Quark fried, blown capacitor...

[BrianG]

For any kind of clamping diode or device, you select the voltage that you want to clamp at. It's not like a capacitor where you choose one about 15% higher voltage than the voltage you plan to run at. Kinda like a regulator; you don't choose a 6v if you want 5v...

[zeropointbug]

No, i know, it's just that 6s Lipo is 25.2 volts, I know no one runs it, and the controller is meant for 25v.... just a thought. I guess for most our setups, the 25v would work.

Brian, have you found some good low-esr caps on Digikey.com? I can't seem to find any, only a few surface mount types, which I don't want, as you prob as well I'm assuming?

[BrianG]

No, I haven't really looked yet. And no, no surface mount caps for me...

[zeropointbug]

Got yah...

I have looked and looked for a supplier good low-esr caps, I have given up. There are those ebay ones, which are 35v/1500uF for $19 for 25 caps (good deal), and the one 25v/2700uF Novak cap, which is enough voltage rating for up to 4s LiPo; the Novak one sells for $10, which all you are paying for is the Novak sticker, maybe they ARE good caps, but thats not the cost...

Here are some good options:

Novak 25v/2700uF low-esr (unknown manufacturer)

http://www.shopatron.com/product/pro....56.7400.0.0.0

Nichicon 35v/1500uF low-esr (better choice IMO)

http://cgi.ebay.com/25-Nichicon-35V-...QQcmdZViewItem

[zeropointbug]

Or, if you want 200 capacitors, there are 330uF caps, like the ones IN the Quark . $26 for 200 caps.

http://cgi.ebay.com/200-Nichicon-35V...QQcmdZViewItem

[BrianG]

Wow, 200! I was thinking of about 20, lol. But yeah, the ones they use in the Quark, just more of them would be nice to use.

[zeropointbug]

HERE is the data sheet on those, they are Nichicon PM series caps. They are the best series that have 35v rated.

The 330uF has 0.06ohm (100khz), and the 1500uF has 0.024ohm (100khz). So five of the 330uF to get roughly the same capacity as the 1500uF will get you 0.012ohm, pretty good.

I am trying to decide between these. :024:

[zeropointbug]

I still want to try those 140 farad Maxwell Ultra capacitors, it appears as thought they have better ESR than these low-ESR caps. Even factoring in the frequency coefficient for ESR

DC ESR: .0072ohm

1KHz ESR: .0036ohm

It looks like to me they would do a better job with such low esr, and very large capacitance available. Plus you would have extremely high power on tap. Batts would only deliver the average current draw. They are rated for 530 amps with a short circuit, and 15,000 watts/kg. $5 a cap, you would need 8 for 4s LiPo.

[suicideneil]

You guys are true electronics buffs, and I salute thee. You got me thinking too- what would be the effect of wiring another capacitor (of the same type it uses) to a HVMaxx esc? I know it only runs at 14.4-16.8v,but the capacitor is rated for 2700µF & 25v; would it be beneficial in terms of what you've been talking about- smoother/better performance & reduced back emf etc?....

[BrianG]

ZPB: OK, but 140F will make one hell of an arc! Current would be theoretically infinite, only limited by the battery capacity and wire gauge. IIRC, last time this came up, I suggested that you permanently wire a cap to each pack to avoid arcing from the initial connection. Just use a resistor on initial charge to force a lower charge. This would definitely let you use much lower C-rate batteries since the cap will take care of just about all current pulses. And you'll get closer to the actual Ah of a battery since it won't be running as hard. However, will the price and weight increase be worth it? I say you try it and see - I'd MUCH rather spend your money ;)

Are you sure those are the ones that Quark uses? If so, I'll probably wire an extra two on the Quark and a couple on the MM.

suicideneil: Running a capacitor at lower voltage is perfectly fine, it will actually extend the life of the internal insulator since it's not running close to its max. The only downside is the slightly bigger physical size. It'll be better for your batteries, but not sure if you'll see a noticeable increase in performance. But it sure won't hurt - just make sure you hook it up right. :)

[zeropointbug]

Did you read the whole thread back? The main reason why you should solder a few more caps on the input power leads of an esc is for reliability, number one, those caps take an electrical pounding, and stand back up for some more... well not always, like what happened to mine.

Having more capacitance/less impedance caps

-They make for more efficient FET switching, prob a cooler running controller, thus more reliable
-Take load off the stock caps in the esc (which they are limited by space)
-May or may not achieve more power, smoothness. depends.

You can go ahead and try using that Novak cap if you want, see how it works. Having more capacitance is like a heatsink for electricity, the more the better.

I wouldn't call myself a BUFF, but thanks! :027:

I understand more physics stuff, not electronics per say.

[zeropointbug]

Originally Posted by BrianG

ZPB: OK, but 140F will make one hell of an arc! Current would be theoretically infinite, only limited by the battery capacity and wire gauge. IIRC, last time this came up, I suggested that you permanently wire a cap to each pack to avoid arcing from the initial connection. Just use a resistor on initial charge to force a lower charge. This would definitely let you use much lower C-rate batteries since the cap will take care of just about all current pulses. And you'll get closer to the actual Ah of a battery since it won't be running as hard. However, will the price and weight increase be worth it? I say you try it and see - I'd MUCH rather spend your money ;)

Are you sure those are the ones that Quark uses? If so, I'll probably wire an extra two on the Quark and a couple on the MM.

suicideneil: Running a capacitor at lower voltage is perfectly fine, it will actually extend the life of the internal insulator since it's not running close to its max. The only downside is the slightly bigger physical size. It'll be better for your batteries, but not sure if you'll see a noticeable increase in performance. But it sure won't hurt - just make sure you hook it up right. :)

Yeah, of course Brian! That wouldn't be an arc, that would be critical mass occuring at contact! :005:

I would have them as part of the pack, but wired with little wire as possible, and shorten all wires between it and the esc. I can also set current levels of 0.01 amps with the power supply I use, as you might know.

About it being worth it, I really do think it would be worth it, not considering weight, just looking at performance, reliability, and much longer cycle life of the battery. I wouldn't doubt it would pay for itself even before the battery is at end of life. I can see it AT LEAST tripling the battery cycle life.

BTW, I made a mistake on price, they are $15 for the 120F U-cap, you need 8. They have a calender life of 10 years, cycle life of 500,000.

It looks as though they have a high energy version and a high power version. The 120F being power, and 140F being energy, the 120F has a lower ESR of 0.0025ohm, the 140F is 0.0035ohm.

I was just thinking, this really sucks. I mean, we are talking about what is arguably the best large scale esc on the market right now. First we have to accept that it won't do 6S as it should; now we are hearing that when used for its intended purpose, we should add caps. Well, why didn't they add them?

As for the Caps in the ebay link. Are those the ones that we need? If so, I'll grab 'em.

Just to be sure, all I'm doing with them is soldering one leg to the + input and one leg to the - input on the esc where the battery wires are soldered to the esc, right?

I've had such crap luck with escs lately, that if this Quark lets me down, I'm liable to start a bon fire with all of my r/c stuff.

Oh, and why not smt caps? Form factor? Performance?

OK, never mind. I found a pic of one and I see why not.