Puma board for FreeEMS

[TonyS]

XOR is on the top drawing, and the 280 ohm is Fred's testing. I don't think it's on the schematic at this point.

Could Preston's problem be caused by using a 74"LS"86 as shown on the schematic, instead of the 74"HC"86 as shown on the BOM? If he used the "LS" version, the logic output High voltage may be too low to trigger the ignitors.
-Huff

[nitrousnrg]

XOR is on the top drawing, and the 280 ohm is Fred's testing. I don't think it's on the schematic at this point.

Could Preston's problem be caused by using a 74"LS"86 as shown on the schematic, instead of the 74"HC"86 as shown on the BOM? If he used the "LS" version, the logic output High voltage may be too low to trigger the ignitors.

Damn, the schematic shows "LS", which is wrong. That field got translated to the BOM automagically. fyi, Fred and me have the HC version.

280ohm was 330ohm, not sure when we changed that. The schematic shows 330ohm in Spin2 branch, Spin1 has a 1k or 1.6k resistor in the protection circuit.

http://i260.photobucket.com/albums/ii15 ... 0_8641.jpg
Note how the legs are bent and destressed?

I saw stress simulations on a device installed exactly like that and didn't like it much, although is better than a normally installed TO220. I recall that pic is from an OEM EMS... Linds' car maybe. Anyhow, injector drivers are the only TO220 devices left (by default), and that could change. The default population should be without P&H.

[Fred]

+1 for default without P&H
+1 for HC on my Puma
280 = 330 + 1k in parallel ;-)
330 ain't good enough unless its some logic level input (which I thought these ignitors are, but they aren't...).
Yes, the pic is from Lindsay's old Toyota, one of the most reliable vehicles you can buy at any price ;-)
ca7, I was going to ask if I could have your babies, but they deliver max of 35mA and max of 70mA for VCC or GND, so it's still not good enough, unfortunately, as that is what I have on this ms box which needs to run tonight that I built and am now not happy with...

For now, in lieu of a functional 12v solution, i'm running with two 2n5401 transistors (on the ms2, 4 on the puma soon) and 2 10ohm 1/4 current limit resistors, this should give close to 35mA continuous from both channels without risking damage. I've tested this on my hyundai in single channel mode and it works OK (no change). The output hovers around 300mV when off and solid 5v when on.

Jared has a good solution, or one that appears to be. Depleted mode N channel mosfet with logic level drive and strong pull up. when not driven, it would keep the output at zero v and with 5v on it, it would allow the pull up to provide the base current to the ignitor. Conversely, for someone that needs a low side drive, the same spot could probably house a normal logic level FET or IGBT ign driver such as bp273 or whatever it is. If the pinouts are chosen well, it could be pretty universal. The XOR has been gold today, as I inverted stuff back and forward, just swapped resistor and it definitely worked correctly, no changing settings/code... yay.

Fred.

[nitrousnrg]

Ok, I've 2 news:

* I just found a big, nice and compatible darlington (FJB102) for the P&C driver, and it comes in SMD version. In my mind, it goes in the upper side of the board, and the DIP8 goes in the bottom side, because its optional and TH. The same footprint can be used for a regular MOSFET with logic input in the default setup.
In 15 days I'm setting up an order of components for testing (yes! I'm gonna test stuff this time!). Critical circuits to test are the switching supply and this darlington. Have to order some 5v references for Vref input too. I'm going to try the P&H in both linear and switching mode to check how noisy it is (the board has different grounds, so I've some expectations).

* Tomorrow I've a meeting with the president of a brand new free economic zone, near my city. Inside the zone, I can import and export without taxes, and its intended for electronics production, hence they have (or will have, its really new) pick&place machinery, tools for PCB fabrication, and things like that. Many things depend on the next hours.

That way, my boards can be assembled here saving $ for the customers (no taxes, no extra shippings around the world), and helping a bit the very fucked up Argentinian economy:-) Maybe this effort could generate interest and motivation amongst designers and engineers, it would be really nice to see argy industries waking up as it used to be many years ago.

About the depletion mode mosfet, if it works, its in Spin2. I need some help to find those fets, couldn't find the way in newark and digikey.

Also, the Puma website could be benefited from Fred's idea about github pulls. We're going to test that during the week.

[jharvey]

I searched octoparts like this

http://octopart.com/partsearch/#search/ ... tion%20fet

Then I set the filter to a + voltage for Vgs, and got this guy

http://datasheet.octopart.com/LND150N3- ... -20983.pdf

There seems to be several depletion mode mosfets, and many specify N channel, however, they appear to require a - voltage at the gate to activate. Which is the opposite of what I expect. Perhaps I'm reading the datasheet's wrong, and a - voltage is really a + to the MCU pin. I know there are exceptions to the rule of thumb about enhanced mode being Normally Open and depletion being Normally Closed, and N channel being a + polarity on the gate, with P channel being a - voltage on the gate. However, it seems like they all are exceptions, so I'm a bit confused really. Might be best to search MFG sites, then use part numbers to search for vendors.

[jharvey]

That datasheet indicates the MOSFET is n channel depletion, and the wave forms shown indicate it works as we want. However it has a very small current capability.

From http://www.electronics-tutorials.ws/tra ... ran_6.html we see the exact opposite of what's noted on supertec's datasheet. It specifies the activation polarity as being opposite of we want. I think this may simply be a terminology issue, and I think that any n channel depletion will be OK. Probably a good one to include in Marco's testing purchase. I'll see if I can find a couple MOSFET's even if the datasheet seems to indicate the polarity is backwards from the expected.

[nitrousnrg]

* Tomorrow I've a meeting with the president of a brand new free economic zone, near my city. Inside the zone, I can import and export without taxes, and its intended for electronics production, hence they have (or will have, its really new) pick&place machinery, tools for PCB fabrication, and things like that. Many things depend on the next hours.

This ^ went great, Pumas can be built there, and every thing manufactured in those facilities is tested to comply with international certifications. I will use the place for other projects if I can.
Plus, an University in La Plata is ready to test the boards under MIL-STD 810G, and the quotation was reasonable, so if Spin2/3 are tough enough (Ha! I made rhyme) we would have military stuff in our cars.

There seems to be several depletion mode mosfets, and many specify N channel, however, they appear to require a - voltage at the gate to activate

Ummm, a depletion mode mosfet-N needs a negative Vgs to to cut the current thorugh it, and with 0v it allows the current to flow. I wasn't aware about exceptions

This figure is consistent with my old class notes
http://www.esacademic.com/pictures/eswi ... el.svg.png
I've a better pic (with actual values on its axis!) but is the same, and ^ was online somewhere.

Do you want a mosfet that allows current to flow at Vgs=0v and cuts it at Vgs >0 (p-mosfet depletion, although Vds<0 and Id<0),
or one that flows no current at 0v and makes it flow at Vgs > 0 (n-mosfet enrichment)?

A draft of the intended circuit could help me understand the desired behavior :-)

[jharvey]

From what I'm reading, I agree, it appears we are looking for a P channel, even though my understanding was that we wanted an N channel. However, that 150N3 clearly shows the wave forms as different, so I question if I'm simply flipping terminology. Here's my pencil sketch of the circuit layout.



I guess the polarity flips when you change enhanced vs depleted. So my above sketch has errors, but does show the basic layout. I'll have to learn a bit more I guess, I haven't use a depletion yet, just read about them, so gut feel and experience is limited there yet for me. I understand N channel is typically low side, which we are, and P channel is typically high side. Also N channel is really common, but P channel is next to impossible to find. So I have lots of things that are telling me it should be N channel, but the data sheets typically tell me different. Except that 150N3... This shows a chart that claims the polarity flips with channel type.

http://www.electronics-tutorials.ws/tra ... ran_8.html

[jbelanger]

We aren't modifying the bootloader because if we do, we're not compatible with the TA card anymore, and people who don't want or need SMD setups are fucked. And that is totally unacceptable to me, and to many others too. That's why, and it's not much of a restriction, really. You'd still have to do it anyway, as during resets some of those pins are in high impedance state too...

Bullshit.

The Puma board doesn't use the TA card and the bootloader has no impact on the code. So if changing the bootloader and specifying some wiring that allows you not to have oodles of useless components then that's a much better choice. The new bootloader can be 100% compatible with the PC side of the loader and still initialize some pins to a safe state. And the transition during the reset can be handled by some control pin that does have the hardware protection.

Fred putting such unnecessary limitations on this is not only shortsighted it can also prevent some people from using FreeEMS as a basis for their own project. I know it would in my case. For example, I'd want to be able to have a bootloader that uses CAN instead of RS232 or USB.

Jean

[nitrousnrg]

Good,
What I don't like is the big current being drained to ground when the igniter is not active.

Whats wrong with this one?:

It can flow plenty mAmperes to drive the igniter. Pullup/down should be in the MCU-side, before the XOR.

And Jean, I'm with you, and I just realized that Puma doesn't have a reset button.