Well crap.

G-force · Post by **G-force** » Wed Nov 09, 2005 10:50 am

I spent all this money to buy some 3.3V regs at digi-key and now I find out they don't work. poopy!

Here's how I set it up, if you can tell me what I did wrong, or why they are so evil, that would help me a lot.

I put a 16V 1000 microfarad capacitor on the 7.2V 2000 mah battery line into the regulator. The on the 3.3V out line I put a 16V 470 microfard capacitor. I also put a big heatsink on the reg. The regulator I used was a 3.3V 1.5 amp, part # LM1086IT-3.3-ND

When I turned it on all I got was the startup screen of the game for 5 seconds before the reg got really hot (so fast the heatsink couldnt draw it away fast enough) and the N64 shut off.

This is the exact same results I got when I tried this a year or two ago.

I tested the battery, the output voltage, everything I could think of. Any ideas at all before I toss these things into the evil abyss of hell itself?

Maniac_Man · Post by **Maniac_Man** » Wed Nov 09, 2005 12:53 pm

maybe too much amperage draw from the n64? I always just use any car charger i can find. the other day i found one that was less than a square inch. i use the voltage from the led that most car chargers have.

Post by **Gamelver** » Wed Nov 09, 2005 1:49 pm

hmm....well, first of all, make sure the caps are across Vin and Vout (one side on positive, the other to ground)....second of all, isn't it supposed to be a 10uF tantalum cap for both? perhaps you just looked at a different circuit diagram....anyway, it's probably just the heat, since the caps are just to stablize the voltages, I believe. try a really big heatsink (like an N64 one) and if it works, then you'll know that it's the heat, and you'll have to get a heatsink big enough to accomodate it. or, you can get a fan

.

superdeformed · Post by **superdeformed** » Wed Nov 09, 2005 1:54 pm

Does it work without the N64 connected to it? If so, check all of the connections on the N64 side and make sure nothing is shorting to ground. If not, I'm sure you've already done this, but I suggest you check all of the connections on the regulator (correct polarity on capacitors between input/output and ground, make sure nothing is touching the tab except for Vout, etc). Also, those capacitors are complete overkill (in reference to the datasheet), try something smaller on the output. You should also check the current draw on it, because it definitely should not be more than 1.5 amps.

edit:

Gamelver wrote:anyway, it's probably just the heat, since the caps are just to stablize the voltages, I believe

Yeah, the caps are just to even out the input/output voltages. Technically you don't need the input one, but it doesnt hurt to have it there and it helps if there's a spike in current or voltage from the regulator. And the only reason it should shut down from heat after only a couple of seconds would be way too much current (or something wrong with the regulator itself). Just adding a bigger heatsink isn't a good way to permanently deal with a problem like that.

Anyways, my guess is that the output capacitor is just too big and it's drawing too much current to fill up when you turn everything on.

Turbo Tax 1.0 · Post by **Turbo Tax 1.0** » Wed Nov 09, 2005 2:17 pm

try wiring 2 in a parrelal thats what helped my 7808 out it barely gets hot now

Post by **Gamelver** » Wed Nov 09, 2005 2:22 pm

superdeformed wrote:Anyways, my guess is that the output capacitor is just too big and it's drawing too much current to fill up when you turn everything on.

*smacks forehead*...of course!! that might be it....try it without a cap on the output, since it doesn't really need to be perfect, since the N64 can handle +/- a few hundred millivolts on the 3.3V line

G-force · Post by **G-force** » Wed Nov 09, 2005 2:37 pm

I kinda missed the circuit, and just winged it

I was using elctrolytic caps.

I already tried using the reg without any capacitors at all and I still got the same results. Oh and the heatsink's not the problem, it's huge and it just can't dissipate the heat fast enough. I'm not sure if the heat is the problem though.

The only thing I can think of is putting two of them in parrallel

Turbo Tax 1.0 · Post by **Turbo Tax 1.0** » Wed Nov 09, 2005 3:29 pm

i think 2 in a parralel will work because the same thing happened with my screen's regulator yoiu might also be able to try one of those variable regulators from rad shack if this doesnt turn out well

timmeh87 · Post by **timmeh87** » Mon Nov 14, 2005 10:27 pm

you shouldnt use (big) electrolytic caps as decouplers. they have too much internal inductance, which entirely defeats the purpose.

G-force · Post by **G-force** » Thu Nov 17, 2005 2:52 pm

I'll keep that in mind next time, thanks.

Well finally I think this is getting somewhere. I was looking through my parts that I had ordered a while ago, and I happened upon some regulators from mouser. One was a 5V 1A, and the other was a 3.3V 1A. I remember somebody saying that the N64 only used something like .8 amp. So I set it up so that the 7.2V from the batterie would go to the 5V reg and the 5V would go into the 3.3V reg, then into the N64. This way the heat produced could be somewhat equally spread out into two TO-220's and 2 seperate heatsinks.

So it went like this: 7.2V battery>5V reg>3.3V reg>N64

On testing I found that the N64 worked fine (no resetting). There was still a bit of heat, but I didn't have heatsinks on at the time (I only ran it for 30 seconds before turning it off). I may need a very small inside fan to cool the heatsinks, but hopefully not.

Here is what was odd though. The screen (Hip gear 2.3" LCD) had only a weird bussing noise from the speakers, and the video was slighty garbled. I could't tell if it was the LCD getting some kind of weird interference, or the N64 couldn't pull enough power through the 1 amp regulator. I will try hooking the video and audio from the N64 to a normal TV, and see if I get the same result. If I do, it might be time to try some shielded wire.
Oh yeah, I didn't use any capacitors to stabilize the voltage, could that be the cause?

Post by **marshallh** » Thu Nov 17, 2005 2:54 pm

You need capacitors, and make sure all your grounds connect to the same place. What you are experiencing is feedback looping where interference is caused by the signals boucning around before they are grounded (hard to explain). But car audiophiles know all about it and make sure you take care of that.

G-force · Post by **G-force** » Thu Nov 17, 2005 3:42 pm

Yup. You're right. I just read the data sheet for it. It says:

"AME1117 requires a capacitor from VOUT to GND to provide compensation feedback to the internal gain stage. This is to ensure stability at the output terminal. Typically, a 10mF tantalum or 50mF aluminum electrolytic is sufficient.
(Note: It is important that the ESR for this capacitor does not exceed
0.5W.)"

So I need to go grab a 50 microfarad off some scrap electronics. /me looks around a bit.... Ah here we go, I found some that say "25V 47 uf". Will they work for this? One for each regulator output right?
(Sorry, I am more than a bit noobish at some of this. But hey, I'm learning)
And man, that earlier capacitor really was overkill. I didn't realize how much till now.

And I did ground everything to the same place, the N64 mobo. All I did was extend all the connections to/from the N64. Can you give me more details on what I should look for to fix this feedback you were talking about?

nos_slived · Post by **nos_slived** » Thu Nov 17, 2005 4:02 pm

The capacitors alone should fix most of the problems.

G-force · Post by **G-force** » Thu Nov 17, 2005 4:35 pm

I tried the above mentioned capacitors on the output lines of the regulators (One cap on each regulator output), I still got the exact same result. Perhaps a tantalum capacitor would work then.

Also, something weird is happening now that the batteries are not fully charged. They are 15 min rechargeable Ni-Mh, AAx6. When they are freshly charged they measure at 1.4V each, then after some use, quickly drop to 1.2V each. Right now they are at 1.3V. The N64 is resetting after the N64 logo and a few developers logos (the game is avalanche 1080) appear. I assumed that the caps would fix it, but since they didn't, I'm clueless.

What's really weird is the total voltage of all 6 batteries is dropping from 7.8V to 6.7V when the N64 turns on. Funky huh?

Please help me out here guys. I'm all out of ideas, and it's so close to being done! I mean REALLY done. I just have to screw it together once I fix this.

EDIT: Here's the data sheet for the regulators I'm using right now. Heh, meant to psot it earlier, but forgot. http://www.analogmicro.com/Datasheet/ame1117.pdf

timmeh87 · Post by **timmeh87** » Fri Nov 18, 2005 3:58 am

its normal for the voltage of batteries to drop under heavy load.

regarding the regs:
they are 1A regs, with "over current" protection. this implies to me that they will try to shut off if the '64 tries to draw more than 1A. which it probably will. i reccomend that you go read my thread about how to get yourself a free dc-dc.

also, using two regs in series might work, but youre dropping your overall efficiency to like 20%. which is terrible. a properly hooked up dc-dc is ~80% efficient. it also does not heat up nearly as much, eliminating the need to waste even MORE power to cool it. the only thing you gotta worry about is that they need at least 9V on the input side.

also, tantalium caps would probably be better than the electrolytics. (although it might not actually help your problem. but you gotta eliminate every variable right?) just make sure you DONT hook them up backwards. ive heard they self destruct rather violently.

what is the brand/model of the batteries you are using? they might not be able to handle the ~2A of current you are trying to pull out of them. if you can find a datasheet, look for the max current. if it is quoted in units of "C", "C" means "times the max capacity". so if you have a 2200mAh battery with a max discharge of .5C, then the max discharge current is 1.1A