Can't Load Old Schematic

Opening my mouth where I might not know what I'm talking about, but here goes anyway! 

If you open that file with a text editor, do you find "SUPPLY1" listed twice in that part?  It may be as easy as using your editor to change that name, or delete the extra definition line.

Make a backup first... 

Doug, I actually tried that right after I posted the question.  In the first attempt it just went to another error involving the same signal, so I deleted that line also.  The schematic now loads and looks correct, but I'm sure there are missing connections.  A better 'hack' at this might be to change the signal name on these 2 lines to some new, unique name.  I haven't tried that yet.

The 2nd error message is: <instance part="SUPPLY1" gate="GND" x="332.74" y="172.72"/>.  Note that this has coordinates in it, which I presume would indicate where the particular connection to this signal occurs, but what coordinate system is it?  I can only set the grid to 'normal' units such as inches or millimeters.

On 1/03/2014 6:48 a.m., Greg Marshall wrote:

Doug, I actually tried that right after I posted the question.  In the

first attempt it just went to another error involving the same signal,

so I deleted that line also.  The schematic now loads and looks correct,

but I'm sure there are missing connections.  A better 'hack' at this

might be to change the signal name on these 2 lines to some new, unique

name.  I haven't tried that yet.

 

The 2nd error message is: <instance part="SUPPLY1" gate="GND" x="332.74"

y="172.72"/>.  Note that this has coordinates in it, which I presume

would indicate where the particular connection to this signal occurs,

but what coordinate system is it?  I can only set the grid to 'normal'

units such as inches or millimeters.

 

The coordinate system for sch file is millimeters

You can verify this for your self. Set the sch grid to MM. next locate a

part on the schematic, say R1 and view its INFORMATION. Record its xy

location

Now look for the part in the file in the 'instances> instance' section.

It will be towards the bottom. It will have the same XY there.

Quickest way to find the right line is search in the file for "x="60.96"

inserting the x you noted from INFORMATION. THere will likely only be

one fine result, the one you want.

Setting the grid to inches, performing a save and then inspecting the

file again, there will be no change to the xy location of the part

proving it's mm.

HTH

Warren

Warren, you're right!  And going to the specified coordinates I found that there was a component missing in the edited file.  So I'm hoping that just replacing/reconnecting this component will fix the problem completely.

I hadn't tried setting the grid to mm because I didn't think it was that big, but it is:  B size is 11x17 inches, so the width in mm is about 430.

Thanks,

Greg

As described earlier, I edited the .sch file to change the 2nd instance of "SUPPLY1" to "SUPPLY31" (a name that was not previously used) for the 2 different uses. The schematic now loads correctly and forward/back annotation to the board layout seems to work.  However, I'm still nervous about the integrity of this fix, so if anyone can better explain the situation I'd really like to hear it.

Thanks,

Greg

As described earlier, I edited the .sch file to change the 2nd instance of "SUPPLY1" to "SUPPLY31" (a name that was not previously used) for the 2 different uses. The schematic now loads correctly and forward/back annotation to the board layout seems to work.  However, I'm still nervous about the integrity of this fix, so if anyone can better explain the situation I'd really like to hear it.

Thanks,

Greg

Did you run an electrical rule check?  If forward/backward annotation is working and you have a good ERC, I think you'll be in pretty decent shape.  Of course, just because the spellchecker passes doesn't mean the grammar is correct, so it would probably be good to go over the schematic visually, and if possible, use LTSpice.  At some point, you might use a gerber viewer to see if things look good too...

There are a number of warnings in ERC, a couple of which were clearly not there before (such as a wire overlapping a pin).  That is, I did ERC on it 2 years ago, accepted a few warnings and fixed everything else.  ERC now shows a few "accepted" warnings, but I'm sure there used to be more, and some of the new ones make no sense.  I can easily fix them, but this is disconcerting.

Is it possible that these inconsistencies are the result of changes to library components?  Only a couple of components have been edited and the errors are not necessarily related to those components.

I loaded up a design I did about a year and a half ago, and I got some inconsistencies myself.  Three of the errors are definitely related to library issues.  Hmmm...

Do you by any chance have a different copy of the files from two years ago?

And I'm probably full of silly questions, but are you using the same computer, same operating system and same version of EAGLE as you did when you created the design?

When I did the original design I archived the .sch and .brd files when I sent it off to have the board made.  Some time later I needed to look at the design, so tried to load it and got this error message.  At that time everything should have been exactly the same as before except that I had added some parts to the library (for another project - not any parts used in this design).  At that time I just used a PDF I had saved of the schematic and didn't worry about it because I didn't think I'd ever be changing the design.  Now I should be in exactly the same configuration.  I have downloaded the latest version of Eagle CAD, but am using the old version for this project just to be sure.  So other than the library additions everything should be the same:  Same files, same software, same computer, same OS.

Of course, the likelihood is that I inadvertently messed up the file without realizing it at some point.

I am proceeding, cautiously, with my changes.  I'll let you know if it blows up!

On 1/03/2014 12:26 p.m., Greg Marshall wrote:

When I did the original design I archived the .sch and .brd files when I

sent it off to have the board made.  Some time later I needed to look at

the design, so tried to load it and got this error message.  At that

time everything should have been exactly the same as before except that

I had added some parts to the library (for another project - not any

parts used in this design).  At that time I just used a PDF I had saved

of the schematic and didn't worry about it because I didn't think I'd

ever be changing the design.  Now I should be in exactly the same

configuration.  I have downloaded the latest version of Eagle CAD, but

am using the old version for this project just to be sure.  So other

than the library additions everything should be the same:  Same files,

same software, same computer, same OS.

 

Of course, the likelihood is that I inadvertently messed up the file

without realizing it at some point.

 

I am proceeding, cautiously, with my changes.  I'll let you know if it

blows up!

 

I suspect Eagle hick-cupped on you in the past.

The XML of the schematic file is quite easy to understand. With an XML

editor like the free FOXE from First Objects you get a good view of the

tree structure and tabs. That helps to see when the xml is malformed.

Right at the beginning you said:

...error message: line 16474, column 66:

redefinition of name 'SUPPLY1' in tag

You should be able now to see where the corruption occurred and change

it properly

All the best

Warren

Warren, I looked at it with FOXE as you suggested, but have to admit that I know nothing about XML.  I didn't see anything suspicious in the structure, it simply had 2 occurrences of 'SUPPLY1' in both the 'PARTS' and 'INSTANCES' sections.  It's not clear (to me) that the 2nd occurrence in one section relates to the 2nd occurrence in the other, but by changing the name of the 2nd occurrence in each section to 'SUPPLY31' the problem seems to be resolved.

On 1/03/2014 3:37 p.m., Greg Marshall wrote:

Warren, I looked at it with FOXE as you suggested, but have to admit

that I know nothing about XML.  I didn't see anything suspicious in the

structure, it simply had 2 occurrences of 'SUPPLY1' in both the 'PARTS'

and 'INSTANCES' sections.  It's not clear (to me) that the 2nd

occurrence in one section relates to the 2nd occurrence in the other,

but by changing the name of the 2nd occurrence in each section to

'SUPPLY31' the problem seems to be resolved.

 

--

To view any images and attachments in this post, visit:

http://www.element14.com/community/message/105131

 

If you are inclined, how about uploading here the section of the XML

from  (inclusive)

Your solution is probably a safe fix but I would be interested in seeing

the original corrupt code.

Warren

Warren,

Below is the <parts> section with most of it snipped out. Note occurrences of "SUPPLY1" near the beginning and end:

<parts>

<part name="D1" library="diode" deviceset="1N4004" device=""/>

<part name="FRAME1" library="frames" deviceset="FRAME_B_L" device=""/>

<part name="C3" library="rcl" deviceset="C-US" device="025-025X050" value="0.1uF"/>

<part name="C4" library="rcl" deviceset="C-US" device="025-025X050" value="0.1uF"/>

<part name="C6" library="rcl" deviceset="C-US" device="025-025X050" value="0.1uF"/>

<part name="C7" library="rcl" deviceset="C-US" device="025-025X050" value="33pF"/>

<part name="C8" library="rcl" deviceset="C-US" device="025-025X050" value="33pF"/>

<part name="SUPPLY1" library="supply2" deviceset="GND" device=""/>

<part name="SUPPLY2" library="supply2" deviceset="GND" device=""/>

<part name="SUPPLY3" library="supply2" deviceset="GND" device=""/>

.

.

.

<part name="U2" library="uln-udn" deviceset="ULN2068B" device="N"/>

<part name="C16" library="rcl" deviceset="C-US" device="025-025X050" value="0.1uF"/>

<part name="SUPPLY1" library="supply2" deviceset="GND" device=""/>

<part name="F1" library="ptc-ntc" deviceset="B57164" device=""/>

<part name="C17" library="rcl" deviceset="CPOL-US" device="E2-5" value="33uF"/>

<part name="J1" library="GBM" deviceset="PWR-JACK-2.5MM" device=""/>

</parts>

And here is a similarly edited part of the <instances> section:

<instances>

<instance part="D1" gate="1" x="60.96" y="251.46"/>

<instance part="FRAME1" gate="G$1" x="0" y="0"/>

<instance part="FRAME1" gate="G$2" x="325.12" y="2.54"/>

<instance part="C3" gate="G$1" x="111.76" y="233.68"/>

<instance part="C4" gate="G$1" x="195.58" y="198.12"/>

<instance part="C6" gate="G$1" x="330.2" y="116.84"/>

<instance part="C7" gate="G$1" x="101.6" y="106.68"/>

<instance part="C8" gate="G$1" x="116.84" y="106.68"/>

<instance part="SUPPLY1" gate="GND" x="53.34" y="215.9"/>

<instance part="SUPPLY2" gate="GND" x="109.22" y="91.44"/>

.

.

.

<instance part="U2" gate="A" x="330.2" y="147.32"/>

<instance part="C16" gate="G$1" x="332.74" y="182.88"/>

<instance part="SUPPLY1" gate="GND" x="332.74" y="172.72"/>

<instance part="F1" gate="1" x="38.1" y="251.46" smashed="yes" rot="R90">

<attribute name="NAME" x="40.132" y="253.7206" size="1.778" layer="95"/>

<attribute name="VALUE" x="43.307" y="253.492" size="1.778" layer="96" rot="R90"/>

</instance>

Does this tell you anything?

Greg

On 2/03/2014 1:05 p.m., Greg Marshall wrote:

Warren,

 

Below is the <parts> section with most of it snipped out. Note

occurrences of "SUPPLY1" near the beginning and end:

 

<parts>

<part name="D1" library="diode" deviceset="1N4004" device=""/>

<part name="FRAME1" library="frames" deviceset="FRAME_B_L" device=""/>

<part name="C3" library="rcl" deviceset="C-US" device="025-025X050"

value="0.1uF"/>

<part name="C4" library="rcl" deviceset="C-US" device="025-025X050"

value="0.1uF"/>

<part name="C6" library="rcl" deviceset="C-US" device="025-025X050"

value="0.1uF"/>

<part name="C7" library="rcl" deviceset="C-US" device="025-025X050"

value="33pF"/>

<part name="C8" library="rcl" deviceset="C-US" device="025-025X050"

value="33pF"/>

<part name="SUPPLY1" library="supply2" deviceset="GND" device=""/>

<part name="SUPPLY2" library="supply2" deviceset="GND" device=""/>

<part name="SUPPLY3" library="supply2" deviceset="GND" device=""/>

.

.

.

<part name="U2" library="uln-udn" deviceset="ULN2068B" device="N"/>

<part name="C16" library="rcl" deviceset="C-US" device="025-025X050"

value="0.1uF"/>

<part name="SUPPLY1" library="supply2" deviceset="GND" device=""/>

<part name="F1" library="ptc-ntc" deviceset="B57164" device=""/>

<part name="C17" library="rcl" deviceset="CPOL-US" device="E2-5"

value="33uF"/>

<part name="J1" library="GBM" deviceset="PWR-JACK-2.5MM" device=""/>

</parts>

 

And here is a similarly edited part of the <instances> section:

 

<instances>

<instance part="D1" gate="1" x="60.96" y="251.46"/>

<instance part="FRAME1" gate="G$1" x="0" y="0"/>

<instance part="FRAME1" gate="G$2" x="325.12" y="2.54"/>

<instance part="C3" gate="G$1" x="111.76" y="233.68"/>

<instance part="C4" gate="G$1" x="195.58" y="198.12"/>

<instance part="C6" gate="G$1" x="330.2" y="116.84"/>

<instance part="C7" gate="G$1" x="101.6" y="106.68"/>

<instance part="C8" gate="G$1" x="116.84" y="106.68"/>

<instance part="SUPPLY1" gate="GND" x="53.34" y="215.9"/>

<instance part="SUPPLY2" gate="GND" x="109.22" y="91.44"/>

.

.

.

<instance part="U2" gate="A" x="330.2" y="147.32"/>

<instance part="C16" gate="G$1" x="332.74" y="182.88"/>

<instance part="SUPPLY1" gate="GND" x="332.74" y="172.72"/>

<instance part="F1" gate="1" x="38.1" y="251.46" smashed="yes"

rot="R90">

<attribute name="NAME" x="40.132" y="253.7206" size="1.778" layer="95"/>

<attribute name="VALUE" x="43.307" y="253.492" size="1.778" layer="96"

rot="R90"/>

</instance>

 

Does this tell you anything?

 

Greg

 

--

Thanks Greg

The XML looks well formed. For some reason Eagle permitted a part of the

same name or made an error when writing out during the save, Anyway your

solution is fine.

Since about v6.5 Jorge reports there have been more checks made when a

board loads so maybe this is why it has shown up now.

All the best

Warren