Searching \ for '[EE]: Current &' in subject line. ()
Help us get a faster server
FAQ page: www.piclist.com/techref/power.htm?key=current
Search entire site for: 'Current &'.

No exact or substring matches. trying for part
'[PIC]: (Basic) question about current & LEDs'
2002\03\03@215724 by

The PIC16F877 data sheet says any one pin can source a max of 25 mA. If I
want to drive a blue LED that has a forward current rating of 30 mA, can I
just connect that LED between the pin and ground, without a resistor?

Either the output voltage on the pin will drop because of the 25 mA source
limit, or the LED will draw 20 mA and the PIC will be damaged. I'm not sure
which will happen.

If I do a simple R = V / I calculation on a series LED & resistor, I come up
with a 200 ohm resistor. Is that the right way to determine the necessary
current-limiting resistor in this situation? Doesn't the LED also limit
current (it has a 3.5 - 4 V voltage drop)?

TIA,

--
Rick

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:

> The PIC16F877 data sheet says any one pin can source a max of 25 mA.
If I
> want to drive a blue LED that has a forward current rating of 30 mA,
can I
> just connect that LED between the pin and ground, without a resistor?
>
> Either the output voltage on the pin will drop because of the 25 mA
source
> limit, or the LED will draw 20 mA and the PIC will be damaged. I'm not
sure
> which will happen.
>
> If I do a simple R = V / I calculation on a series LED & resistor, I
come up
> with a 200 ohm resistor. Is that the right way to determine the
necessary
> current-limiting resistor in this situation? Doesn't the LED also
limit
> current (it has a 3.5 - 4 V voltage drop)?
>

The correct calculation is the following:
Series resistance = ( Supply voltage - LED's forward voltage ) / Desired
current
This way, supposing that Your power is 5V and a blue LED's forward
voltage is 4V, the resistor is:
(5V - 4V) / 25mA = 40 Ohm.

Only a blue LED vill work reliably in this circuit. If You replace it by
a red one, it will result a higher current, and the PIC may be damaged.

Sincerely:
Bela

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

> The PIC16F877 data sheet says any one pin can source a max of 25 mA. If I
> want to drive a blue LED that has a forward current rating of 30 mA, can I
> just connect that LED between the pin and ground, without a resistor?

No!  The data sheet doesn't say that 25mA is the max the pin CAN source,
only that it is the max you are allowed to let is source for all the other
specs to remain valid.

********************************************************************
Olin Lathrop, embedded systems consultant in Littleton Massachusetts
(978) 742-9014, olinembedinc.com, http://www.embedinc.com

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

Here's some other ideas about LED's:

1.  I would NEVER drive an LED directly from a PIC lead or transistor,
always use a current limiting resistor to protect the product.  Having said
this, I recently finished a project for a blinking LED bicycle tail-light
where I did exactly what I said not I would never do, but I was careful to
check the current involved while making it, and it is a one-off, not
production.

2. You should consider using the PIC to drive an external transistor any
time you are pushing the limits of the PIC's output port.  A 2n2222A or
similar transistor can be bought for two bits, a penny in volume, and will
produce a lower Vcesat voltage drop that your PIC port.  At 25 mA, the PIC
port starts to have a signifigant voltage across the output trasistor,
enough to change the calculation of your limiting resistor.

--Lawrence Lile

{Quote hidden}

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics

'[EE] Relatively High Current & Voltage (60V, 1A) d'
2010\04\27@223554 by
I have an application which basically consists of powering a device
which requires both 57VDC, and an interruption of that same voltage once
per second, for a total of around 100uS.   Rise/Fall times of under 1uS,
so it has to be 'driven' both high and low.

I've been spoiled to date with the LMD18200.. existing devices with the
same requirements only use 29VDC, which is well in the range of the
LMD18200.  I can actually drive two devices with each '200 - one on each
half of the H-bridge.   And, as far as I can tell, the LMD18200 is
nearly indestructable.   I'm not sure I've ever actually had a failure
out of hundreds (maybe almost thousands) of units shipped - and this is
in an environment where shorts, overcurrents, and impulses (aka
lightning hits) happen not infrequently.

Unfortunately the higher voltage of the newer devices (which I do not
build or control), necessitates using something other than the
LMD18200.   Using external FET's and FET drivers, I can definitely get
into my timing range, but overcurrent protection is an addon.  Along
with short circuit survivability and static-discharge (nearby lighting)
survivability.   By the time I put together what I think I need to do
this, it has grown from a clean design to something rube-goldberg-like.
Current sensing resistors, comparators, logic gates, etc. etc. etc.

What I'm hoping for is that I've either missed some LMD18200-like chip
which handles everything.. or perhaps something like that which uses
external fets and maybe sense resistors.  I will say that I have also
dealt in the past with issues where the devices I'm powering have a
large enough inrush current that doing simple overcurrent protection is
an issue.  By the time you've backed off the protection enough to allow
the device to power up, you've also effectively disabled short-circuit
protection (I think the L6206 may have been the part I was having issues
with in this regard, before I came across the LMD18200).  But in any
case, I guess I'm really mainly looking for some ideas which I'm
obviously missing here...

Thanks.

-forrest

Probably "a bit close to the wind, but perhaps BTS6163D ?has some value.
Only 1/4 of what you have now :-(.

http://www.infineon.com/dgdl/BTS6163D_DS_v10.pdf?folderId=db3a304314dca389011537739e37155f&fileId=db3a3043183a95550118606cdf2336a2

Can you "pedestal" your existing driver to raise it's voltage and do
something extra for the low side?

Much messier, but may give you it's functionality on the upper side.
Or lower (choose one).

R
TDA3654 is made for TV vertical deflection purposes.
You MAY be able to press something similar into service.
This is a 60V 1.5A part but there will robably be other TV oriented
ICs of better specs.

Horrendous datasheet access here.

http://pdf1.alldatasheet.com/datasheet-pdf/view/19290/PHILIPS/TDA3654.html

R
> Can you "pedestal" your existing driver to raise it's voltage and do
> something extra for the low side?

That was my thinking too, use two chips, one on a positive supply, one on a
negative, and when you want 0 volts across the load the positive supply one
switches to the low state, the negative supply one switches to the high
state.

>> Can you "pedestal" your existing driver to raise it's voltage and do
>> something extra for the low side?

> That was my thinking too, use two chips, one on a positive supply, one on a
> negative, and when you want 0 volts across the load the positive supply one
> switches to the low state, the negative supply one switches to the high
> state.

Some thinking through is in order, of course :-).
An off high side driver on the pedestal will tend to see full supply
voltage when the lower driver is turned on unless some way of
isolating it is used. Sounds messy.

Here's an app note ofr "Intersil's" HIP4081A with 95V rating driving a
quad of external N Channel MOSFETS.

http://www.intersil.com/data/an/an9405.pdf

Can achieve load current control with external circuitry (shown) and
has programmable dead time control etc.

Not as good as what you had but lots of voltage.

HIP4081A datasheet here

http://www.intersil.com/data/fn/fn3659.pdf

Despite the AN and DS length and detail some functionality is obscure.

Russell
> Here's an app note ofr "Intersil's" HIP4081A with 95V rating driving a
> quad of external N Channel MOSFETS.
>
>    http://www.intersil.com/data/an/an9405.pdf
>
> Can achieve load current control with external circuitry (shown) and
> has programmable dead time control etc.
>
> Not as good as what you had but lots of voltage.
>
> HIP4081A datasheet here
>
>      http://www.intersil.com/data/fn/fn3659.pdf
>
> Despite the AN and DS length and detail some functionality is obscure.

The HIP4081A is a nice chip. We had occasion to design the radiation
hardened version into a voice coil drive mechanism for a Sterling Helium
Cooler for a space instrument which is now flying on the Planck spacecraft.
I wasn't directly involved in doing that part of the design, but as I
recall, the engineer who did wasn't wandering about muttering about 'that
useless chip' like I have seen engineers sometimes do before they change
chips mid design.

Yeah, I had actually found the HIP4081A...  sort of the definition of
the best case scenario I had found so far...

The thought process was sort of, if all else fails, I'll use a HIP4081A
and oversize the FET's so that perhaps I can have a bit more leniency in
the short-circuit protection area.   Having not yet done the analysis,
it does seem that perhaps using say, 20A FET's in a TO-220 package (no
heatsink), and protecting it with a 1A PTC self-resetting fuse might
just be sufficient if appropriate impulse supression is added.

Other ideas?

-forrest

On 4/28/2010 3:49 AM, Alan B Pearce wrote:
{Quote hidden}

Forrest Christian wrote:
> I have an application which basically consists of powering a device
> which requires both 57VDC, and an interruption of that same voltage once
> per second, for a total of around 100uS.   Rise/Fall times of under 1uS,
> so it has to be 'driven' both high and low.
>
I built an RF amp using a pair of IRF740. If you can drive the gate
capacitance (I was  using about 8 Ohm source impedance) you can get sub
100ns switching even on 900V SMPSU FETS.  Some will do 10A. I think I
got about 80W RF using 36V supply at 14MHz

As a rule of thumb, the higher the current rating, the higher the gate
capacitance and lower drive impedance needed.  60V power FETs in TO220
can certainly do 50MHz, maybe 70MHz.

They show source resistors and the driver has 3 leads per FT - one of which
goes to the source to, presumably, get an optimum drive circuit - no ground
currents via the standard Gnd line etc. I thought at first that the source
R's were used on chip for current control but at a moderately quick skim
through that doesn't appear to be the case. BUT they show external current
control with an opamp and you could implement something similar with little
effort. If you wanted to do similar on the high side you could float an
opamp up there and use the high side gate boot strap supply to power the op
amp (as N channel gate must go comfortably above HV+. OR you could use high
side sense resistors and a current mirror to reflect the monitor current
down to ground where it's easier to deal with.

It's almost harder to talk about than to rough out a cct on paper. (or in LT
Spice if you are not me and Olin).

The 1A PTC could be a dread polyfuse.

Depends on how much overload current you expect. FETs are some of the best
fuses going ! :-(.

Russell

On 28 April 2010 22:19, Forrest W Christian <forrestcimach.com> wrote:

{Quote hidden}

> -
On Wed, Apr 28, 2010 at 6:19 AM, Forrest W Christian <forrestcimach.com>wrote:

> Yeah, I had actually found the HIP4081A...  sort of the definition of
> the best case scenario I had found so far...
>
> The thought process was sort of, if all else fails, I'll use a HIP4081A
> and oversize the FET's so that perhaps I can have a bit more leniency in
> the short-circuit protection area.   Having not yet done the analysis,
> it does seem that perhaps using say, 20A FET's in a TO-220 package (no
> heatsink), and protecting it with a 1A PTC self-resetting fuse might
> just be sufficient if appropriate impulse supression is added.
>
> Other ideas?
>
> -forrest
>

Don't oversize the FETs too much. The HIP4081A uses charge pumps for the
high side gates and you could have issues with switching speed if the gate
charge is too large.

A PTC is probably too slow to protect an FET. It might be OK if you drive
through a 1 or 2 ohm resistor. You only require 1 Amp, so the I*R drop of 1
or 2 volts might be acceptable.

--
Martin K.

> -----Original Message-----
> From: piclist-bouncesmit.edu [piclist-bouncesmit.edu] On
Behalf
> Of Russell McMahon
> Sent: 28 April 2010 12:41
> To: Microcontroller discussion list - Public.
> Subject: Re: [EE] Relatively High Current & Voltage (60V, 1A) driver
> protection.
>
> The 1A PTC could be a dread polyfuse.
>

You've probably posted them before, but what issues have you suffered
with polyfuses?  Are they devices you would avoid using if at all
possible?

Regards

Mike

=======================================================================
This e-mail is intended for the person it is addressed to only. The
information contained in it may be confidential and/or protected by
law. If you are not the intended recipient of this message, you must
not make any use of this information, or copy or show it to any