Practical Electronics for Inventors, 3rd

[TomC]

@TomC this is an AMAZING contribution, thanks!  I just found this thread after posting this errata question on EESE http://electronics.stackexchange.com/questions/123891/bipolar-transistor-switch-base-current-calculation-example-from-pefi-seems-wrong

I don't see this issue in the last errata pdf you published.  Can you have a look at the question and see if there is really an error in the text, or if I'm just misunderstanding something about how BJTs work?

Hi cdwilson,

Thanks a lot for bringing this up. I agree that the equation at the top of page 437 is incorrect. This equation refers to the diagram at the top of Fig. 4.52 and, as you pointed out in your EESE post, the equation should read: I_B = (V_CC - 0.6V)/R₁

I'll add this to the next revision of the errata.

[TomC]

The latest revision of the Unofficial Errata is uploading and should be available in about 30 minutes:

https://onedrive.live.com/redir?resid=967A90CA47FD025B!172&authkey=!ACEbpvA4f9gUlxc&ithint=file%2c.pdf

Thanks to the EEVblog members that reported errors or problems with the text the following pages have been added or modified:

213
214
215
229
437
446

[neslekkim]

That was some comprehensive errata!:), Is this submitted to the authors for corrections?
To bad this book is not as the Oreilly ebooks, that is updated from time to time, so one can download them when they do new reprints of the books.

[TomC]

That was some comprehensive errata!:), Is this submitted to the authors for corrections?
To bad this book is not as the Oreilly ebooks, that is updated from time to time, so one can download them when they do new reprints of the books.

Hi neslekkim,

I've tried to pass on some of these errors to the publisher ever since the early days of the second edition, but got no response. The errata compiled at Bucknell hit the same firewall for years. Perhaps they pay attention only when they are ready to work on a new edition. That's OK, I guess offering interim updates is not part of their business model. However, I'd like to pass on this information to current readers in the hope that it will enhance their experience.

Since I first published the errata there has been quite a bit of input from EEVblog members resulting in additional entries. Hopefully this will lead to an overall far more error free text for all to enjoy.

[cdwilson]

@TomC I sent an email to Simon Monk <evilgeniusauthor@gmail.com> letting him know this thread existed.  See his response below.  I bet if you get in touch with him directly, he may have some swing with the publisher.

Thanks very much!

I have passed this on to my editor. I know someone there was working through another list of errata.

Si.

[neslekkim]

So the Author itself is not interrested in corrections?, hm.. Do the editors have any knowledge about the subject, and can validate what's reported?, hm..

[Rigby]

So the Author itself is not interrested in corrections?, hm.. Do the editors have any knowledge about the subject, and can validate what's reported?, hm..

MUCH more likely that he has someone already working on errata and is not interested in duplicating work.

[TomC]

The latest revision of the Unofficial Errata is uploading and should be available in about 30 minutes:

https://onedrive.live.com/redir?resid=967A90CA47FD025B!172&authkey=!ACEbpvA4f9gUlxc&ithint=file%2c.pdf

Thanks to the EEVblog members that reported errors or problems with the text the following pages have been added or modified:

225
228
229
230

[cdwilson]

On Page 23, Figure 2.18 shows a graph of the "Response of Ohmic Materials" with Voltage on the Y axis, and Current on the X axis.  Should these axis labels be swapped?  For example, in a conductor, a small voltage should product a large current.  However, the graph seems to indicate that a small current would product a large voltage in conductors, which doesn't seem correct.

[TomC]

On Page 23, Figure 2.18 shows a graph of the "Response of Ohmic Materials" with Voltage on the Y axis, and Current on the X axis.  Should these axis labels be swapped?  For example, in a conductor, a small voltage should product a large current.  However, the graph seems to indicate that a small current would product a large voltage in conductors, which doesn't seem correct.

Hi cdwilson,

Thanks for pointing this out!

As I see it, the labels "conductors" and "insulators" need to be swapped. That way a small voltage will produce a large current on conductors but a very high voltage would be required to produce a very small current on insulators. This will have the same general effect as swapping the labels on the axes, but given the current angle of the graph lines I think it will be more representative of the actual ohmic range of conductors and insulators. I will add this correction to the next revision of the errata.

Again, thanks a lot for the input!

[TomC]

The latest revision of the Unofficial Errata is now available:

https://onedrive.live.com/redir?resid=967A90CA47FD025B!172&authkey=!ACEbpvA4f9gUlxc&ithint=file%2c.pdf

Thanks to the EEVblog members that reported errors or problems with the text the following pages have been added or modified:

23

[cdwilson]

Couple questions on page 35:

1) The errata says equation 2.11 should use the integral form of Fourier's law.  Can you provide more insight about why the differential form is incorrect?  Is it because the differential form (-k*deltaT) is equal to the heat flux density, and when you integrate the heat flux density over the total surface, you get the total total heat transferred per unit time (i.e. the total power)?  Also, in this Wikipedia article http://en.wikipedia.org/wiki/Thermal_conduction#Integral_form the integral form ends in dA while the errata ends in dS.  Which is correct (and why)?

2) equation 2.12 is defined as P_heat = -k*A*deltaT/L, where deltaT = T_hot - T_cold.  Should deltaT be defined instead as T_cold - T_hot?  Since T_cold - T_hot will always be a negative quantity, P_heat will be a positive quantity.

3) Why were the brackets on the gradient equation changed to square brackets?  Do these denote a vector or something?

[TomC]

Couple questions on page 35:

1) The errata says equation 2.11 should use the integral form of Fourier's law.  Can you provide more insight about why the differential form is incorrect?  Is it because the differential form (-k*deltaT) is equal to the heat flux density, and when you integrate the heat flux density over the total surface, you get the total total heat transferred per unit time (i.e. the total power)?  Also, in this Wikipedia article http://en.wikipedia.org/wiki/Thermal_conduction#Integral_form the integral form ends in dA while the errata ends in dS.  Which is correct (and why)?

2) equation 2.12 is defined as P_heat = -k*A*deltaT/L, where deltaT = T_hot - T_cold.  Should deltaT be defined instead as T_cold - T_hot?  Since T_cold - T_hot will always be a negative quantity, P_heat will be a positive quantity.

3) Why were the brackets on the gradient equation changed to square brackets?  Do these denote a vector or something?

Hi cdwilson,

Thanks for posting your concerns in this forum!

1) I think you got the idea. It's not that the differential form is incorrect, but in the context of what the author says it seems to me that he meant the integral form. The differential form refers to the "heat flux density" or heat rate per unit area which is a vectorial quantity, the integral form refers to "heat rate" through a body as a whole which is a scalar quantity. In the caption for Fig 2.24 it seems to me that the author defines Pheat as "the rate of heat transfer, or power due to heating", also at the top of page 37 he defines Pheat as thermal current (heat flow). These statements again lead me to believe that on equation 2.11 he was talking about the integral form. There is a Wikepedia article on "heat current" where the integral form is used to define it:

http://en.wikipedia.org/wiki/Heat_current

The Wikipedia article you mention is where I got the formula for the errata, but I did the research for this errata entry back in 2009 or 2010 and since then the article has undergone many edits. Either version is OK in my opinion since dS and dA are defined as the same quantity. The link for the article just before they edited the formula is below:

http://en.wikipedia.org/w/index.php?title=Thermal_conduction&oldid=337138011

I surmise that the reason for the change is to avoid confusion between the S used as a limit and the S in dS.

2) As I understand it this is by design and the negative sign represents the direction of the flow from hot to cold. However I've seen one article where the author omitted the minus sign on a similar formula.

3) There is no special meaning for the brackets, I use the Open Office formula editor and as I remember the scaling of parentheses didn't work well at the time I did this.

@ A reminder to other members:

Please jump in at any time and offer your opinion!
Specially if you have expertise on one of these complex subjects that are glossed over in the book!

[cdwilson]

Hi Tom,

Thanks for the detailed reply.  Agreed, if anybody else wants to jump in and provide more insight, please feel free!

1) I think I understand the difference between heat flow (dQ/dt) and the "heat flux density" as defined on this page http://en.wikipedia.org/wiki/Thermal_conduction#Fourier.27s_law  However, I'm having a hard time figuring out if this "heat flux density" is synonymous with "heat flux", or if they are actually two different things (for example, in E&M they are different, flux density is flux/unit area).  https://en.wikipedia.org/wiki/Heat_flux doesn't provide any clarification, and seems to use them synonymously.  From the Wikipedia definition of heat flux, it appears that it already is defined per unit area, which makes me totally confused as to what "heat flux density" is (per unit area, per unit area???)

2) I think the negative sign is required in the differential form (-k*A*gradient_T) because the negative sign counteracts the "downhill" gradient, providing a positive heat flow along the gradient from hot to cold.  However, when you integrate this equation to obtain the heat flow through a uniform object, you end up with delta_Q/delta_t=k*A*(T1-T2)/l where T1=hot and T2=cold.  See the example given on this page http://www.brighthubengineering.com/hvac/54035-the-one-dimensional-heat-conduction-equation/ Also check out Physics for Scientists & Engineers, 3rd Ed, Giancoli, Page 503.  I'm trying to find a clear explanation of this online so I can verify, if you find one, please post it. 

[TomC]

Hi cdwilson,

You pose some interesting questions and arguments for which I have no ready answer. I'm not an expert on this subject but I'll do some more research and give you my opinion. Right now I have to mow the lawn so I'll take a better look at this tonight!

[cdwilson]

Hi Tom,

RE: #1 below, https://en.wikipedia.org/wiki/Flux seems to indicate that there are two different usages for "flux":

1) Flux as a surface integral (including E&M).  Defined as the surface integral of the vector field.  In electrical fields for example, the E-field is the electric flux density, and the surface integral is the total electric flux out of the surface.

2) Transport phenomena (including heat flux).  Defined as flow rate of property per unit area.  Since the definition on Wikipedia already includes per unit area, it would make more sense (to me) if this quantity was called "heat flux density" instead of "heat flux".  I'm totally confused why in one context (#2) "flux" would be per unit area, and in another context (#1) "flux" would be the total over the entire surface...

I wasn't familiar with Fourier's law before coming across it in the text.  I could be totally wrong about all of this 

Btw, I think part of the reason why I was initially confused was because I didn't realize that the differential form and the integral form were actually different quantities (i.e. they weren't just different forms for heat flow).

If we figure out the remaining details about heat flux vs. heat flux density, I think the errata could be improved if both the differential and integral forms are shown in the explanation, with proper labels (heat flow vs. heat flux [density???]).

[TomC]

Hi cdwilson,

Intuitively I believed that "heat flux" and "heat flux density" were the same until you questioned it and aired some valid concerns. However, after researching this further, I'm again convinced that they are the same. I think  the word "density" added to "heat flux" just sort of reaffirms the formal meaning of "heat flux". Go figure that this reaffirmation can be confusing if you know the formal meaning of "heat flux" beforehand. I found quite a few corroborating articles on the web by searching for: define "heat flux density". Here is a link to one that spells it out unequivocally:

http://www.norenproducts.com/thermal-solution-components

I also did some research on the comments that you had about equation 2.12 and I know believe, as you indicated, that something needs to be changed so that you don't end up with negative results all the time. The easiest to implement would be to just delete the minus sign on the RHS of the equation. Another possibility would be to delete delta_T and replace it with Tcold - Thot. This would also require changing the paragraph below the equation to something like:

Figure 2.24 shows a picture of the situation. Here delta_T = Thot - Tcold, measured at points across the length L of the material. The material may be steel, silicon, copper, PCboard material, and so on.

A number of web articles influenced my thinking on this but below is a link to a PDF that helped a lot:

http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=7&ved=0CEwQFjAG&url=http%3A%2F%2Fwww.me.umn.edu%2Fcourses%2Fold_me_course_pages%2Fme3333%2Fessays%2Fessay%25203.pdf&ei=ClYWVMTzAeTbsAT7zYKIAg&usg=AFQjCNFd-VZKZj6KbYW7I9uINUH9-dAgNw&bvm=bv.75097201,d.cWc&cad=rja

The reason I'm not proposing to change the definition of delta_T to Tcold - Thot as you initially suggested is because of conflict with the thermal resistivity equations that follow. Otherwise this would be more mathematically precise as indicated in the text of the above PDF.

[cdwilson]

Thanks Tom.  I agree on the "density" suffix, thanks for looking into this and clearing it up for me. 

The easiest to implement would be to just delete the minus sign on the RHS of the equation.

I think this is the correct change to make.  It is consistent with the equation/explanation in the physics textbook I have which also removes the minus sign and uses delta_T = T_hot - T_cold. 

Seriously man, many thanks for your responses and hard work on the errata doc.  It has saved me from many head scratching moments while starting to read through this text 

[robrenz]

Seriously man, many thanks for your responses and hard work on the errata doc.  It has saved me from many head scratching moments while starting to read through this text 

I still scratch my head even when the formulas are corrected   

[TomC]

The latest revision of the Unofficial Errata is uploading and should be available in about 30 minutes:

https://onedrive.live.com/redir?resid=967A90CA47FD025B!172&authkey=!ACEbpvA4f9gUlxc&ithint=file%2c.pdf

Thanks to the EEVblog members that reported errors or problems with the text the following pages have been added or modified:

35

[cdwilson]

Page 381, example 7, part c, shouldn't the answer be 7.5 ohms instead of 8?  (did they just round up?)

[TomC]

Page 381, example 7, part c, shouldn't the answer be 7.5 ohms instead of 8?  (did they just round up?)

Hi cdwilson,

I think you are right and the author just rounded the answer. It seems that in this section all the answers involving numbers greater than 1 are rounded to a whole number.

[nowlan]

Havent looked at the question myself, but you would normally have to round to nearest E12 value in real life.

1.0, 1.2, 1.5, 1.8, 2.2, 2.7, 3.3, 3.9, 4.7, 5.6, 6.8, 8.2 and 10.0
times a power of ten.

8.2 - 7.5 = 0.7
7.5 - 6.8 = 0.7

[TomC]

Hi nowlan,

You are right that when selecting resistors you normally round to one of the E values depending on the precision you need. However, in this case we are looking at the calculation for the input impedance of a transformer. The exact value is 7.5 ohms, but the author rounded to a whole number (8 ohms).

[cdwilson]

Page 52, Figure 2.40, the middle plot shows a non-linear curve for Current vs. Resistance (I = V/R).  Shouldn't this be linear for resistors?