Received Puma boards / My Puma project!

Marcos' unmaintained, but still in-use, Puma for FreeEMS circuit board/hardware design!
Post Reply
User avatar
KW1252
LQFP112 - Up with the play
Posts: 166
Joined: Tue Jan 15, 2008 5:31 pm

Received Puma boards / My Puma project!

Post by KW1252 »

Got a nice little post package today, my Puma boards arrived! :mrgreen:

The boards are really impressive, though few ideas did come to mind, don't know how much of these been discussed elsewhere (I haven't been as active on the forum as I should've been :oops: )

Injector pins are packed really tight. There's a bit of estate to spread out the LM1949s and give a little more space to the headers, this would allow use of meatier connectors; peak phase reaches 4A which might deal some punishment on regular pins and it would be nice to have latching housings on the pins. The stepper controller suffers a similar condition; it would benefit greatly from using a polarized latching connector.

There are few SMD components really close to the edges, I wonder if they'll touch the rails easily. The mounting holes of course eliminate the need, but it'd be a nice choice.

Fuses and USB connector have been discussed about; for the USB I'll make an adapter board to Mini-B connector, and the fuses are mounted on a chassis holder. For production spin, the fuses could be SMD or through-hole miniature types? (there are holders for both types if soldering a fuse doesn't sound too great an idea.) Also, I'm guessing fiducials are to be used in later spins?

I guess I'll spend a few days hunting for components, already looking forward to soldering the beast together :)
User avatar
nitrousnrg
LQFP144 - On Top Of The Game
Posts: 468
Joined: Tue Jun 24, 2008 5:31 pm

Re: Received Puma boards / My Puma project!

Post by nitrousnrg »

So happy to hear that!!
I'm starting to get worried about some boards not reaching their owners. gearhead, was the second, and I still don't have news from him.

About the connector: there are bigger things that those. There is an extra (unneeded) pin for almost every high power output, and more ground pins are needed too. A big change in the connectors is coming. I was thinking on a specific connector for the injectors, but couldn't buy it, and the mfg deadline was getting close, so it eneded that way.
There are few SMD components really close to the edges, I wonder if they'll touch the rails easily.
Oh, R67 and R72 are really close to the edge. Moving them right now, thanks thanks.

Fuses were removed for spin2, now there is a lot space to put interesting stuff there.

Remeber to cut the MCU traces! And ask, ask. My plan is to go hard on documentation next week, since I won't have the car available for testing.

Have fun, Mr moderator.
Marcos
User avatar
jharvey
1N4001 - Signed up
Posts: 1607
Joined: Tue Jun 10, 2008 5:17 pm

Re: Received Puma boards / My Puma project!

Post by jharvey »

KW1252 wrote:I haven't been as active on the forum as I should've been :oops:
True true, you typically offer good content, and have been lacking :)
KW1252 wrote:Fuses and USB connector have been discussed about;
Perhaps you can make your adapter available to others as well? I suspect others will be interested in such an adapter. Do you plan a self made, or a mfg'ed PCB? Have you seen the above post that notes the thru hole/SMT connector? Also in KICAD, there's a foot print that can handle both the more rugged thru hole mini usb connector, and the SMT mount style. If you use KICAD to make your adapter board, that footprint is already made.

The fuse theory was a bit buggered. The idea on DFH was that the fuse was bridged by design. The theory was that it would include a trace between the pads, such that if you did short it, this trace was a bit thinner than the rest of the PCB trace. For example, a 1amp circuit, would be designed to handle 2amps, but the trace between the fuse holders would be designed for 1.5amps. Then if you shorted the board, the weak point would be in a known location, and you could fix it with by adding a fuse. Marcos didn't happen to catch that in spin1, so it's just a normal fuse holder, which seem to be out of place, as fuses should be up stream. It's being removed from future spins as it really should be fused at the source, and leaving this kind of protection on the PCB encourages bad practice.

About the PCB lead times, I've had issues with receiving international mail on other projects. I once ordered some SSR's from futurlec. They claim a 2 week lead time, but it took over 8 weeks for them to arrive. They shipped them that day, but they sat in customs for a long time. If gearhead is in the states, it might be a while before it makes it.

Good feed back, thanks for helping make it better.
User avatar
KW1252
LQFP112 - Up with the play
Posts: 166
Joined: Tue Jan 15, 2008 5:31 pm

Re: Received Puma boards / My Puma project!

Post by KW1252 »

nitrousnrg wrote:So happy to hear that!!
I'm starting to get worried about some boards not reaching their owners. gearhead, was the second, and I still don't have news from him.
Snail mail can be a fickle thing... This was actually quite speedy delivery compared to some cases :)
About the connector: there are bigger things that those. There is an extra (unneeded) pin for almost every high power output, and more ground pins are needed too. A big change in the connectors is coming. I was thinking on a specific connector for the injectors, but couldn't buy it, and the mfg deadline was getting close, so it eneded that way.
Okay, understandable :) Such changes in a design as extensive as this take their time, better to stick with the existing arrangement if there's no time to properly audit the changes. Hoping to see latching polarized connectors in the future ;)
There are few SMD components really close to the edges, I wonder if they'll touch the rails easily.
Oh, R67 and R72 are really close to the edge. Moving them right now, thanks thanks.

Fuses were removed for spin2, now there is a lot space to put interesting stuff there.
This was great news :)
Remeber to cut the MCU traces! And ask, ask. My plan is to go hard on documentation next week, since I won't have the car available for testing.

Have fun, Mr moderator.
Damn, that piece of information had escaped me! I'll be sure to look into it. It seems there is no easy way to disable the internal MAP sensor except by not installing it, I'm guessing the next spin will have a jumper for selection of internal/external sensor?

Building the BOM for the electronics supplier I noticed there's a substantial difference in MPX4100 packages, turns out the MPX4100A is a free-standing unit, hardly wider than the footprint and without the hose nipple. And it also costs half the MPX4100AP with mounting tabs does! The enclosure needs venting though, and preferably to some static pressure spot too. I was thinking a hose from the enclosure to the space between the door and door jamb, it's in pretty constant pressure regardless of speed. A plastic fuel filter should keep the contaminants out well enough.
jharvey wrote: Perhaps you can make your adapter available to others as well? I suspect others will be interested in such an adapter. Do you plan a self made, or a mfg'ed PCB? Have you seen the above post that notes the thru hole/SMT connector? Also in KICAD, there's a foot print that can handle both the more rugged thru hole mini usb connector, and the SMT mount style. If you use KICAD to make your adapter board, that footprint is already made.
I was thinking about real low-tech approach to this but if there's a symbol ready... I'll have to look into it of course!

As for the connection to the enclosure connector, I first thought about doing a similar sandwiched solution the version with prototype board uses. However, upon further consideration, I'm going to opt for ribbon cable. First, I'm going to need cables anyway with the various connectors on the board. Also I'm probably going to mount the board on rubber pads, so mechanical insulation from the connector would be necessary. Finally, upon discovering the Cinch enclosure I'm really interested in using it, and aligning the boards might be difficult; so, I'll choose ribbon cable and latching connectors.
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Received Puma boards / My Puma project!

Post by Fred »

Re the connectors, realise that they are supposed to go to another board not to a wire or out to a device. IE, the other board will have those pins firmly plugged and soldered (and keyed by design) and the other board will provide connection to the world.

Re the fuses, say goodbye.

Re the adapter usb board, good idea.

Re AAP sensing, constant pressure is one idea, but what you actually want is the pressure where the exahaust exits the car, map takes care of input pressure, and after that you might need yet another for pressure where the turbo draws its air to get its efficiency and properly calibrate things, but this is all another story, the main thing is that the AAP NEEDS a barb on it to be properly useful so it can be referenced to the best possible place for that person's application.

And, what marcos said re the injector pins, ALL of those 12v out and batt v out pins are useless and will be gone next time round, leave them unconnected/unpopulated and you'll have lots of room :-)

Welcome back to the board, slacker, I hope we'll hear more of you this time, also, don't lose this board like you did with your TA :-p ;-) :-)

Fred.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
User avatar
KW1252
LQFP112 - Up with the play
Posts: 166
Joined: Tue Jan 15, 2008 5:31 pm

Re: Received Puma boards / My Puma project!

Post by KW1252 »

Ah, true. The +12v gets shafted obviously. I was a bit distracted about them, but yeah, it's just the battery voltage doing extra miles :) I was thinking about a construction where the main board and connector board are separate and connected by wire harness (which would be very flexible for future revisions)

Actually, when I did a bit of research why manufacturers do use AAP, it turned out it was to keep turbo pressure ratio in check. Makes a lot of sense, the wastegate operates at (more or less) absolute pressure, and if the turbo tried to keep up with it at high up, lots of little turbo pieces would result... As such, I wouldn't need it at all right now, but a trip to the Alps someday..? Who knows :)

Fuel pump control is low side controlled, but I think I'll convert it to high-side (via a relay.) This is because if there's wire damage (wear or accident), it'll be more likely to shut down the engine and systems rather than keep them powered uncontrollably. Not a huge issue since they are fed from ignition key controlled switched source anyway, but I like keeping it on the safe side.
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Received Puma boards / My Puma project!

Post by Fred »

The fuel pump FET is designed to drive a/many relay(s) in a massively overkilled/robust way. It's not designed to drive a fuel pump directly, however on a low power setup with a small fuel pump, it could do that, depending upon the FET chosen.

AAP is useful without boost. The low exhaust back pressure affects the VE curve in ways that aren't consistent with just MAP control. If you live in Holland, don't bother with AAP, if you live in NZ, do, etc.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
User avatar
KW1252
LQFP112 - Up with the play
Posts: 166
Joined: Tue Jan 15, 2008 5:31 pm

Re: Received Puma boards / My Puma project!

Post by KW1252 »

Fred wrote:The fuel pump FET is designed to drive a/many relay(s) in a massively overkilled/robust way. It's not designed to drive a fuel pump directly, however on a low power setup with a small fuel pump, it could do that, depending upon the FET chosen.
True, I wasn't thinking as much driving the actual motor low-side, just the way the relay is operated. I meant I was going to use a relay (a well-behaving PCB or solid state DC relay) within the enclosure to skip any complicated circuit changes and problems with high-side FET control. I'm going to control also igniter and injector voltage feed with it, and I'm thinking daytime running lights too... but I'll have the signal on the ECU connector as high side.
AAP is useful without boost. The low exhaust back pressure affects the VE curve in ways that aren't consistent with just MAP control. If you live in Holland, don't bother with AAP, if you live in NZ, do, etc.
Makes sense, but any variable exhaust systems, eg. valve timing or muffler bypass (handy when the car is regularly inspected for street legality,) mess that up good. Exhaust pressure sensor would be more usable. Just use long metal tubes and/or some old heater core in the pressure line to cool it :)

I'm still looking for the footprint of the mini-USB for KiCAD, I don't seem to have it in my libraries. Perhaps I need to update them.
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Received Puma boards / My Puma project!

Post by Fred »

Bump for activity and update!
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
Post Reply