Searching \ for '[EE] Driving Dot Matrix LED Displays...' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/displays.htm?key=led
Search entire site for: 'Driving Dot Matrix LED Displays...'.

Exact match. Not showing close matches.
PICList Thread
'[EE] Driving Dot Matrix LED Displays...'
2009\02\24@173928 by solarwind

picon face
Hey all

Today I bought three 5x7 nice looking dot matrix LED displays. I can't
wait to make something nice out of it.

However, I didn't buy anything to drive it, which gives me a chance to
discuss the matter. How do I drive it?

I mean, I know that the LEDs are arranged in rows and colums which
means I can hook each pin up to the microcontroller and let it drive
the LED, but that would occupy 7 pins for each LED module. Another
website I came across
(http://www.best-microcontroller-projects.com/led-dot-matrix-display.html)
suggested the use of a 4017 CMOS decade counter to "walk" each row
while each LED in the column was driven by the microcontroller using a
multiplexing persistence-of-vision method. Since the 4017 CMOS can
only source 1 mA, a transistor array was used to supply enough
current. This raises several questions.

1. Do resistors need to be used to limit current? If so, how would
they be arranged?
2. I don't plan on using any CMOS 4017 to drive my LED displays.
Instead, I'll probably be using a 74HC595 or larger (serial to
parallel shift register). Can these serial to parallel shift registers
source/sink enough current to drive the LEDs?
3. What's the best way to drive such displays?

--
solarwind

2009\02\24@175951 by Marcel Birthelmer

picon face
>
> 3. What's the best way to drive such displays?
>
>
If it's a one-off, you might be able to use a sampled MAX7221 to do the job.
These things are about $10 apiece, but they're very easy to use - they
handle the multiplexing automatically, and you can specify things like
brightness and what to display by SPI.
They're meant for 7-segment displays (up to 8 digits), but I seem to
remember that they have a mode by which you can output arbitrary signals
(not just 0-9) on the pins. That way, each "digit" would be a column, and
each "segment" would be one LED in that column. Again, check the data sheet
to see if it's possible to make it work with your particular display.
- Marcel

2009\02\24@180143 by Mark Rages

face picon face
On Tue, Feb 24, 2009 at 4:39 PM, solarwind <spam_OUTx.solarwind.xTakeThisOuTspamgmail.com> wrote:
> Hey all
>
> Today I bought three 5x7 nice looking dot matrix LED displays. I can't
> wait to make something nice out of it.
>

...

> 1. Do resistors need to be used to limit current? If so, how would
> they be arranged?

Yes.  In series.

> 2. I don't plan on using any CMOS 4017 to drive my LED displays.
> Instead, I'll probably be using a 74HC595 or larger (serial to
> parallel shift register). Can these serial to parallel shift registers
> source/sink enough current to drive the LEDs?

I don't know, what does the datasheet say?

> 3. What's the best way to drive such displays?
>

It depends on how you define "best".  Your '595 is a reasonable choice.

Regards,
Mark
markrages@gmail
--
Mark Rages, Engineer
Midwest Telecine LLC
.....markragesKILLspamspam@spam@midwesttelecine.com

2009\02\24@182221 by solarwind

picon face
I just discovered that Maxim produces ICs that drive dot matrix
displays via SPI. They are very expensive but I ordered two samples
and they have to be approved first. I also sampled 2 MAX232 but those
didn't need to be approved.

2009\02\24@182721 by Bob Blick

face
flavicon
face

On Tue, 24 Feb 2009 17:39:25 -0500, "solarwind"
<x.solarwind.xspamKILLspamgmail.com> said:
> However, I didn't buy anything to drive it, which gives me a chance to
> discuss the matter. How do I drive it?

You could use a pic with enough pins, or buffer it. But basically
everything will be rows-and-columns. Typically in your multicharacter
displays you energize one row and send out a pattern to all columns.
then energize the next row and send out a new pattern, repeat to 7 and
repeat again. That way you end up with a 1/7 duty cycle which is much
brighter than if you did it the other way where you could have many,
many columns.

> 1. Do resistors need to be used to limit current? If so, how would
> they be arranged?

Resistors in series with the columns, since load on the rows depends on
what the pattern on the columns might be.

> 2. I don't plan on using any CMOS 4017 to drive my LED displays.
> Instead, I'll probably be using a 74HC595 or larger (serial to
> parallel shift register). Can these serial to parallel shift registers
> source/sink enough current to drive the LEDs?

For the columns, the 74HC595 is OK in room lighting. The row drivers
need to be much stronger. But you can get very strong versions of the
74HC595. Three different versions (there are more but these are the ones
that I typically lean toward): TPIC6C595, TPIC6595, TPIC6B595. Get the
datasheets for them and have a look.

Cheers,
Bob

--
http://www.fastmail.fm - Accessible with your email software
                         or over the web

2009\02\24@185532 by Byron Jeff

flavicon
face
On Tue, Feb 24, 2009 at 05:39:25PM -0500, solarwind wrote:
> Hey all
>
> Today I bought three 5x7 nice looking dot matrix LED displays. I can't
> wait to make something nice out of it.
>
> However, I didn't buy anything to drive it, which gives me a chance to
> discuss the matter. How do I drive it?

With a lot more power than you think.

> I mean, I know that the LEDs are arranged in rows and colums which
> means I can hook each pin up to the microcontroller and let it drive
> the LED, but that would occupy 7 pins for each LED module.

You still need to address the power aspects of multiplexing.

> Another
> website I came across
> (http://www.best-microcontroller-projects.com/led-dot-matrix-display.html)
> suggested the use of a 4017 CMOS decade counter to "walk" each row
> while each LED in the column was driven by the microcontroller using a
> multiplexing persistence-of-vision method. Since the 4017 CMOS can
> only source 1 mA, a transistor array was used to supply enough
> current. This raises several questions.

Now you are getting closer to the mark.

>
> 1. Do resistors need to be used to limit current?

Absolutely.

> If so, how would they be arranged?

One resistor for each LED in the column.

> 2. I don't plan on using any CMOS 4017 to drive my LED displays.
> Instead, I'll probably be using a 74HC595 or larger (serial to
> parallel shift register). Can these serial to parallel shift registers
> source/sink enough current to drive the LEDs?

No. Time for the multiplexing discussion. Say for example you are driving
15 columns. That means that each column is only active 1/15th of the time.
So to get brightness that matches a non multiplexed LED you need to drive
the LED with 15 times the current. If that is 15 mA for example then you'd
need to drive 225 mA through each LED.

This also means that each column driver would need to be able to sink a
maximum of 225 mA x 7 = 1525 mA.

You're going to need some serious drivers to drive the displays.

One way to cut down on these numbers is to change the duty cycle. For
example say you drive each display's row drives separately using a 74HC573
coupled to a set of transistors. Then you drive the same column on all
three displays at the same time. Now you've changed the column duty cycle
from 1/15th of the time to 1/5th of the time. So now you only need to drive
each LED with 5 x 15 mA = 75 mA. Now the column drivers still need to be
very stiff because it's sinking up to 21 LED. So 21 x 75 mA = 900 mA. But
then you only need 5 of them, instead of the 15 that you needed originally.

> 3. What's the best way to drive such displays?
>

There is no best way to do it. Each configuration has a tradeoff between
the complexity of the driver and the amount of power required to drive the
LED effectively. The key is managing the duty cycle.

Hope this helps,

BAJ

2009\02\24@185823 by Bob Blick

face
flavicon
face

On Tue, 24 Feb 2009 15:26:36 -0800, "Bob Blick" <.....bobblickKILLspamspam.....ftml.net>
said:

> For the columns, the 74HC595 is OK in room lighting. The row drivers
> need to be much stronger. But you can get very strong versions of the
> 74HC595. Three different versions (there are more but these are the ones
> that I typically lean toward): TPIC6C595, TPIC6595, TPIC6B595. Get the
> datasheets for them and have a look.

I forgot to mention that depending on the polarity of your displays, you
might have to use transistors to buffer the rows, since the TPIC chips
are open-drain outputs, they can only sink current, not source it.

Cheers,

Bob

--
http://www.fastmail.fm - Access your email from home and the web

2009\02\24@190822 by solarwind

picon face
Just found that this chip does it all: MAX6952EPL

2009\02\24@191652 by Byron Jeff

flavicon
face
On Tue, Feb 24, 2009 at 06:22:19PM -0500, solarwind wrote:
> I just discovered that Maxim produces ICs that drive dot matrix
> displays via SPI. They are very expensive but I ordered two samples
> and they have to be approved first. I also sampled 2 MAX232 but those
> didn't need to be approved.

That's certainly one way to do it. The problem usually is that chips like
that are often made of unobtainium. Maxim is especially notorious for this.

I would advise using this as a teachable moment. Using just 74HC595 and
transistors you can drive arbitrary sized displays with ease.

BAJ

2009\02\24@191751 by Bob Blick

face
flavicon
face

On Tue, 24 Feb 2009 19:08:19 -0500, "solarwind"
<EraseMEx.solarwind.xspam_OUTspamTakeThisOuTgmail.com> said:
> Just found that this chip does it all: MAX6952EPL


And at $16 each, I would expect it to brew coffee.

Is there a Maxim FAQ we can point you to? Basically, Maxim parts are:

Cool. They make a chip that does what I want!
Easily sampled.
Expensive!
Unavailable when you wish to purchase them.

So a lot of people avoid anything Maxim.

-Bob



--
http://www.fastmail.fm - I mean, what is it about a decent email service?

2009\02\24@193147 by solarwind

picon face
On Tue, Feb 24, 2009 at 7:17 PM, Bob Blick <bobblickspamspam_OUTftml.net> wrote:
> And at $16 each, I would expect it to brew coffee.

I know what you mean...

> Is there a Maxim FAQ we can point you to? Basically, Maxim parts are:
>
> Cool. They make a chip that does what I want!

Word.

> Easily sampled.

Love it.

> Expensive!

Hate it. But see above.

> Unavailable when you wish to purchase them.

Hate it. But see above.

> So a lot of people avoid anything Maxim.

Even MAX232? I just ordered two of those.

How long do the Maxim IC free samples take to get here and is there a
limit to how many you can order?

2009\02\24@211553 by William \Chops\ Westfield

face picon face

On Feb 24, 2009, at 4:31 PM, solarwind wrote:

>> So a lot of people avoid anything Maxim.
>
> Even MAX232?

Yes.  Everyone makes equivalent chips (pin compatible), and they're  
all cheaper.
I think I wound up with National 14C232 or something like that last  
time I bought a batch.  (although current digikey prices for the  
national part are not so good.  Hmm.)

BillW

2009\02\24@215115 by Matt Pobursky

flavicon
face
On Tue, 24 Feb 2009 16:17:49 -0800, Bob Blick wrote:
>
> On Tue, 24 Feb 2009 19:08:19 -0500, "solarwind"
> <@spam@x.solarwind.xKILLspamspamgmail.com> said:
>> Just found that this chip does it all: MAX6952EPL
>>
>
> And at $16 each, I would expect it to brew coffee.
>
> Is there a Maxim FAQ we can point you to? Basically, Maxim parts are:
>
> Cool. They make a chip that does what I want!
> Easily sampled.
> Expensive!
> Unavailable when you wish to purchase them.
>
> So a lot of people avoid anything Maxim.

There is much truth to these words. I have heard through my distribution
friends that Maxim is getting better but I suspect they still have a  long
way to go... at least to get any level of trust with me.

I have two real-world stories I can tell.

One of my good friends is a salesman for one of the largest electronics
distributors in the world. They are a Maxim distributor.

We were sitting around one evening having a few cold adult beverages and I
casually mentioned I was considering a Maxim battery charger IC for a
medical device I was designing. He got a very sour look on his face and
shook his head.

"That bad?", I said. Head nodding by my friend followed.

"So who uses Maxim parts? Hobbyists and companies who order millions of
parts a year?", I asked.

"Basically, yes. Or people who can order 100 from DigiKey and put them on
the shelf. Do yourself a favor and use TI or someone else instead. Parts
you'll actually be able to get." was his reply. And he *sells* them.

The 2nd story was from another friend who owns a contract manufacturing
company. He was lamenting to me that he couldn't ship product to one of his
biggest customers -- he couldn't get a part the customer had specified
-- as it happens the *exact* display driver IC Solarwind is considering.
Mind you he had a blanket order with his Maxim distributor for about 5000
pieces/month covering a whole year year.

None could be found anywhere in the Maxim distribution world. Lead time 22-
26 weeks. He ultimately found and bought some through a grey market dealer
(with the customer's approval) and paid almost 5x what his contract price
was through regular distribution. Crazy.

Every time I consider designing Maxim parts into one of my projects I try
to remind myself of these stories.

Matt Pobursky
Maximum Performance Systems

2009\02\24@220129 by Matt Pobursky

flavicon
face
On Tue, 24 Feb 2009 16:17:49 -0800, Bob Blick wrote:
>
> On Tue, 24 Feb 2009 19:08:19 -0500, "solarwind"
> <KILLspamx.solarwind.xKILLspamspamgmail.com> said:
>> Just found that this chip does it all: MAX6952EPL
>>
>
> And at $16 each, I would expect it to brew coffee.
>
> Is there a Maxim FAQ we can point you to? Basically, Maxim parts are:
>
> Cool. They make a chip that does what I want!
> Easily sampled.
> Expensive!
> Unavailable when you wish to purchase them.
>
> So a lot of people avoid anything Maxim.

I meant to mention in my earlier reply -- if you are looking for an
excellent alternative to the Maxim LED display driver ICs you should take a
look at Austria Microsystems LED drivers. They have a whole line of Maxim
compatible (pin and functional) LED display drivers. They are less than
half the price of Maxim, most have some nice functional improvements (like
2.7V operation) and are actually available. They have a reasonable sampling
program too.

They don't have a huge distribution network but are worth looking at. I use
them in several agricultural products and they have performed flawlessly.
For instance I use several thousand AS1108WL-T's a month and they cost less
than $2 ea.

http://www.austriamicrosystems.com/eng/Products/Lighting-Management/LED-Drivers

No affiliation with them, just a satisfied customer.

Matt Pobursky
Maximum Performance Systems


2009\02\24@220150 by William \Chops\ Westfield

face picon face

On Feb 24, 2009, at 3:55 PM, Byron Jeff wrote:

>>
>> 1. Do resistors need to be used to limit current?
>
> Absolutely.

Dissenting opinion:  No, you don't.


> No. Time for the multiplexing discussion. Say for example you are  
> driving
> 15 columns. That means that each column is only active 1/15th of the  
> time.
> So to get brightness that matches a non multiplexed LED you need to  
> drive
> the LED with 15 times the current. If that is 15 mA for example then  
> you'd
> need to drive 225 mA through each LED.

Except that LED brightness does NOT scale linearly with current, at  
least not up to 15x "typical" drive current.  And apparent brightness  
doesn't necessarily scale the way you'd expect, either.  You don't  
figure out how to pulse multiplexed displays with huge currents, you  
buy displays whose LEDs are bright enough with easily achievable  
currents and multiplex rates.  A reasonable dot-matrix LED display  
probably already picked pretty good LED chips... You can also play  
tricks like lighting fewer than a full row of LEDs at one time, to  
reduce current consumption for the single common lead.

For example, http://members.cox.net/berniekm/super.html
Four digit display, lights one segment at a time (x32 multiplexing!),  
no resistors, still seems to be bright enough and doesn't burn out the  
PIC...

(and, while it does use resistors, this charlieplexed hack of mine  
lights one of 20 LEDs at a time, with a max current of about 30mA, and  
also has acceptable brightness.
http://www.instructables.com/id/How-to-drive-a-lot-of-LEDs-from-a-few-microcontrol/)

BillW

2009\02\24@233030 by Byron Jeff

flavicon
face
On Tue, Feb 24, 2009 at 10:00:58PM -0500, William Chops Westfield wrote:
>
> On Feb 24, 2009, at 3:55 PM, Byron Jeff wrote:
>
> >>
> >> 1. Do resistors need to be used to limit current?
> >
> > Absolutely.
>
> Dissenting opinion:  No, you don't.

Bill,

Normally I don't deal in absolutes. But in this case Solarwind is the
audience. It's my personal assessment that explaining things to him in
absolutes gets better results than offering complex nuanced arguments.

In short he knows little about the process of which he asks. So keep it
simple. That means that when driving LEDs, use resistors to limit current.

{Quote hidden}

I know that both of the above points are true. But once again to simplify
the explanation, I went with KISS.

>  You don't  
> figure out how to pulse multiplexed displays with huge currents, you  
> buy displays whose LEDs are bright enough with easily achievable  
> currents and multiplex rates.

But that horse has already left the barn. Solarwind started with the
discussion "I bought these displays." No characteristics given. I pulled 15
mA to drive from thin air. It could be 1 mA to drive brightly, or 20 mA.
Who knows?

>  A reasonable dot-matrix LED display  
> probably already picked pretty good LED chips... You can also play  
> tricks like lighting fewer than a full row of LEDs at one time, to  
> reduce current consumption for the single common lead.

Again that adds complexity to the design. I gave a sample that reduced the
duty cycle and the expected load.

The point I was trying to drive home is that most likely that to drive the
display effectively is going to require some real drivers.
>
> For example, http://members.cox.net/berniekm/super.html
> Four digit display, lights one segment at a time (x32 multiplexing!),  
> no resistors, still seems to be bright enough and doesn't burn out the  
> PIC...

I'd be very leery to trust the current limit of the chip to limit the
current to the LEDs. What the harm in throwing in 180 ohm resistors to make
sure the current is capped?

> (and, while it does use resistors, this charlieplexed hack of mine  
> lights one of 20 LEDs at a time, with a max current of about 30mA, and  
> also has acceptable brightness.
> http://www.instructables.com/id/How-to-drive-a-lot-of-LEDs-from-a-few-microcontrol/)

Of course the brightness depends on the LEDs, right?

BAJ

2009\02\24@235829 by solarwind

picon face
On Tue, Feb 24, 2009 at 10:00 PM, Matt Pobursky <RemoveMEpiclistTakeThisOuTspammps-design.com> wrote:
{Quote hidden}

Thanks. Just filled out their survey with some random nonsense. My
samples should be on their way!

2009\02\25@013814 by Bob Blick

face
flavicon
face
Byron Jeff wrote:
> No. Time for the multiplexing discussion. Say for example you are driving
> 15 columns. That means that each column is only active 1/15th of the time.

Reality check here. In "real" LED signs, you do not multiplex the
columns. You multiplex the rows. So in Solarwind's case he has a 7x15
matrix, so done right it will be 1/7 multiplexed.

Cheerful regards,

Bob

2009\02\25@022011 by solarwind

picon face
On Wed, Feb 25, 2009 at 1:38 AM, Bob Blick <spamBeGonebobblickspamBeGonespamftml.net> wrote:
> Reality check here. In "real" LED signs, you do not multiplex the
> columns. You multiplex the rows. So in Solarwind's case he has a 7x15
> matrix, so done right it will be 1/7 multiplexed.

Hmm, that makes sense to me as this would better adapt to more modules
added on...

--
solarwind

2009\02\25@070743 by olin piclist

face picon face
solarwind wrote:
>> So a lot of people avoid anything Maxim.
>
> Even MAX232? I just ordered two of those.

TI, ST, and others make knockoffs that are a lot cheaper.  I won't use
Maxxim parts in a real design unless there is a very very good reason, which
so far hasn't happened.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2009\02\25@071626 by olin piclist

face picon face
solarwind wrote:
> Just filled out their survey with some random nonsense.

This is not a good idea.

> My samples should be on their way!

Or not.  They're random nonsense filters may be better than you think.

********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2009\02\25@072149 by Byron Jeff

flavicon
face
On Wed, Feb 25, 2009 at 02:19:29AM -0500, solarwind wrote:
> On Wed, Feb 25, 2009 at 1:38 AM, Bob Blick <TakeThisOuTbobblickEraseMEspamspam_OUTftml.net> wrote:
> > Reality check here. In "real" LED signs, you do not multiplex the
> > columns. You multiplex the rows. So in Solarwind's case he has a 7x15
> > matrix, so done right it will be 1/7 multiplexed.
>
> Hmm, that makes sense to me as this would better adapt to more modules
> added on...

Like everything else we've discussed here, this is not set in stone because
you run into the same issues. First you have to set enough column bits for
the entire row. Second you have to have a row driver that can source/sink
enough current to handle all of the LEDs in the module row being lit. So
you end up subdividing the row drivers as you add more modules.

The example that I gave where you drive multiple columns accomplishes the
same feat, with the same technology, with a better duty cycle (1/5).

Since there is no one single way to do it, you have to work with some
guiding principles:

1) Multiplexing will require more power than driving an LED continuously.

2) Managing the duty cycle is critical.

3) Serial to parallel chips like the 595 family are your best friend.

4) Subdivide packs to keep power management reasonable.

5) Driving LED modules can be done with garden variety technology. So you
don't need specialized chips to do the task.

There's no one single way to do it. Noodle around with some different
configurations until you settle on one that works for you.

BAJ

2009\02\25@081350 by solarwind

picon face
On Wed, Feb 25, 2009 at 7:17 AM, Olin Lathrop <RemoveMEolin_piclistspamTakeThisOuTembedinc.com> wrote:
> solarwind wrote:
>> Just filled out their survey with some random nonsense.
>
> This is not a good idea.
>
>> My samples should be on their way!
>
> Or not.  They're random nonsense filters may be better than you think.

I just got another email saying that the order has been in the mail
and has been shipped out from Asia; "in transit".

2009\02\25@081536 by solarwind

picon face
On Wed, Feb 25, 2009 at 7:21 AM, Byron Jeff <byronjeffEraseMEspam.....clayton.edu> wrote:
> 1) Multiplexing will require more power than driving an LED continuously.

Really? I thought it requires _less_.

> 2) Managing the duty cycle is critical.

Yea.

> 3) Serial to parallel chips like the 595 family are your best friend.

Yea.

> 4) Subdivide packs to keep power management reasonable.

Yea.

> 5) Driving LED modules can be done with garden variety technology. So you
> don't need specialized chips to do the task.

Yea.

> There's no one single way to do it. Noodle around with some different
> configurations until you settle on one that works for you.

Yea.

2009\02\25@092210 by joseph

face
flavicon
face


Original Message:
-----------------
From:  EraseMEolin_piclistspamembedinc.com (Olin Lathrop)
Date: Wed, 25 Feb 2009 07:17:05 -0500
To: RemoveMEpiclistEraseMEspamEraseMEmit.edu
Subject: Re: [EE] Driving Dot Matrix LED Displays...


solarwind wrote:
> Just filled out their survey with some random nonsense.

This is not a good idea.

> My samples should be on their way!

Or not.  They're random nonsense filters may be better than you think.

I'll agree.  I have found most distributers to be very accomodating to
hobbyists.  I've received most samples I've ever requested from companies
like AMD, Maxim, and Microchip, even though said use was for only a hobby
project with a production of no more than two.  In times past, your
hobbyists are what turned into HP, Apple, and Microsoft.  Maybe these parts
suppliers view a few dollars spent on samples a good investment.

Joe


--------------------------------------------------------------------
mail2web.com – Enhanced email for the mobile individual based on Microsoft®
Exchange - http://link.mail2web.com/Personal/EnhancedEmail


2009\02\25@094445 by Thomas Coyle

picon face
I've always been impressed with Maxim's accommodation; I don't ever recall
having a sample request turned down, and I've always been honest about the
intended use. Sometimes it's for something potentially-marketable, other
times it's just for obscure hobbyist work.
What really impresses me is that they've always sent me the samples via
Fedex. :)
Is there a consolidated list anywhere of the hobbyist-sample-friendly
companies?
On Wed, Feb 25, 2009 at 6:21 AM, RemoveMEjosephspam_OUTspamKILLspamkirtland.com <RemoveMEjosephTakeThisOuTspamspamkirtland.com>wrote:

{Quote hidden}

>

2009\02\25@111501 by olin piclist

face picon face
solarwind wrote:
>> 1) Multiplexing will require more power than driving an LED
>> continuously.
>
> Really? I thought it requires _less_.

To the first order approximation it is the same power.  All the LEDs get lit
to the same apparent brightness, so it's roughly the same power.  However,
multiplexing causes individual LEDs to be run with high power for part of
the time instead of the average power all the time.  This requires higher
current, causing higher I*R losses.  The light output to current ratio for
LEDs also falls a bit for higher currents.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2009\02\25@131042 by Benjamin Grant

flavicon
face
just as a note in general -- I have acquired plenty of free samples in order
to build the EEG recorder I've mentioned before and never had to lie about
my intended use. I'd just be honest about your intentions - they don't
really care in general
ben

2009\02\25@134859 by Michael Algernon

flavicon
face
>
> On Feb 25, 2009, at 6:13 AM, solarwind wrote:
>
> On Wed, Feb 25, 2009 at 7:17 AM, Olin Lathrop <olin_piclistSTOPspamspamspam_OUTembedinc.com
> > wrote:
>> solarwind wrote:
>>> Just filled out their survey with some random nonsense.
>>
>> This is not a good idea.
>>
>>> My samples should be on their way!
>>
>> Or not.  They're random nonsense filters may be better than you  
>> think.
>
> I just got another email saying that the order has been in the mail
> and has been shipped out from Asia; "in transit".

Would you like someone to do that to your company ?
MA


2009\02\25@150428 by Byron Jeff

flavicon
face
On Wed, Feb 25, 2009 at 01:48:53PM -0500, Michael Algernon wrote:
> >
> > On Feb 25, 2009, at 6:13 AM, solarwind wrote:
> >
> > On Wed, Feb 25, 2009 at 7:17 AM, Olin Lathrop <spamBeGoneolin_piclistSTOPspamspamEraseMEembedinc.com
> > > wrote:
> >> solarwind wrote:
> >>> Just filled out their survey with some random nonsense.
> >>
> >> This is not a good idea.
> >>
> >>> My samples should be on their way!
> >>
> >> Or not.  They're random nonsense filters may be better than you  
> >> think.
> >
> > I just got another email saying that the order has been in the mail
> > and has been shipped out from Asia; "in transit".
>
> Would you like someone to do that to your company ?

He's a teenager. Doesn't have a company.

No reason to lie though.

BAJ

2009\02\25@154828 by solarwind

picon face
On Wed, Feb 25, 2009 at 9:44 AM, Thomas Coyle <KILLspamzxcasdspamBeGonespamgmail.com> wrote:
> I've always been impressed with Maxim's accommodation; I don't ever recall
> having a sample request turned down, and I've always been honest about the
> intended use. Sometimes it's for something potentially-marketable, other
> times it's just for obscure hobbyist work.
> What really impresses me is that they've always sent me the samples via
> Fedex. :)

I can't wait to get them.

> Is there a consolidated list anywhere of the hobbyist-sample-friendly
> companies?

Haven't found one yet, but would be really cool to make one!




On Wed, Feb 25, 2009 at 2:17 PM, Byron Jeff <EraseMEbyronjeffspamEraseMEclayton.edu> wrote:
> He's a teenager. Doesn't have a company.
>
> No reason to lie though.

I had to lie because I thought that these companies will only give out
samples to potential multi-million dollar businesses that buy millions
of chips. Why would they waste all that money plus shipping costs to
send these out to a hobbyist? I guess I was wrong. I won't do it
again.

2009\02\25@162429 by Alan B. Pearce

face picon face
>just as a note in general -- I have acquired plenty of free
>samples in order to build the EEG recorder I've mentioned before
>and never had to lie about my intended use. I'd just be honest
>about your intentions - they don't really care in general

Well, to a point.

Just watch the small print, as many companies will disclaim any liability
when their product is used in life support or critical applications, without
written authorization from them. Now if you are mentioning that you intend
to use these samples for an EEG machine, as this is in the medical field
they may refuse the samples. It sounds like you have the samples, so they
may regard an EEG as not a problem, despite potentially being connected to a
person in a manner that could electrocute them if something failed.

2009\02\25@165716 by Tamas Rudnai

face picon face
On Wed, Feb 25, 2009 at 8:48 PM, solarwind <@spam@x.solarwind.x@spam@spamspam_OUTgmail.com> wrote:

> I had to lie because I thought that these companies will only give out
> samples to potential multi-million dollar businesses that buy millions
> of chips. Why would they waste all that money plus shipping costs to
> send these out to a hobbyist? I guess I was wrong. I won't do it
> again.
>

THat's the idea of the 'sample', right? They want to win potential customers
that will make real orders - at least that's what they hope when sending the
free sample to your door using Fedex. If you need a component to your
project you supposed to buy one. Sorry, that's my opinion.

Tamas
--
Rudonix DoubleSaver
http://www.rudonix.com

2009\02\25@182537 by Benjamin Grant

flavicon
face
my eeg was for a Duke class, the intention was made clear in the request for
sample. Sure, they won't give samples for life saving devices, but an EEG
for proof of concept(we made a halter monitor EEG) isn't made to diagnose or
for life support purposes.  I mean... you could electrocute yourself with
any device medical or not if you built incorrectly.. but my EEG ran off a 9
V battery so I wasn't too worried about it
ben

tamas - your opinion is yours to have, but if you write to a company and
fill out their form with exactly what you'er using it for and they're fine
with that, why would that be morally ambiguous? I.e. ordering EEG
componenets from TI and telling them you are not producing these
commercially nor do you plan on it. Why do they still give them to you?
Because they're not dumb, they know a percentage of electrical engineering
studetns will go on to design large scale products later in life and will
hopefully remember the quality products company A provided them. Anyway, I
mean lying to obtain something is a lot different than being honest, and if
TI and Maxim are willing to supply under such circumstances, doesn't seem
like your argument holds much weight "You're supposed to by one"...
according to whom? Again, if companies want to entice you so that you'll buy
from them in the future... that's their decision

2009\02\25@194911 by Tamas Rudnai

face picon face
I am not arguing with this - just thinking that for students there are other
options companies usually have, like campus programs where they support
classes or projects in universities etc or just have special prices for
teachers/students. Hobbyst is a hobbyst, not a student. However, everything
has a marketing value including statistical ones like a making survey what
components are used by whom to help targeting the products, or if someone
makes a popular open design project with their selected devices then it
makes business for sure. I just have my opinion about people exploiting
things like this - collecting free samples no matter if those are useful or
not, but it's free so why not having it... not my style, that's all I wanted
to say.

Tamas


On Wed, Feb 25, 2009 at 11:25 PM, Benjamin Grant <spamBeGonebenjamin.grantspamKILLspamduke.edu>wrote:

{Quote hidden}

> -

2009\02\25@204331 by Joseph Bento

face
flavicon
face

On Feb 25, 2009, at 5:48 PM, Tamas Rudnai wrote:

{Quote hidden}

I will generally request a sample when the price break requires a  
minimum purchase of many units at Digikey or Mourser.  Less  
frequently, neither may have the part.  In either case, the  
manufacturers have been very generous at honoring my infrequent  
requests.  One time the request was for a VFD driver made by Maxim.  
Digikey had a 100 piece minimum for this part.  Maxim sent me two when  
I clearly stated it was for a hobby project.  I have since learned  
that Maxim also has a direct order option with no minimum that I have  
since used when needing additional parts.

Joe





2009\02\26@074630 by olin piclist

face picon face
solarwind wrote:
> I had to lie because I thought that these companies will only give out
> samples to potential multi-million dollar businesses that buy millions
> of chips.

No, you didn't have to lie.  You engaged in deliberate fraud to try to get
something for free.  You could of course always buy the parts or use
alternatives.  Nobody is obligated to give you their products for free.
Whether you like it or not, it's their call to make.

> Why would they waste all that money plus shipping costs to
> send these out to a hobbyist?

Some don't, some do.  Those that do probably figure it's not worth the
trouble trying to distinguish the hobbyists from the real customers, or they
figure some hobbyists might turn in to real customers down the road.  As
long as the total sample volume is tiny compared to the for-profit volume,
they probably just send the samples.  Some companies, like Microchip, get
abused more than others by ditbags trying to get free stuff.  Microchip
regularly adjusts their sample policy as a result, most recently just a few
months ago.  A while back they had people getting free samples and then
selling them on ebay.  I can't blame them from clamping down, but
unfortunately this has made things more difficult for legitimate guys like
me.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2009\02\26@075142 by olin piclist

face picon face
Benjamin Grant wrote:
> but if you write to a company
> and fill out their form with exactly what you'er using it for and
> they're fine with that, why would that be morally ambiguous? I.e.
> ordering EEG componenets from TI and telling them you are not
> producing these commercially nor do you plan on it. Why do they still
> give them to you? Because they're not dumb, they know a percentage of
> electrical engineering studetns will go on to design large scale
> products later in life and will hopefully remember the quality
> products company A provided them.

I think you're reading too much into this and making incorrect assumption
about how the information you provide is used.  I don't know for sure
either, but I wouldn't be too suprised if the samples get sent anyway and
the information gets forwarded to sales that then decide to follow up or
not.  In other words, the stuff you fill out probably has nothing to do with
whether you get the samples, only what level of sales followup you get.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

More... (looser matching)
- Last day of these posts
- In 2009 , 2010 only
- Today
- New search...