Unoffical empeg BBS

Quick Links: Empeg FAQ | RioCar.Org | Hijack | BigDisk Builder | jEmplode | emphatic
Repairs: Repairs

Topic Options
#115080 - 06/09/2002 11:27 ~OT: 5V power....VSS serial interface capacity
jimhogan
carpal tunnel

Registered: 06/10/1999
Posts: 2591
Loc: Seattle, WA, U.S.A.
I have two issues that, while they are not really Empeg-specific, I thought would sit better here than in OT.

#1 5V power supply

I need to build a 5V DC power supply to run a Netgear WAP on my boat. It's OEM AC power adapter is rated at 1.0 amp output. The new power supply will run off the 12-volt batter bank on the boat.

I found a schematic for a very simply step-down power supply that even an electronics idiot like me might be able to follow. It says that the 7805 can be increased to 1A output by adding a heatsink, but it doesn't say if any other changes need to be made (other than a heatsink). Also, I have in my possession a 7805 CK that does the same thing and is rated up to 3A.

My question, if I built this with the 3A-capable 7805CK, would I need to adjust the values of the other components (capacitors)?

Also, one of the statements on that page "Change the transformer from where you take the power to the circuit to a model which can give as much current as you need from output" just didn't make sense.

#2 Taking speed sensor (VSS) input on the serial port

I am installing another axle-mounted, drop-to-ground pulse, speed sensor on my car. I estimate that with my tire size it will generate between 750-800 pulses per mile, so say 750 pulses per minute at 60MPH and 1500 PPM at 120MPH.

At the high limit of 120 MPH, that would nmake around 25 pulses per second. My intent is generally to follow the interface diagram here and the author of that software mentions the shortcoming of RS232/serial handling under Windows. my question is whether this rate of serial port pulses-per-second would be an issue if I were going to try to implement it on a Linux laptop or the Empeg. Is there any other serial library of something that I should be looking at? I think this may have come up in a previous speed sensor thread, but i couldn't find it...

(Man, I am in *serious* work avoidance mode this Friday morning!)
Thanks,
_________________________
Jim


'Tis the exceptional fellow who lies awake at night thinking of his successes.

Top
#115081 - 06/09/2002 13:48 Re: ~OT: 5V power....VSS serial interface capacity [Re: jimhogan]
mtempsch
pooh-bah

Registered: 02/06/2000
Posts: 1996
Loc: Gothenburg, Sweden
I now realize that I've forgotten the design rules for the 78xx... Oh well...

Have you tried to measure the actual current draw of the WAP? Might well be that it doesn't draw more than 0.5-0.6A or so.

Also, one of the statements on that page "Change the transformer from where you take the power to the circuit to a model which can give as much current as you need from output" just didn't make sense.

Well, not if you run from a battery bank capable of delivering 100s of ampere... But if your power comes out of a transformator (110V -> say 11V), you need one that can deliver the amount of current you want. If you get a trafo that can/is designed to deliver 0.5A, you shouldn't try to draw 3A out of it...

Some pointers though - the big cap on the left is there to smooth out the ripples that you have when your power comes from a trafo and diode rectifier bridge. The more current you draw, the bigger the cap. I seem to recall a rule of thumb of 1000uF per 1A.
But running from batteries, the need is greatly reduced... I'd probably go with 470uF or so, for any ripples that might be when running the motor/alternator.

IIRC there should also be smaller ceramic caps right by the regulator, both on input and output - my mind says 100nF, but I'm far from sure...

Whatever model you get, I'd try to attach it to something that could help cool it - at 1A you'll be dumping about 8W in the regulator.

/Michael
_________________________
/Michael

Top
#115082 - 06/09/2002 13:53 Re: ~OT: 5V power....VSS serial interface capacity [Re: jimhogan]
number6
old hand

Registered: 30/04/2001
Posts: 745
Loc: In The Village or sometimes: A...
re: VSS speed sensor.

any box that can run Linux can easily do this - 25 pulses a second is nothing.

But - the problem you have (with the Empeg in particular) is that there are no spare input pins to the Empeg, the 2 that the serial interface has are multiplexed for phone and display dim {headlight sense).

So, you could use one of them, but you'd have to sacrifice one of these 2 features.

For Rallying, I suspect the Cellphone mute is probably the one to be used here - heck who wants incoming phone calls when trying to stay on the road right?

Its then a simple Kernel mod to "count pulses" and then store this information somewhere accessible [e.g. /proc/empeg_XXXXX file].

Second issue then is having a program to show this information.

For a regular Linux OS [other than Empeg], then any regular serial port will do the trick, but again you'd have to write some software to do the polling and display.

But using a non windows OS for this software/hardware could be done.
Its a matter of someone doing the programming.




Top
#115083 - 06/09/2002 15:08 Re: ~OT: 5V power....VSS serial interface capacity [Re: jimhogan]
pca
old hand

Registered: 20/07/1999
Posts: 1102
Loc: UK
The 78xx series regulators are pretty bomb-proof. Assuming you're using a recent, overload protected version, it's almost impossible to kill them.

The same circuit will work with almost any current rating. I'd suggest adding another 100nF capacitor across the input to the thing as well, and if the current is going to be substantial (over an amp), double the size of the input and output electrolytics. You might as well use 220uF on the input and 100uF or above on the output, this should make it work with any capacity 7805 you want. It'll work with lower values, but will have more ripple under load.

Use a reasonable heatsink, since linear regulators aren't noted for efficiency (they simply dump the excess energy as heat, so if you're pulling 1A @ 5V, ie 5W, and the thing is running from a 12v supply, you are dissipating 1A @ 7V or 7W, ie the difference), and you can't go far wrong.

pca
_________________________
Experience is what you get just after it would have helped...

Top
#115084 - 09/09/2002 08:40 Re: ~OT: 5V power....VSS serial interface capacity [Re: jimhogan]
JBjorgen
carpal tunnel

Registered: 19/01/2002
Posts: 3584
Loc: Columbus, OH
as an alternative, you could purchase the Arise ACE-845v. It would give you 6A of 5v power and also a source of regulated 12v for any other electronics you wanted to run. This info is a bit stale (several months) but when I bought mine they were $77.
_________________________
~ John

Top
#115085 - 09/09/2002 09:49 Re: ~OT: 5V power....VSS serial interface capacity [Re: jimhogan]
jimhogan
carpal tunnel

Registered: 06/10/1999
Posts: 2591
Loc: Seattle, WA, U.S.A.
Do I love this BBS, or what? Look at responses you get for a marginally on/off-topic questions posed by the electronics village idiot!

mtempsch:
Have you tried to measure the actual current draw of the WAP? Might well be that it doesn't draw more than 0.5-0.6A or so.


I haven't, but one of the reasons I went with the 3A-capable 7805 is that I may try to provide a small 5V bus to power 2-3 items (like the used GPS I just bought for $25 from BG Micro)

Also, one of the statements on that page "Change the transformer from where you take the power to the circuit to a model which can give as much current as you need from output" just didn't make sense.

Well, not if you run from a battery bank capable of delivering 100s of ampere... But if your power comes out of a transformator (110V -> say 11V), you need one that can deliver the amount of current you want. If you get a trafo that can/is designed to deliver 0.5A, you shouldn't try to draw 3A out of it...


Ah, what he's saying makes sense now..


Some pointers though - the big cap on the left is there to smooth out the ripples that you have when your power comes from a trafo and diode rectifier bridge. The more current you draw, the bigger the cap. I seem to recall a rule of thumb of 1000uF per 1A.
But running from batteries, the need is greatly reduced... I'd probably go with 470uF or so, for any ripples that might be when running the motor/alternator.

IIRC there should also be smaller ceramic caps right by the regulator, both on input and output - my mind says 100nF, but I'm far from sure...

Whatever model you get, I'd try to attach it to something that could help cool it - at 1A you'll be dumping about 8W in the regulator.


One application for this power supply is that driver's display VFD. I may try to built it into the box and amd thinking I can put an aluminum heat sink along the back -- if I'm reading it right the body/shell of the 7805CK is ground/earth, so should be OK if it is exposed...

number6:
any box that can run Linux can easily do this - 25 pulses a second is nothing.

But - the problem you have (with the Empeg in particular) is that there are no spare input pins to the Empeg, the 2 that the serial interface has are multiplexed for phone and display dim {headlight sense).

So, you could use one of them, but you'd have to sacrifice one of these 2 features.


Got it. Actually, I think my only Empeg application will be to display a big clock from a GPS source. The answer is still relevant, though as we are thinking of running software on other linux devices (whether laptop with Linux of Zaurus). Yes, 25PPS seemed like nothing. I'm not sure what Glenn's (the TSD computer software author) problems were, but suspect they were probably Windows-specific when he switched from the original DOS/parallel port version...

For Rallying, I suspect the Cellphone mute is probably the one to be used here - heck who wants incoming phone calls when trying to stay on the road right?

CORRECT!

Its then a simple Kernel mod to "count pulses" and then store this information somewhere accessible [e.g. /proc/empeg_XXXXX file].
Second issue then is having a program to show this information....

I'm going to look at some source for stuff like gpsd and at Glenn's serial source to see if I can make heads and/or tails!

pca:
The 78xx series regulators are pretty bomb-proof. Assuming you're using a recent, overload protected version, it's almost impossible to kill them.

The same circuit will work with almost any current rating. I'd suggest adding another 100nF capacitor across the input to the thing as well, and if the current is going to be substantial (over an amp), double the size of the input and output electrolytics. You might as well use 220uF on the input and 100uF or above on the output, this should make it work with any capacity 7805 you want. It'll work with lower values, but will have more ripple under load.

Use a reasonable heatsink, since linear regulators aren't noted for efficiency (they simply dump the excess energy as heat, so if you're pulling 1A @ 5V, ie 5W, and the thing is running from a 12v supply, you are dissipating 1A @ 7V or 7W, ie the difference), and you can't go far wrong.


Thanks! I think can follow that (Yes, finally a project that may fall within my meager soldering and electonics skills!!) I got a few fairly large heatsinks from AllElectronics that I think will fit nicely on the back of this display/project box.

meatballman:
as an alternative, you could purchase the Arise ACE-845v. It would give you 6A of 5v power and also a source of regulated 12v for any other electronics you wanted to run. This info is a bit stale (several months) but when I bought mine they were $77.


Thanks for that link. As I said up above, I'm thinking of supplying a 5V bus for a number of devices (computer, scanner, WAP), so that might be a cleaner alternative, say for the boat. In the car, though, there is so little space, I think I'm going to take a whack at building that simple circuit with the suggested improvements.

/Jim now wondering about RF/shielding!!

Thanks!!
_________________________
Jim


'Tis the exceptional fellow who lies awake at night thinking of his successes.

Top