PDA

View Full Version : Power problems



pilot00
04-05-2009, 02:09 AM
I have a 7.5 v 1amp power supply that I purchased for Parallax. I am running a ping sensor, three servos and a two line lcd.
The lcd flickers and goes out. Tech support suggested a larger capacitor (I am on a breadboard). I changed to a 2200 mf from the
1000 mf cap at the five volt output. The lcd continues to flicker. One servo is a standard that is panning. The other two are simulating
a tracked robot's motion. When the tracked simulation changes direction the change in current cuts out the lcd entirely.
According to my calculations and going over them with the support staff, I should not have a problem with power.
Any ideas?

Thanks

mctrivia
04-05-2009, 02:16 AM
seperate the motors from the screen with diodes. then place large caps at least 1000uF after each diode. if you have a 1000mF great but be careful with super caps like that you need current limiting resisters.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Need to make your prop design easier or secure? Get a PropMod (http://propmodule.com) has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.

pilot00
04-05-2009, 02:32 AM
I'll try the caps and diodes and see what happens. Thanks. I will post the results. I will check out the PropMod.

Philldapill
04-05-2009, 03:26 AM
Just for future reference, pilot00, you are probably using the wrong notation for your capacitors. mF means "milli-Farad", while uF means "micro-Farad". A 1000mF capacitor is a 1 Farad capacitor - a HUGE capacitor. Your capacitors are more likely to be 1000uF, which is 1mF, not 1000mF.

mctrivia
04-05-2009, 03:29 AM
and I just assumed he was trying super caps(not really.)

I do have this 1F cap very fun to play with (http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=P6955-ND)

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Need to make your prop design easier or secure? Get a PropMod (http://propmodule.com) has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.

Tesla
04-05-2009, 03:33 AM
Little of subject but i have 2 3000F 2.7v boost caps from maxwell technology........those are a force to be reckoned with. they will heat up jumper cables if you discharge it with them.

sorry to get off subject

mctrivia
04-05-2009, 03:46 AM
were did you get 3000F caps? to bad voltage is so low. I can invision making a pretty powerful rail gun with caps that big in the 1000V range.


How much?

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Need to make your prop design easier or secure? Get a PropMod (http://propmodule.com) has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.

Post Edited (mctrivia) : 4/4/2009 7:51:47 PM GMT

kwinn
04-05-2009, 03:54 AM
3000F at 1000V would be more like a rail cannon or possibly orbital launcher. Charging them might even cause a blackout http://forums.parallax.com/images/smilies/wink.gif

mctrivia
04-05-2009, 03:59 AM
orbital launcher is actually what I would like to build but I already figured out it would take over a 1km long rail gun to do it in reality and a ridiculous amount of power.

Would be really cool though.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Need to make your prop design easier or secure? Get a PropMod (http://propmodule.com) has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.

pilot00
04-05-2009, 07:32 AM
I tried the caps and diodes. Seems to be somewhat better. LCD continues to flicker but the power does not fail.
Thanks

pilot00
04-05-2009, 07:34 AM
Thanks for the milli versus micro I am aware of that, though. Is there a character set that I can access when using the forums???
Thanks

Philldapill
04-05-2009, 08:35 AM
The only character, and the most common, is just "u" to designate mu for micro. You could always just say "microFarad" or just "uF". Most everyone knows what you mean. :)

BTW, Tesla, I would also like a 1.5GJ capacitor bank. Man, the destruction you could do with something like that...(0.5*3000F*1000V^2) LOL

mctrivia
04-05-2009, 09:04 AM
do you know the current draw?

there are 2 reasons I can think why it may still blink.

1) cap size not enough for current. or you put the cap in series instead of between power and ground after the diode.

2) near by high frequency signal interfering. this can be a real problem if you don't have good grounding. high frequency signal lines should be surrounded by ground lines.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Need to make your prop design easier or secure? Get a PropMod (http://propmodule.com) has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.

mctrivia
04-05-2009, 09:07 AM
Philldapill with 1000/2.7 caps you could still build a marcs generator to get that power.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Need to make your prop design easier or secure? Get a PropMod (http://propmodule.com) has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.

Cluso99
04-05-2009, 09:31 AM
In 1970 my lecturer told us that a 1F capacitor was basically an impossibility - It would take the size of the lecture room. That's progress !!

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Links to other interesting threads:

· Home of the MultiBladeProps: TriBladeProp (http://forums.parallax.com/showthread.php?p=786418), SixBladeProp (http://forums.parallax.com/showthread.php?p=780033), website (Multiple propeller pcbs) (http://bluemagic.biz/cluso.htm)
· Single Board Computer:·3 Propeller ICs·and a·TriBladeProp board (ZiCog Z80 Emulator) (http://forums.parallax.com/showthread.php?p=790917)
· Prop Tools under Development or Completed (Index) (http://forums.parallax.com/showthread.php?p=753439)
· Emulators: Micros eg Altair, and Terminals eg VT100 (Index (http://forums.parallax.com/showthread.php?p=778427))
· Search the Propeller forums (via Google) (http://search.parallax.com/search?site=parallax&client=parallax&output=xml_no_dtd&proxystylesheet=parallax&proxycustom=<HOME/>&ie=&oe=&lr=)
My cruising website is: ·www.bluemagic.biz (http://www.bluemagic.biz)·· MultiBladeProp is: www.bluemagic.biz/cluso.htm (http://www.bluemagic.biz/cluso.htm)

mctrivia
04-05-2009, 09:55 AM
in traditional caps yes. supper caps don't work the same. to bad those 3000f caps are $150 I think I could run a prop off 1 of those for 10 years before it dropped to 1.8 v the lowest my boost supply can use.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Need to make your prop design easier or secure? Get a PropMod (http://propmodule.com) has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.

Peter Jakacki
04-05-2009, 11:05 AM
pilot00 said...
I have a 7.5 v 1amp power supply that I purchased for Parallax. I am running a ping sensor, three servos and a two line lcd.
The lcd flickers and goes out. Tech support suggested a larger capacitor (I am on a breadboard). I changed to a 2200 mf from the
1000 mf cap at the five volt output. The lcd continues to flicker. One servo is a standard that is panning. The other two are simulating
a tracked robot's motion. When the tracked simulation changes direction the change in current cuts out the lcd entirely.
According to my calculations and going over them with the support staff, I should not have a problem with power.
Any ideas?

Thanks


Just getting back to what you said originally because this is very important. It doesn't matter what calculations you do because once the motors immediately change direction you can at least expect effectively a short-circuit across your power supply. It is bad enough when a motor first starts but if it is already moving in one direction and then suddenly you reverse it then you are also "fighting" the back emf of the motor as it appears as a DC generator connected in reverse to your power. Arrange if you can a "dead" time in between changing direction and if possible PWM the motor up to speed. You will immediately find that you can get away without those enormous capacitors although they are never a bad thing to have. Once again though, use the diode + capacitor feed for the logic whenever you are hanging your motors off the same supply.

*Peter*

shanghai_fool
04-05-2009, 11:43 AM
Cluso99 said...
In 1970 my lecturer told us that a 1F capacitor was basically an impossibility - It would take the size of the lecture room. That's progress !!


I have almost 100 1.0F @5.5V that I purchased 10 yrs. ago for a project that never got off the ground. They are 3/4" dia. x 1/4" thick. So how much energy would that bag store?

Donald
·

mctrivia
04-05-2009, 11:46 AM
0.5*100F*5.5V^2=1.5kJ in paralel.

0.5*1F*550V^2=151kJ in series

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Need to make your prop design easier or secure? Get a PropMod (http://propmodule.com) has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.

mctrivia
04-05-2009, 11:54 AM
now imagine a marx generator (http://video.google.ca/videosearch?q=marx+generator&oe=utf-8&rls=org.mozilla:en-US:official&client=firefox-a&um=1&ie=UTF-8&ei=jCrYSaz_DZuwtAPlnMmlCg&sa=X&oi=video_result_group&resnum=4&ct=title#) with 3000F 1000V caps. That would be one powerful bolt.


not as much power. 1000x 495-3953-ND would give you 3,000,000 V enough power to jump over 3m

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Need to make your prop design easier or secure? Get a PropMod (http://propmodule.com) has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.

Post Edited (mctrivia) : 4/5/2009 4:02:49 AM GMT

Philldapill
04-05-2009, 11:56 AM
mctrivia, the arrangement dependance struck me as odd with the energy being different if they are in series or parallel. Then a light bulb went off. You are forgeting that the capacitance will be 1/100th of 1F when they are in parallel. The energy would be about 1.5kJ if they are in parallel OR series. Don't worry - it happens to the best of us. :)

mctrivia
04-05-2009, 12:02 PM
in parallel caps add up. I think you ment series. your right i forgot about the fact they divide in series.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Need to make your prop design easier or secure? Get a PropMod (http://propmodule.com) has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.

Philldapill
04-05-2009, 12:11 PM
Ha! You're right. See, it happens to the best of us!

Carl Hayes
04-05-2009, 09:27 PM
Philldapill said...
The only character, and the most common, is just "u" to designate mu for micro. You could always just say "microFarad" or just "uF". Most everyone knows what you mean. :)

BTW, Tesla, I would also like a 1.5GJ capacitor bank. Man, the destruction you could do with something like that...(0.5*3000F*1000V^2) LOL
If you are using Windows XP, you can add the Greek language support, very easy to do in Control Panel under Regional and Language Options.·

Then you can use whatever characters you want, and you can write μF or whatever you like.· Although, being a Florida Gator, I think mention of UF enhances any discourse.

Σεε?

With Icelandic characters, you can even write Anglo-Saxon.· Don“t you šink žęt“s nice?


▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i@arrl.net (mailto:nn5i@arrl.net)

Post Edited (Carl Hayes) : 4/5/2009 1:34:09 PM GMT

Carl Hayes
04-05-2009, 11:13 PM
pilot00 said...
I tried the caps and diodes. Seems to be somewhat better. LCD continues to flicker but the power does not fail.
Thanks
Get rid of the diodes.· All they're doing is reducing your available supply voltage.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i@arrl.net (mailto:nn5i@arrl.net)

Peter Jakacki
04-05-2009, 11:33 PM
Carl,

If he has them hooked-up correctly the diode feed followed by the cap should be going to the logic only whereas the motor connects directly to the input power. The motors don't really need any big caps. With a momentary heavy load on the supply the logic supply should hold up long enough until the motor starts moving.

*Peter*

Philldapill
04-06-2009, 02:02 AM
Carl, if I understand it correctly, and if this is the way he has them laid out, the diodes are preventing the caps from draining out with the sudden rush of current when the motor turns on. I had a similiar problem with a high-power(50A) mosfet switching circuit. The supply for the logic was the same as the supply for the motor(yes, bad practice). What would happen, is when the mosfets would turn on for the motor, there would be a massive voltage drop, causing my logic supply to bleed out to the motor. Often, this would cause a reset or other problems. Adding a diode in between the positive supply, and my logic supply caps fixed the problem. Current could fill up the capacitors, but would block the caps from discharging when the motor turned on.

Carl Hayes
04-06-2009, 03:39 AM
Philldapill said...
Carl, if I understand it correctly, and if this is the way he has them laid out, the diodes are preventing the caps from draining out with the sudden rush of current when the motor turns on. I had a similiar problem with a high-power(50A) mosfet switching circuit. The supply for the logic was the same as the supply for the motor(yes, bad practice). What would happen, is when the mosfets would turn on for the motor, there would be a massive voltage drop, causing my logic supply to bleed out to the motor. Often, this would cause a reset or other problems. Adding a diode in between the positive supply, and my logic supply caps fixed the problem. Current could fill up the capacitors, but would block the caps from discharging when the motor turned on.
But there are already diodes in the power supply.· Current cannot backdrain into the power supply, so those additional diodes do nothing.

Well, I suppose current could drain into the motor from the capacitors; that means merely that the power supply is inadequate to begin with.· The solution is not to place an additional 3/4 of a volt (or more) additional drop in line with the part of the circuit that is already suffering power loss.

Me, I'd power the motors and the electronics separately, or get an adequate power supply for both.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i@arrl.net (mailto:nn5i@arrl.net)

mctrivia
04-06-2009, 03:46 AM
If you take a motor and reverse its direction it will act as a dead short for a split second. This will drain all power from the system no mater how powerful your power supply is.

My car battery is rated at 1000A but when I start my car the lights will dim and the radio would shut off if not for the diode and .5F cap in there to keep it running.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Need to make your prop design easier or secure? Get a PropMod (http://propmodule.com) has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.

Fred Hawkins
04-06-2009, 05:23 AM
on Windows systems, µ = alt 0181. If you're using a laptop without a keypad, you have to shift into the numlock mode. (Read your manual)

Tesla
04-06-2009, 07:44 AM
I did buy them for 150$ a pop from someone that retails the BoostCap series of maxwell tech. They are rated 600amp discharge and charge. They are rated to completely short out with no damage. The low voltage is due to how they are constructed.

Maxwell does sell "power cache" of those cells wired up for 144v with balancing circuits, but i dont have that kind of money, with each 3kF cell at 100$ a pop.

I once hooked a small cdrom brushed dc motor to with with a cd on it and it spun for days.

Philldapill
04-06-2009, 07:46 AM
Carl, I think we can all agree that ideally, we should always have an "adequate" power supply. However, sadly, I didn't get one for christmas. :(

Since it sounds like he doesn't have an adequate power supply, the diodes seem to be needed. You can always get schottky diodes. I've got some with a 0.33V drop. You can get some even lower... $0.05 in diodes is well worth it compared to a beefier power supply, but that's just my opinion...

Tesla
04-06-2009, 08:12 AM
Is it just the LCD backlight or whole screen cutting out on you?

Carl Hayes
04-06-2009, 11:23 AM
kwinn said...
3000F at 1000V would be more like a rail cannon or possibly orbital launcher. Charging them might even cause a blackout http://forums.parallax.com/images/smilies/wink.gif
Wow, 1500 megajoules.· That's 'prox 417 KWH of energy, about 50-75 bucks on your light bill.· It'd take a while to charge it fully, and it would pack a terrific punch.· Lessee -- you could use it to tase woolly mammoths, Kzinti, or Bandersnatchi.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i@arrl.net (mailto:nn5i@arrl.net)

Philldapill
04-06-2009, 12:16 PM
Wow, Carl, I never thought about it like that... That's more than my electric bill in my apartment last MONTH!

I'm salivating at the thought of a 1.5GJ bank.

Now that's what I call EE porn.

pilot00
04-11-2009, 07:47 AM
Back to my original problem. I seperated the servos to battery power with common ground. No flicker and the voltage regulators stay cool. Any other ideas??
Thanks.

Toby Seckshund
04-11-2009, 05:02 PM
Everybody is worrying about the positive side of the power supplies. I was pondering on the negative sideas well, are all the "earths and supplies" fed dirrectly from the source, in a star?

Toby Seckshund
04-11-2009, 07:26 PM
And another thing....

I have found the reset cct of the prop (demo board ) a bit trigger happy. My 'scope being swithed off resets it even when not connected apart from the mains (via 9V SMPSU, 5V reg then 3.3V reg )

kwinn
04-11-2009, 10:49 PM
Attached are 3 schematics I use as a starting point for any power supply I build. I put a 3 pin terminal strip between the bridge rectifier/transformer center tap and the regulator board, and connect any motor/drive circuit directly to the terminal strip. I make sure one of the connections on the terminal strip is used as the common/ground point for all the circuitry to avoid problems with ground loops and noise pickup.
This layout works very well and the only time I have had problems with the microcontroller/logic resetting is when the motor draws more current than the transformer can supply for more than a couple of cycles of the AC line. If that happens you need a heftier transformer and bridge rectifier. It also works very well if the transformer/rectifier is replaced with a battery.

pilot00
04-12-2009, 12:41 AM
Howdy,

I scavanted a motorola cell phone v 500 ma supply and found it successful in eliminating the problem.
Thanks for all the suggestions.

Toby Seckshund
04-12-2009, 05:51 PM
I just went back to your original post question. You said that you purchased a 7.5V @ 1A PSU "for" parallax, did you mean from Parallax or some other some other place to be used for the Prop.
I only ask because there are all sorts of crap PSUs out there which may give the stated volts at that stated current, when open cct they give out more than double that with enough lump, bumps and spikes to rival the tail end of a hedgehog. Your phone charger will almost certainly be a switch mode and so properly regulated.

pilot00
04-14-2009, 09:21 AM
I fixed the power problem by using a separate supply for the servos.

Thanks for all the help.

Carl Hayes
04-14-2009, 01:54 PM
mctrivia said...
·supper caps don't work the same
I have always preferred breakfast caps, myself.·

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i@arrl.net (mailto:nn5i@arrl.net)

mctrivia
04-14-2009, 02:06 PM
I am sure that is a joke but it got lost some where. never herd of a breakfast cap

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Need to make your prop design easier or secure? Get a PropMod (http://propmodule.com) has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.

SRLM
04-14-2009, 02:10 PM
It was in reference to the meal time: http://en.wikipedia.org/wiki/Supper

mctrivia
04-14-2009, 02:13 PM
ok yes my spelling sucks

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Need to make your prop design easier or secure? Get a PropMod (http://propmodule.com) has crystal, eeprom, and programing header in a 40 pin dip 0.7" pitch module with uSD reader, and RTC options.