Bob Cordell Interview: Negative Feedback

[Johan Potgieter]

BWRX,

May I just slip in a final one, please? Use gold, support South-Africa! (We need it.)

But yes - your ruling saves me from .... never mind.

 

[PB2]

Hi Pete,

(Your ASIC design stuff sounds coool!)

It depends. If it's just a quickie design, just "playing around", with no pcb layout yet, then I have to somehow just estimate what typical PCB trace widths and lengths might be, and use a corresponding impedance for them (often just the LR components). For i/o and power supply connections, I usually assume wires or cables for interconnections and use a small set of spice .PARAM statements for each one, so I can just set the lengths in inches and the impedances are automatically calculated and their values inserted.

Even though the trace and wire characteristics might be "just estimates", note that they can then be used to help determine, for example, which ones' lengths or thicknesses might be critical.

For inductors, I always include at LEAST the series resistance. For through-hole resistors, I usually just put something like 0.5pF in parallel with each one. For capacitors, it's often very important to include at least the ESR (equiv series R).

For film and ceramic caps, I try to get the datasheets, and look at the manufacturers' websites. If they only give the %dissipation factor, which is tan(delta) x 100 (where tan(delta) is the tangent of the loss angle), then I use tan(delta) = 2 x Pi x f x C x ESR and solve for ESR. If a max value for DF is all they give, I have to use that. Also, Kemet's website has software that will give good capacitor models, but not for electrolytics.

For electrolytic capacitors, they often only give ESR at one f, such as 100 kHz. But it varies significantly with f. Usually I can get lucky and find a figure for DF at some other f, on the manuf website, from which I can calculate the ESR for a second f. If so, for transient simulations at least, I can assume a linear ESR vs f (the best I can do without more data) and then use a spice .PARAM statement for each type and value of capacitor, to calculate the ESR for the freq param being used in a transient run. For leakage current, they usually say something like "I=.01CV or 3uA, whichever is greater". So I usually just use a spice .param statement to calculate Rleak=1/(.01C), for each capacitor type and value. I posted an example of a similar procedure for modeling an electrolytic's ESR and leakage, here: http://www.diyaudio.com/forums/showthread.php?s=&postid=1228677&highlight=#post1228677 .

For "AC Analysis" spice runs, the param statements to calculate ESR vs f cannot be used, unfortunately. But you can put a G-Laplace source is series with a capacitor, to get the ESR to vary with f, using something like the linear model for ESR vs f mentioned above. You can look in the message archive of the LT-SPICE group at http://www.yahoogroups.com , for a more-detailed answer about that, which was posted on June 22, 2007. Also, look in the group's Files area, mainly in the Tut section, to get many examples for modeling R, L, and C, with parasitics and with temperature dependence, monte carlo runs, sensitivity analysis, etc., etc., etc.

Also, I have posted a two-winding transformer model that calulates its L and R parameters based on simple RMS voltage measurements that the user can enter, which is at http://www.fullnet.com/~tomg/gooteesp.htm . But, it doesn't model the interwinding capacitance or the two intra-winding capacitances. However, if you can short both the primary and secondary and measure, somehow (another whole can of worms), the C between them (interwinding), and then short each one in turn and measure the other's C, which is actually measuring each intra-winding C in parallel with the interwinding C, then you can calculate each intra-winding C.

- Tom Gootee

http://www.fullnet.com/~tomg/index.html

Hi Tom,

Interesting thoughts there thanks!

I believe that, as I'm sure you know, there is a balance between detail in the model and the required accuracy. Certainly, RF and high speed designs in the 100 MHz to 1 GHz range require fine detail, however I prefer to keep it simple when possible. Interesting issue in audio, that we need 20 kHz bandwidth, or let's say 100 kHz for some margin, yet the transistors often have ft's of 100 MHz or higher. So we probably do need a bit more accuracy in order to see coupling, and oscillation issues, perhaps even ground lift/coupling issues but it makes sense to me to keep the parasitic details to a minimum if possible.

Here's an interesting article from 1986 in the Intusoft newsletter:
http://www.intusoft.com/nlpdf/nl06.pdf

And here are a few on capacitors:
http://www.intusoft.com/nlhtm/nl65.htm

http://www.intusoft.com/nlhtm/nl44.htm#modeling

http://www.kemet.com/kemet/web/home...CD004EBC04/$file/TechTopics Vol4No5 Sep94.PDF

Here's a discussion of a power amplifier thermal model:
http://www.intusoft.com/nlpdf/nl11.pdf

Here's the index to the Intusoft Newletters:
http://www.intusoft.com/newsletters.htm

You do a lot of nice work Tom, thanks for sharing it here and on the web.

Pete B.

PS. Seems this didn't get moved to the SPICE thread so I replied here.

 

[PB2]

Re: Re: Re: Re: Distortion vs bias current

That's why I have chosen the 21193/4 model, which seems to me to be more realistic.

From OnSemi:

.MODEL mj21194 npn
+IS=4.02325e-14 BF=10000 NF=1.1488 VAF=10000
+IKF=0.377331 ISE=2.16244e-09 NE=2.49213 BR=0.1
+NR=1.5 VAR=1.70851 IKR=3.77331 ISC=1.00031e-16
+NC=3.99945 RB=0.1 IRB=0.1 RBM=0.1
+RE=0.00775691 RC=0.0387846 XTB=0.1 XTI=1
+EG=1.05 CJE=1.07724e-08 VJE=0.975489 MJE=0.524369
+TF=1e-08 XTF=2.16157e+06 VTF=0.184568 ITF=5.56361
+CJC=2.68609e-10 VJC=1.64862 MJC=0.242322 XCJC=0.1
+FC=0.910137 CJS=0 VJS=0.75 MJS=0.5
+TR=1e-07 PTF=0 KF=0 AF=1


BF=10000???? Does this figure pass a sanity check?

 

[PMA]

Cool down, mine has BF=69.3372 for 21193 and BF=41.975 for 21194.

 

[Edmond Stuart]

[snip]
Why burden you with high end requirements?

Indeed, you're perfectly right. Why 'burden' myself with high-end requirements and that kind of BS, while my knowledge and experience is rather limited. As a matter of fact, I'm only involved in extreme high-end.

Cheers,

 

[PB2]

Cool down, mine has BF=69.3372 for 21193 and BF=41.975 for 21194.

The data I showed was fresh from OnSemi at the time that I posted. I'm a bit annoyed that there seems to be no quality control on models from OnSemi. Seems they should be required to fix this being an ISO9000 company.

The MJ and MJL versions of the parts have different models from what I've seen. Tried to find a good pair for the 21195/96 and most of them looked suspicious in one way or another. The first pair I tried in simulation did not pass a sanity check, which is what sent me off checking models.

Onsemi needs to but some quality checks into their models.

Where did you get your models, PMA?

Pete B.

 

[Bob Cordell]

The data I showed was fresh from OnSemi at the time that I posted. I'm a bit annoyed that there seems to be no quality control on models from OnSemi. Seems they should be required to fix this being an ISO9000 company.

The MJ and MJL versions of the parts have different models from what I've seen. Tried to find a good pair for the 21195/96 and most of them looked suspicious in one way or another. The first pair I tried in simulation did not pass a sanity check, which is what sent me off checking models.

Onsemi needs to but some quality checks into their models.

Where did you get your models, PMA?

Pete B.

This is why I chose to use Andy_c's RET models for my crossover distortion simulations. Although it would be nice to have the exact right value for RE of the BJT, it will largely just be absorbed into the external RE in the Class-AB stage. It would be relatively more important for simulations using RE = 0.1 as opposed to the simulations using RE = 0.22 that I did.

Cheers,
Bob

 

[PMA]

RE = 0.1 ohm is what I use in my designs, I prefer it to higher values. I would admit maximal value of 0.15 ohm.

 

[PMA]

Where did you get your models, PMA?

Pete B.

They are from MC8 library.
AFAIK we should solve this in the SPICE thread.

 

[gootee]

Hi Tom,

Interesting thoughts there thanks!

I believe that, as I'm sure you know, there is a balance between detail in the model and the required accuracy. Certainly, RF and high speed designs in the 100 MHz to 1 GHz range require fine detail, however I prefer to keep it simple when possible. Interesting issue in audio, that we need 20 kHz bandwidth, or let's say 100 kHz for some margin, yet the transistors often have ft's of 100 MHz or higher. So we probably do need a bit more accuracy in order to see coupling, and oscillation issues, perhaps even ground lift/coupling issues but it makes sense to me to keep the parasitic details to a minimum if possible.

Here's an interesting article from 1986 in the Intusoft newsletter:
http://www.intusoft.com/nlpdf/nl06.pdf

And here are a few on capacitors:
http://www.intusoft.com/nlhtm/nl65.htm

http://www.intusoft.com/nlhtm/nl44.htm#modeling

http://www.kemet.com/kemet/web/home...CD004EBC04/$file/TechTopics Vol4No5 Sep94.PDF

Here's a discussion of a power amplifier thermal model:
http://www.intusoft.com/nlpdf/nl11.pdf

Here's the index to the Intusoft Newletters:
http://www.intusoft.com/newsletters.htm

You do a lot of nice work Tom, thanks for sharing it here and on the web.

Pete B.

PS. Seems this didn't get moved to the SPICE thread so I replied here.

Hi Pete!

THANKS for those extremely interesting and useful links about spice modeling! Those ought to keep me busy for a while.

And I certainly agree that, as you mentioned, the modeling should be kept as simple as possible, for the task at hand. (Paraphrasing Einstein: Make it as simple as possible, but no simpler.) But, of course, there might be times when I like to inject RF into audio circuits, to see what happens.

Generally, though, if it's not clear to me whether or not a certain type of parasitic might be significant, it can be a bit of a pain, since I might want to include it everywhere and then, if there is a significant overall difference, might have to use a "process of elimination" in order to find out which instances of the parasitic were significant, to get back to a "minimal" model. That process can provide useful information. But I guess/hope that the more experience I get, the less I'll need to do that (the "hard way", at least).

- Tom Gootee

http://www.fullnet.com/~tomg/index.html

 

[PB2]

East Coast Burning Amp

I just dropped my daughter off at Skidmore, maybe we should have an East Coast Burning Amp.

No interest?

 

[PB2]

Hi Pete!

THANKS for those extremely interesting and useful links about spice modeling! Those ought to keep me busy for a while.

And I certainly agree that, as you mentioned, the modeling should be kept as simple as possible, for the task at hand. (Paraphrasing Einstein: Make it as simple as possible, but no simpler.) But, of course, there might be times when I like to inject RF into audio circuits, to see what happens.

Generally, though, if it's not clear to me whether or not a certain type of parasitic might be significant, it can be a bit of a pain, since I might want to include it everywhere and then, if there is a significant overall difference, might have to use a "process of elimination" in order to find out which instances of the parasitic were significant, to get back to a "minimal" model. That process can provide useful information. But I guess/hope that the more experience I get, the less I'll need to do that (the "hard way", at least).

- Tom Gootee

http://www.fullnet.com/~tomg/index.html

Yes, I certainly agree, and it would be nice to come up with some guidelines or rules of thumb concerning when parasitics are required. I agree that it comes with experience.

Pete B.

 

[Edmond Stuart]

They are from MC8 library.
....

Where did you get the library, PMA?

 

[PB2]

This is why I chose to use Andy_c's RET models for my crossover distortion simulations. Although it would be nice to have the exact right value for RE of the BJT, it will largely just be absorbed into the external RE in the Class-AB stage. It would be relatively more important for simulations using RE = 0.1 as opposed to the simulations using RE = 0.22 that I did.

Cheers,
Bob

I've been using Andy's models for some time now and I agree that he did an excellent job. I agree, RE= 0 is not a serious issue, especially with an external RE as you mention.

Pete B.

 

[Bob Cordell]

Re: East Coast Burning Amp

No interest?

In girls at Skids? Of course there is. Many of us at RPI dated them.

OOPS!! , you meant amplifiers!!

I could definitely be interested.

Cheers,
Bob

 

[PMA]

I think we have forgotten one important example when exploring NFB here. Let's have a nonlinear element with pure 2nd harmonic nonlinearity. Let's close it into NFB loop and review the resulting spectrum. Maybe jcx has already done it before?

 

[jcx]

you mean this?

http://www.diyaudio.com/forums/showthread.php?s=&postid=920583&highlight=#post920583

or from this thread, a little before the "notching out" 7th stuff:

http://www.diyaudio.com/forums/showthread.php?s=&postid=1277808&highlight=#post1277808

the Cherry article ref in above post gives feedback around 2nd or 3rd nonlinearities as series in feedback factor

 

[PMA]

Yes, the 1st link. Thanks!!

 

[PMA]

That is what some of us spoke about. Feedback creates new harmonics, not present in original circuit output signal before feedback application.

 

[ingrast]

That is what some of us spoke about. Feedback creates new harmonics, not present in original circuit output signal before feedback application.

And in actual music, it appears as a more or less correlated "grass" that does no show up in full under measurement.

The "veil" some listeners complain about when talking about NFB liabilities, though I have not experienced that myself in an obvious way.

Rodolfo