Printed DNA30 Bottom Feeder

[ThreeDJ16]

Diodes have a forward voltage drop. Silicon diodes around 0.7V. Gallium arsenide around 0.3V. This will totally screw up the dna's battery monitoring. When you insert a freshly charged battery at 4.2V, the dna will only see 3.9V. Oh, and at 10A, your diode will be dissipating 3W. If you really want to add reverse battery protection, a diode isn't the way to do it; a P-FET is, one with really low on-resistance (like under 5 milli-ohms).

I swore to myself I wasn't going to post again on this subject.......aw god the devil made me do it.

You state forward voltage drop right? That means when the little arrow on my diode is pointing the other direction. In the circuit I drew, it is reverse of normal battery polarity. So that means no voltage drop and why there would not be 10amps flowing in reverse flow against the diode? It's purpose is to stop current flow in this circuit, not be conducive to it. So there would not be a 3 watt loss. That would only be if it was in series, and then it would totally screw up the dna monitoring system.

OK, devil isn't going to tempt me again.

 

[Rossum]

I think everyone is missing the point that the diode would be reverse biased across the DNA input not in series. So no V drop no power loss when the battery is in correctly.

If that's what you were intending, I did indeed miss it -- sorry.

If you do it that way, when someone reverse connects a battery, the diode becomes what's effectively a short across the battery. That means you better have a fuse as well. Now you're adding two components.

What's wrong with just a P-FET again?

 

[gdeal]

What's wrong with just a P-FET again?

Something like this...

http://www.infineon.com/dgdl/IPD90P...fileId=db3a30431ddc9372011e07e8373a27c4&ack=t

(HT Rossum...)

 

[ThreeDJ16]

If that's what you were intending, I did indeed miss it -- sorry.

If you do it that way, when someone reverse connects a battery, the diode becomes what's effectively a short across the battery. That means you better have a fuse as well. Now you're adding two components.

What's wrong with just a P-FET again?

While i realize my drawing was crude, having two components on it, i thought people could understand that was what i intended before they commented. And since a mosfet does not replace a fuse, that is only one component. An a diode has simple axial leads that does not need an additional breadboard, nor is it placed directory in line with main current path, effectively dropping your power loss to zero.

 

[Rossum]

While i realize my drawing was crude, having two components on it, i thought people could understand that was what i intended before they commented.

Perhaps I missed it somewhere, but I don't recall seeing a drawing you posted. Did you post it in this thread, or was it somewhere else?

And since a mosfet does not replace a fuse, that is only one component.

Nope, a FET doesn't directly replace a fuse. OTOH, I'm not aware of any commercial DNA mods that use fuses, so I'm not sure why a fuse is necessary. It seems to me (I haven't personally tested it) that a FET could offer reverse battery protection with a single component at about the same losses as a fuse. A diode installed in the manner described does indeed not produce any losses (the reverse bias leakage current should be negligible) but the in-line fuse you need to protect the battery and the diode will have a resistance comparable to a FET's ON-resistance.

 

[ThreeDJ16]

Perhaps I missed it somewhere, but I don't recall seeing a drawing you posted. Did you post it in this thread, or was it somewhere else?


Nope, a FET doesn't directly replace a fuse. OTOH, I'm not aware of any commercial DNA mods that use fuses, so I'm not sure why a fuse is necessary. It seems to me (I haven't personally tested it) that a FET could offer reverse battery protection with a single component at about the same losses as a fuse. A diode installed in the manner described does indeed not produce any losses (the reverse bias leakage current should be negligible) but the in-line fuse you need to protect the battery and the diode will have a resistance comparable to a FET's ON-resistance.

Correct! Nothing is required, basically any of this stuff is simply stupid proofing your mod (or drunk moment proofing). Perfectly fine with using a mofet (as it is a well established method all over the internet) or any of a number of different components. It was simply an alternative approach. But the package on the particular P-Fet I have seen folks using does need a mount. And this approach doesn't. Electronics are like many other things in life with many different variables, with many different solutions. Just thought about what Mamu has accomplished using two fuses with two batteries. Her approach creates a high current in reverse that clamps a fuse. Thought it was simple elegance! So wondered about a way to mimic that approach in a single battery mod. Decided it was an interesting theory, and somebody (who shall remain nameless) talked me into posting it.

 

[Aal_]

are you suggesting something like this? just understanding:

 

[ThreeDJ16]

are you suggesting something like this? just understanding:

Almost....your polarity in the drawing is backwards from what i suggested. Installed like that, it would be a dead short with battery installed correctly. Guess it is easier to just post drawing in this thread too...... oh yeah, ressettable fuse doesn't blow, just goes to a very high resistance state. I have tested and they take a while to reset. You will think, they are FUBAR. LoL

And yes, crappy drawing done at like 3am yesterday. Too lazy to redraw........haha

 

[Aal_]

I took this image off google images, i think they are showing the reverse case lol.

 

[Aal_]

Almost....your polarity in the drawing is backwards from what i suggested. Installed like that, it would be a dead short with battery installed correctly. Guess it is easier to just post drawing in this thread too...... oh yeah, ressettable fuse doesn't blow, just goes to a very high resistance state. I have tested and they take a while to reset. You will think, they are FUBAR. LoL

And yes, crappy drawing done at like 3am yesterday. Too lazy to redraw........haha

you're assuming resistance of diode will be far less than resistance of a DNA plugged in backwards? How do you test how much current goes through the DNA in this circuit. I know Mamu did it in practice but I wouldn't risk it.

ADD: with increasing resistance at the fuse, wouldn't current increase at the DNA?

 

[ThreeDJ16]

I took this image off google images, i think they are showing the reverse case lol.

Hah, then just as i said, it is shorted. So cool, where is the web page? I would love to read if it really works and not have to test one of my expensive dna30s not knowing. .......lol.

 

[ThreeDJ16]

you're assuming resistance of diode will be far less than resistance of a DNA plugged in backwards? How do you test how much current goes through the DNA in this circuit. I know Mamu did it in practice but I wouldn't risk it.

ADD: with increasing resistance at the fuse, wouldn't current increase at the DNA?

No, the high resistance fuse acts like a reverse biased diode and stops current flow. Undo the fault and wait a long time, the resistance drops again.

Also, the forward biased resistance of the diode is a known value specified in the datasheet. Since it will basically pose no opposition to current flow, it will allow a rush of current instantly which willl clamp the fuse.

Unless i misread, pretty sure she did try a dna20 with her circuit.

 

[Aal_]

Sorry stupid mistake i thought the fuse is in parallel with the DNA.

No, the high resistance fuse acts like a reverse biased diode and stops current flow. Undo the fault and wait a long time, the resistance drops again.

Also, the forward biased resistance of the diode is a known value specified in the datasheet. Since it will basically pose no opposition to current flow, it will allow a rush of current instantly which willl clamp the fuse.

Unless i misread, pretty sure she did try a dna20 with her circuit.

 

[Aal_]

Hah, then just as i said, it is shorted. So cool, where is the web page? I would love to read if it really works and not have to test one of my expensive dna30s not knowing. .......lol.

The pic is from here.

Circuit Protection - Electrical Engineering Stack Exchange

 

[ThreeDJ16]

The pic is from here.

Circuit Protection - Electrical Engineering Stack Exchange

Awesome!!! Nice to know I wasn't crazy. One guy there stated it was called a crowbar...lol. Guess because if you used fuses that need replacing, it is like hitting them witb a crowbar...haha.

 

[mikepetro]

Awesome!!! Nice to know I wasn't crazy. One guy there stated it was called a crowbar...lol. Guess because if you used fuses that need replacing, it is like hitting them witb a crowbar...haha.

You are crazy, electrical considerations have nothing to do with it.....

 

[Rossum]

And yes, crappy drawing done at like 3am yesterday. Too lazy to redraw........haha

I'm confident this will work -- assuming the diode is robust enough to ensure that the fuse "blows" before the diode does (easy to test w/o the DNA in place, thus posing no risk to the DNA). The DNA should never see more reverse voltage than the forward voltage drop across the diode, which should protect it adequately.

But it's still two components instead of one, and it's really no easier to wire than a FET. I also don't understand why you think the FET needs to be mounted on a board. There's no reason why one can't solder 3 wires directly to it.

 

[kylemd]

Gdeal have you considered redesigning this mod to fit 22mm RTAs ala VTR or Kato Square?

Sent from my Nexus 5 using Tapatalk

 

[Kataphraktos]

Correct! Nothing is required, basically any of this stuff is simply stupid proofing your mod (or drunk moment proofing). Perfectly fine with using a mofet (as it is a well established method all over the internet) or any of a number of different components. It was simply an alternative approach. But the package on the particular P-Fet I have seen folks using does need a mount. And this approach doesn't. Electronics are like many other things in life with many different variables, with many different solutions. Just thought about what Mamu has accomplished using two fuses with two batteries. Her approach creates a high current in reverse that clamps a fuse. Thought it was simple elegance! So wondered about a way to mimic that approach in a single battery mod. Decided it was an interesting theory, and somebody (who shall remain nameless) talked me into posting it.

Thank you for protecting my anonymity.

 

[gdeal]

Gdeal have you considered redesigning this mod to fit 22mm RTAs ala VTR or Kato Square?

Sent from my Nexus 5 using Tapatalk

Not sure how it would work with a tank. Have you seen jonniemac's mod?