en/node/4659 ) Although I haven't been able to get it to solve for facing curves.
> > > 
> > > Is there anyone here that has gotten OpenOffice Solvit to work solving facing curves? It would be great to have a free working solution for this.
> > > 
> > > In the meantime I'll be using the one month free demo version of MS Office.
> > > 
> > > Thanks in advance for any help!
> > > 
> > > --- In MouthpieceWork@yahoogroups.com, "Mike" <mike@> wrote:
> > > >
> > > > It's the "Facing Curve Worksheet." I believe I got it from the Yahoo Workgroup Files, but I'm having trouble finding it again, so maybe I got it elsewhere.
> > > > 
> > > > I'll e-mail it to you so we can be sure to be on the same page.
> > > > 
> > > > So far I have just been playing with the Elliptical Radius and the Curve Start numbers to get the best match and thus minimize the error sum. I'm guessing this is what the solver does, (only better!)
> > > > 
> > > > When I try to do this I get an error message saying "The model is not linear."
> > > > 
> > > > Thanks!
> > > > 
> > > > --- In MouthpieceWork@yahoogroups.com, Keith Bradbury <kwbradbury@> wrote:
> > > > >
> > > > > I am not familiar with Neo Office.  
> > > > > Â 
> > > > > There are 3 paramaters that need to be solved to get an ellipse of best fit for a set of data.  It depends on how the sheet is set up as to which 3 parameters these are.
> > > > > Â 
> > > > > I went looking for the spreadsheet you are trying to use but I could not find one with an error sum cell in it.  What is the file name and where is it located?
> > > > > 
> > > > > 
> > > > > 
> > > > > ________________________________
> > > > > From: Mike <mike@>
> > > > > To: MouthpieceWork@yahoogroups.com 
> > > > > Sent: Friday, April 26, 2013 8:44 PM
> > > > > Subject: [MouthpieceWork] Solver use in Spreadsheet Application
> > > > > 
> > > > > Â  
> > > > > I am using the "Mouthpiece Facing Curve Analysis" (credited to Mojo & A. Donaldson) in Neo Office, and I want to start using the solver to calculate best fit curves. I believe it works very similar to Excel, so I hope that will not be a complicating issue.
> > > > > 
> > > > > What I'm having difficulty getting my head around is how the solver works. My math skills haven't been used in 20 years, and I've never used a spreadsheet solver before.
> > > > > 
> > > > > Is there anyone willing to walk me through it?
> > > > > 
> > > > > This is where I'm at:
> > > > > 
> > > > > 1) I set the Error Sum cell as the "Target Cell" and check on "minimum."
> > > > > 
> > > > > 2) Where it says "By changing cells" I think I need to highlight cell "Elliptical Radius."
> > > > > 
> > > > > When I try this, I get "Err:502" in many of the "Calculated Curve Fit" and "Diff" columns, which I believe means I'm getting multiple possible solutions.
> > > > > 
> > > > > 3) Then there are the "Limiting Conditions." I'm guessing I use these limits to get only one solution?
> > > > > 
> > > > > Thanks in advance for any help!
> > > > >
> > > >
> > > 
> > >
> >
>