Laser, DFB module: TEC reaction speed

[Joerg]

My e-mail address has been showing up here since 1997 (when spam wasn't
a problem).
Modern spam filters keep the level of spam getting into my in-box well
below the regular traffic, so I'm not going to start munging now.

True. I just turned on Mozilla's bayesian spam filter after it got to
over 100 spams a day. Works great. The downside is that it does
occasionally catch an important email so I don't want my ISP to do the
same and toss them. This can be a drag when you are on the road and
without an Internet connection via relatives or via a client. 100
spams/day gets really old on dial-up :-(

Once we turned off the self-tuing algorithm and tuned our PID
controller using the Ziegler-Nichols procedure, we got remarkably fast
settling.

http://ourworld.compuserve.com/homepages/ACTGMBH/zn.htm

I haven't measured the plant behavior yet because it got late and their
engineer had to catch his flight. Will do this week. But I did attempt
my usual tuning procedure which is similar to Z-N. Zeroed I and D, then
eased P up to where it just began oscillating, then added a spritz of D
and so on. It still felt like wading through molasses.

Measuring waste heat fast enough ain't easy either because these are DFB
modules in a can.

Melbourne has just had its coldest Christmas on record, with a maximum
temperature of 14.5C - about 10C cooler than usual. Various mountains
around Victoria and Tasmania had snow cover during the day. My mother's
gas-fired central heating turned itself on, and I had to find myself a
sweater ...

Not much different from California then

They announced a big fat storm rolling in but so far it's just windy.
IIRC folks in the Bay Area have to watch for flash floods. John: Might
want to get the waders ready and stock up on Merlot, Anchor Steam, filet
mignon and other essentials.

 

[Mike Monett]

True. I just turned on Mozilla's bayesian spam filter after it got
to over 100 spams a day. Works great. The downside is that it does
occasionally catch an important email so I don't want my ISP to do
the same and toss them. This can be a drag when you are on the
road and without an Internet connection via relatives or via a
client. 100 spams/day gets really old on dial-up :-(>

There's no need to waste time on spam filters that leave you
wondering if you somehow erased an important email. The best method
is to avoid getting spam in the first place.

Spammers can't send you spam if they don't know your address. I used
to get tons of spam, now I get only one or two spam per year. And I
don't need to waste time on spam filters. Here's how it works.

Get your isp to give you a second email address. Use it to create a
bunch of disposable email addresses. You can use many free services
such as Spammotel, Kasmail, Spamgourmet, etc. Watch out for ones
that silently stop sending email when you reach a quota. The best
one I've found is

http://www.kasmail.com/

Give a unique address to each of your friends and customers. Keep
track of the addresses you assign. When you have finished, kill the
old address that is getting spam.

If you start getting spam, let that person know they may have been
infected with a virus, or may have downloaded trojan software that
harvests addresses. Kill that email address and give them a new one
if you think they will do better in the future.

Use the same procedure with site registration and forums. If you
start getting spam from them, let them know they have a problem.

Get an email client that allows unlimited email addresses so you can
reply to an email and use the correct address. Geminisoft Pimmy is
an excellent client. It doesn't handle html or attachments, so you
can never get infected with the viruses that plague Microsoft.

http://www.geminisoft.com/en/

If you need to put your email address on a web site, use an address
encoder such as http://www.addressmunger.com/

I have two web sites that have been up for over five years and have
never received spam from either one. Be sure to let everyone know
the email address may change, so they should save the url. Here's an
example:

http://silversol.freewebpage.org/contact.htm

That's all there is to it. Instantly, your spam will drop to zero,
and you don't need to worry about erasing false positives.

If you are on the road a lot, most isp's allow you to download email
from their web site. To reply, make sure you use the correct address.

Regards,

Mike Monett

Antiviral, Antibacterial Silver Solution:
http://silversol.freewebpage.org/index.htm
SPICE Analysis of Crystal Oscillators:
http://silversol.freewebpage.org/spice/xtal/clapp.htm
Noise-Rejecting Wideband Sampler:
http://www3.sympatico.ca/add.automation/sampler/intro.htm

 

[[email protected]]

Translation please, for us ohioians.

Google on "cricket game description"

http://www.cricinfo.com/link_to_database/ABOUT_CRICKET/EXPLANATION/

and note that the British national cricket team are now touring
Australia, playing a series of five "Test Matces" against the
Australian national cricket team. Australia have won the first three
matches (and the series) and are now winning the fourth test n
Melbourne.

McGrath and Warne are Austraian bowlers of extraordinary effetiveness.

 

[Lostgallifreyan]

[email protected] wrote in

and are now winning the fourth test

Steady.. > They could, but things aren't so certain now at 84 for 5.
That's a worse position that England were in yesterday.

 

[Terry Given]

ADI says: "The ADN8831 is a highly integrated solution for controlling
the TEC in a laser module used in optical communications systems."

So why should it work for you?

My laser expertise doesn't extend much beyond knowing to put on those
funny goggles before I defeat the interlock. But as a control guy, I
can tell you that just tuning a PID loop isn't going to get you the best
possible performance.

There are two reasons for this. First, any controller that features an
integrator is almost guaranteed to give you overshoot without the
careful application of clever feed-forward schemes or nonlinearities in
the controller. Second, while you can, in theory, achieve near-perfect
control of a second-order system using a PID with clever feed-forward,
thermal systems are, in general, continuous-state systems. So you may
not be able _at all_ to use a PID to do this, even _with_ clever
nonlinearities as well as clever feed-forward.

If this task can be accomplished with clever application of the PID
controller, the first thing you're going to have to do is to pitch that
chip. I'd suggest replacing it with the microcontroller of your choice,
so you don't have to stand on your head to implement the necessary
control rule in hardware and so you're not constrained by the hardware
when you need to change it -- but once you have a control rule in hand
you can certainly implement it with analog circuitry, should that be
your preference.

At any rate, you need a PID controller that looks just like any other
PID controller _except_ for two things. First, set the feed forward and
feedback gains separately for the proportional and differential stages,
such that

prop_term = k_pf * command - k_pb * output
diff_term = d/dt(k_df * command - k_db * output)

this is a great trick. I picked it up from an IEEE paper a few years
ago, tried it and have been hooked ever since. I then found it in one of
Astrom & Wittenmarks books. Which I got for US$1, brand new, because the
hard-cover had been bound upside down. hurrah for manufacturing screw-ups.

Second, instead of using the traditional integrator anti-windup that
just limits the state of the integrator to fixed values, use
output-based integrator limiting. This means you should choose an
output value to limit to, then adjust the integrator state to hold that
value. This is fully discussed in my book, and I'm feeling lazy today,
so I'll just refer you to pp 213 and 214, equation 8.40, and the
surrounding text (Tim Wescott, "Applied Control Theory for Embedded
Systems", Elsevier 2006). Using this sort of integrator limiting will
dramatically reduce, or completely eliminate, the overshoot from
integrator windup, at the cost of having a much slower 'tail' to the
response -- and there are things you can do about that tail.

There's a good chance that using these two techniques together will get
you close enough. Certainly they'll get you much closer than you are
now. If they don't then reflect on this:

In principal you want a controller that puts just the right input into
the plant so that the plant stops where you want and then doesn't move.
For a linear system the output of the controller will be a function of
your desired behavior* filtered by the inverse transfer function of the
plant. You can, in theory at least, generate this sort of result with a
linear discrete-state plant of arbitrary order by implementing a
controller of similar order to push it around. In practice you run into
trouble if you get too ambitious, because your plant model is rarely
kind enough to match what you think it is.

I have had some success in the past using systems that are largely feed
forward with just enough feedback to fix things up at the end of the
excursion. My results with this have been mixed, largely due to folks
not allowing enough time to fix all of the interesting 2nd- and 3rd-
order effects that pop up as soon as you address the 'main' problem. If
you really need this to work, however, and you can spare the development
time, it's a way to go.

* if you have realistic desires, which means not asking for anything
that the plant's going to filter out to any degree.

Hi Tim & Joerg,

Tim, have you done much work with so-called Internal Model Control? Its
pretty much what you just described. of course the hard part is figuring
out the feedforward path. There is a nice description of it in the CRC
Press Control Handbook (ed. Levine). I implemented a really nice current
controller this way a few years back, and had fabulous results (the
control was almost good enough, even with a huge error in my sine(theta)
calculation - flipped the sign for 20 degrees or so) at very low sample
rates. Almost as good a step-response as dead-beat control (OK, 3 not 2
periods), but very robust to parameter disturbance.

Joerg, thats not much use to you though

I hope you guys (and all SED participants) have a safe and joyous
festive season. Would you believe it snowed here and in Aussie the day
before xmas ?!

Regards,

Terry

 

[Terry Given]

Come to think of it, Linear Technology's data sheet for their LTC1923
chip tells you that closing the thermal feedback loop s tricky, and
their application note AN-89 goes into more detail.

Jim Williams hasn't include the quadratic formula relating heat
transferred through the Peltier junction to current driven through the
junction and the temperature difference across the junction, all of
which interact, such that the current-to-heat transferred "gain of the
junction is rather variable - enogh to make a system which is
critically damped at one temperature difference either over-damped or
underdamped at another.

My two papers do include the formula, and describe systems that can
remain critically damped over a range of temperatures.

Hi Bill,

can you email me copies of those papers please? [email protected]

Regards,

Terry_Given

 

[Terry Given]

[email protected] wrote in




Steady.. > They could, but things aren't so certain now at 84 for 5.
That's a worse position that England were in yesterday.

bah humbug! We (NZ) suck so bad at cricket we have to resort to dirty
tricks like the stumnping of (IIRC) Muralitharan when we went to
congratulate one of his team members.

I saw on the news last night, our captain (and one of the better
batsmen) out for 17. Yet we pay these sods around $80k/year each.

Of course our soccer team is far, far worse than the black-craps.

Still, there's always rugby

Cheers
Terry

 

[Tim Wescott]

this is a great trick. I picked it up from an IEEE paper a few years
ago, tried it and have been hooked ever since. I then found it in one of
Astrom & Wittenmarks books. Which I got for US$1, brand new, because the
hard-cover had been bound upside down. hurrah for manufacturing screw-ups.

-- snip again --

Hi Tim & Joerg,

Tim, have you done much work with so-called Internal Model Control? Its
pretty much what you just described. of course the hard part is figuring
out the feedforward path. There is a nice description of it in the CRC
Press Control Handbook (ed. Levine). I implemented a really nice current
controller this way a few years back, and had fabulous results (the
control was almost good enough, even with a huge error in my sine(theta)
calculation - flipped the sign for 20 degrees or so) at very low sample
rates. Almost as good a step-response as dead-beat control (OK, 3 not 2
periods), but very robust to parameter disturbance.

I have a tendency to reinvent methods to solve specific problems, then
find out later that I've done something that was invented in the 50's
(sometimes it's the 1950s, sometimes it's the 1850's, sometimes...).

I've done this more where you have a nice, repetitive input, which makes
adjustments to the feed forward more robust in the face of plant
nonlinearities. Making this work well for an arbitrary input would be
more difficult, but should be possible using adaptive techniques.

I'm going to have to check out Internal Model Control, though.

Joerg, thats not much use to you though

I hope you guys (and all SED participants) have a safe and joyous
festive season. Would you believe it snowed here and in Aussie the day
before xmas ?!

Would you like some nice Pacific Northwest (US) rain? We've got lots of
it this week.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Posting from Google? See http://cfaj.freeshell.org/google/

"Applied Control Theory for Embedded Systems" came out in April.
See details at http://www.wescottdesign.com/actfes/actfes.html

 

[Joerg]

-- snip --

-- snip again --



I have a tendency to reinvent methods to solve specific problems, then
find out later that I've done something that was invented in the 50's
(sometimes it's the 1950s, sometimes it's the 1850's, sometimes...).

I've done this more where you have a nice, repetitive input, which makes
adjustments to the feed forward more robust in the face of plant
nonlinearities. Making this work well for an arbitrary input would be
more difficult, but should be possible using adaptive techniques.

I'm going to have to check out Internal Model Control, though.

I am afraid it's going to be tougher than anticipated because we don't
know what the plant looks like. Lots of individual delay elements. Then
there is the TEC itself which isn't linear at all over +/-10C. Of
course, none of that is spec'd anywhere :-(

In the end it might indeed boil down to a uC with lots of PID and feed
forward parameters in a lookup table that gets switched through as the
DFB module TEC passes through temperature ranges.

Would you like some nice Pacific Northwest (US) rain? We've got lots of
it this week.

If it's any comfort, this stuff has arrived in the Pacific Southwest
about two hours ago. With the usual flash flood warnings and all that.

 

[Terry Given]

-- snip --

-- snip again --



I have a tendency to reinvent methods to solve specific problems, then
find out later that I've done something that was invented in the 50's
(sometimes it's the 1950s, sometimes it's the 1850's, sometimes...).

I've done this more where you have a nice, repetitive input, which makes
adjustments to the feed forward more robust in the face of plant
nonlinearities. Making this work well for an arbitrary input would be
more difficult, but should be possible using adaptive techniques.

I'm going to have to check out Internal Model Control, though.

I just checked; it is, indeed, in the Control Handbook.

It came out of chemical process control (he says, quoting near verbatim)
in the early 70s. The smartest engineer I've ever met mentioned it once,
and as I was stuck with a very low sampling rate along with high
accuracy requirements, I looked into it. It turned out to be easy to
analyse and design, dead easy to stabilise, and on its own gave me
slightly better THD than my BLT+warping discretised PI loops with fancy
dead-time compensation (the dead-time compensation used up to the 7th
harmonic of the dead-time related voltage error waveform), and a higher
CLBW. And the implementation cost me, IIRC, one more MAC per d,q channel
than the quadrature PI loops, but removed truckloads of dead-time
compensation code.

It also lends itself well to adaptive control, inasmuch as its easy to
modify the parameters of the reference plant, and its still easy to
stabilise. its really just a variant of MRAC anyway. Oh yeah, It also
dealt with integrator windup too

I will be using it shortly, for some motor control stuff I'm doing, so
I'll be digging up my old notes and having a read. If I come across any
paper references, I'll post them.



I tested repetitive control for dealing with rectifier/capacitor loads,
and it wasnt as good as IMC

Would you like some nice Pacific Northwest (US) rain? We've got lots of
it this week.

We have some right now, although not really NW its the pacific (which
I'm on the other side of a mountain range from).

Cheers
Terry

 

[[email protected]]

[email protected] wrote in


Steady.. > They could, but things aren't so certain now at 84 for 5.
That's a worse position that England were in yesterday.

Australia seem to have recovered to 372 for 7, against English total of
159.

http://news.bbc.co.uk/sport1/hi/cricket/england/6210269.stm

Between the better bowlers and the better batsmen, the Australians do
seem to be difficult to beat.

 

[Joerg]

Australia seem to have recovered to 372 for 7, against English total of
159.

http://news.bbc.co.uk/sport1/hi/cricket/england/6210269.stm

Between the better bowlers and the better batsmen, the Australians do
seem to be difficult to beat.

I thought your wife is Dutch. That's where soccer is all the rage. Very
few people on the European continent know what cricket is. To be honest,
I don't either.

 

[[email protected]]

I thought your wife is Dutch. That's where soccer is all the rage. Very
few people on the European continent know what cricket is. To be honest,
I don't either.

My wife is Australian. But some of the Dutch do play cricket and there
are cricket teams and cricket competitions in the Netherlands. Soccer
may be a rage n the Netherlands, but field hockey is also popular and
both the mens and womens hockey teams do well in international
competitions - but then again, so do the Australian field-hockey teams.

I'm still playing "trimhockey" myself, which is a Dutch invention -
field hockey played by the relatively elderly to keep fit, with the
players being careful not to damage one another.

 

[[email protected]]

Hi Bill,

can you email me copies of those papers please? [email protected]

Sorry Terry - my attempt to reconstruct your e-mail address failed.
Send me an e-mail - my e-mail address isn't munged - and I'll send you
the two .pdf's.

 

[Terry Given]

Sorry Terry - my attempt to reconstruct your e-mail address failed.
Send me an e-mail - my e-mail address isn't munged - and I'll send you
the two .pdf's.

Oh, the handy underlining completely obscures the underscore. oops,
sorry. will do.

Cheers
Terry

 

[Joerg]

My wife is Australian. But some of the Dutch do play cricket and there
are cricket teams and cricket competitions in the Netherlands. Soccer
may be a rage n the Netherlands, but field hockey is also popular and
both the mens and womens hockey teams do well in international
competitions - but then again, so do the Australian field-hockey teams.

It never caught on in Germany. American football did though. Hockey is
(or at least used to be) big there, occupying quite some TV time.

I'm still playing "trimhockey" myself, which is a Dutch invention -
field hockey played by the relatively elderly to keep fit, with the
players being careful not to damage one another.

That's good not to get hurt. The worst injuries and long term damage
I've seen with sports folks were hockey players. Soccer as well but
those were mostly wear injuries like permanently shot knees.

 

[John Larkin]

I am afraid it's going to be tougher than anticipated because we don't
know what the plant looks like. Lots of individual delay elements. Then
there is the TEC itself which isn't linear at all over +/-10C. Of
course, none of that is spec'd anywhere :-(

In the end it might indeed boil down to a uC with lots of PID and feed
forward parameters in a lookup table that gets switched through as the
DFB module TEC passes through temperature ranges.



If it's any comfort, this stuff has arrived in the Pacific Southwest
about two hours ago. With the usual flash flood warnings and all that.

Advanced control algorithms presuppose that the actual "plant" state
can be accurately measured as the thing to feed back against. It's not
obvious to me that you have an accurate realtime wavelength sensor,
either in the production system or even for initial laser/tec
characterization.

Hell of a storm here last night. There's debris all over, lots of palm
tree bits on Dolores Street, and some traffic lights are out. Jim and
the boys should get it next.

John

 

[Jim Thompson]

Tim Wescott wrote:
[snip]

Would you like some nice Pacific Northwest (US) rain? We've got lots of
it this week.

If it's any comfort, this stuff has arrived in the Pacific Southwest
about two hours ago. With the usual flash flood warnings and all that.

[snip]

Hell of a storm here last night. There's debris all over, lots of palm
tree bits on Dolores Street, and some traffic lights are out. Jim and
the boys should get it next.

John

Yep. It's cloudy here... 50°F. Depressing ;-)

...Jim Thompson

 

[[email protected]]

It never caught on in Germany. American football did though. Hockey is
(or at least used to be) big there, occupying quite some TV time.

Checking ot the mens and womens field hockey results for the last few
Olympic games, Germany seems to do as well as the Netherlands and
Australia.

 

[Tim Wescott]

I am afraid it's going to be tougher than anticipated because we don't
know what the plant looks like. Lots of individual delay elements. Then
there is the TEC itself which isn't linear at all over +/-10C. Of
course, none of that is spec'd anywhere :-(

My understanding of TEC coolers is that they cool at a rate proportional
to KP*I - R*I^2 -- and the 'R' term really is a resistance. If this is
truly the case (and if you can characterize it) then it should be easy
to calculate a desired rate of heating, then calculate the necessary
current.

Of course, that doesn't mean that your laser's response to temperature
changes is at all linear.

In the end it might indeed boil down to a uC with lots of PID and feed
forward parameters in a lookup table that gets switched through as the
DFB module TEC passes through temperature ranges.

Eww. I don't like lookup tables, unless there's some linear
interpolation going on -- it always seems there's something odd
happening at the transitions.

Phil Hobbs's suggestion of using a heater with the TEC going full blast
sounds good to me -- I suppose it's too late to slip a heater in there,
isn't it?
--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Posting from Google? See http://cfaj.freeshell.org/google/

"Applied Control Theory for Embedded Systems" came out in April.
See details at http://www.wescottdesign.com/actfes/actfes.html