What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

SDS CPI2

rv6ejguy

Well Known Member
We've received a lot of emails and PMs here on the new CPI2 so I thought I'd start a fresh thread on that to provide updates.

After all the talk here on alternator failures, over voltage events and LiFePO4 batteries, we spent over a month investigating the best way to ensure power is always available to keep the CPI supplied with power to fire the plugs. We feel we have a good solution now and have been testing that for a couple of weeks, tweaking as we learn more on the bench.

With that mostly sorted, we began preliminary PCB layout about 10 days ago, laying out the rough placement of key components and connectors. Naturally, there are many intertwined aspects which affect each other so all designs are likely to be altered before the prototype stage.

The plan is to be able to control up to 6 coils so the unit can run 12 cylinder engines if desired via the usual waste spark strategy.

We've tried to eliminate all external switches to make for a cleaner, faster installation.

We've done some preliminary studies on the instrument face layout keypad/ display and how to integrate this with the chassis and optional twin PCBs.

So it's coming along but interrupted somewhat as we take care of our regular EM-5/ CPI sales and tech support. We're lining up some new partners overseas to distribute and support the product in other places with high demand and hope to get some OEM deals in place too.

Unfortunately, the product will not be ready for 2017 release as was initially hoped. The keypad vendors usually take at least 30 days to supply product after the design is sent to them. We're not going to rush the product out. It needs to be fully tested to our satisfaction before release.

If all goes well, we hope to finalize face footprint by mid- November.

Thanks to all those who contributed ideas that we could consider and implement into the design. I hope those who are waiting will be satisfied with the design and new features. Please understand that we can't release too many details on all new features due to the competitive nature of this market.

Thanks for your support and interest in the CPI2. I'll update this space as more significant information becomes available.
 
Too soon to say until the design is more fleshed out. We're a long ways from a prototype and even getting quotes on expensive things like keypads.
 
Ross: I'm a happy customer of your six cylinder CPI system. Are you going for the Allison and Merlin market here?


Lee...
 
Glad you're happy with the present CPI. I suppose the new model could be used on the big V12s. I hadn't considered that. We have some customers converting Jag and BMW V12s for replica fighters who expressed some interest and there is a clean sheet aero V12 being developed down under plus the LS based V12 project.

Not a big market but we'll be ready for growth, potentially being able to run 7 and 9 cylinder radials as well with a dual unit in the future when we have time to write software.
 
Great stuff Ross; terrific to see the innovation and market savvy going on with you guys.
 
We've reduced the size of the OV protection and battery switching circuit. Continuing to lay out more of the components.

Have ordered some of the hardware- fittings, displays etc. to be able to start on the prototype board without delays down the line.

Looking for feedback on the keypad colors and LCD back lighting colors. On the LCD we have yellow, blue and white back lights available.

Things are starting to come together a little faster now as items fall into place.
 
Eifis

My panel is very space limited. I would like it best if it would integrate into the EFIS system.
 
Ross - w.r.t. LCD display backlight colours, take a look at human eye response to colours as we transition from day to night vision.

First colour to go is red, then we lose everything in and around the red wavelength.

Of the choices available, blue is the colour we see best at night, thus would be the colour requiring the deepest dimming capability.

For this reason white backlighting might be the best compromise between ability to see the display and ability to dim it so it's not sticking out like a sore thumb in the cockpit at night.
 
The (apparently poorly framed) joke is to use a tri color (RGB) LED, & software lets the user pick their favorite color.
 
We ordered a bunch of displays today from several vendors to evaluate hot and cold performance, contrast and viewability.

Contemplating processors now with an eye towards having enough timer and analog channels for future expansion.
 
We've found some ways to reduce some pins used on the CPU chip to allow the use of the present 80 pin package in the new unit. The higher pin version took up too much board space.

We've received a couple displays now and some more will come next week. Hopefully I can post some photos here so people can comment on what they prefer.

Component layout is slowly progressing. Much thought is going into the best way to have the keypad communicate with dual boards, the backup battery charging strategy and how best to incorporate some of the new features.
 
Will the current crank trigger work with the new CPI2? The reason I ask is, my condition inspection is coming due in December and I'd like to add the SDS crank trigger sensors and flywheel magnets while I have my CS prop off for factory inspection and maintenance.

Thanks
 
Yesterday we were working on software to allow the single instrument face/ keypad to communicate with dual PCBs. Also working on the clearest control strategy for those functions. Leaning towards some colored micro LEDs to differentiate between primary and secondary PCBs to avoid a button push on the keypad.
 
Will the current crank trigger work with the new CPI2? The reason I ask is, my condition inspection is coming due in December and I'd like to add the SDS crank trigger sensors and flywheel magnets while I have my CS prop off for factory inspection and maintenance.

Thanks

I'm do that right now; the prop's off, and just got the SDS today!
 
Ross - w.r.t. LCD display backlight colours, take a look at human eye response to colours as we transition from day to night vision.

First colour to go is red, then we lose everything in and around the red wavelength.

Of the choices available, blue is the colour we see best at night, thus would be the colour requiring the deepest dimming capability.

For this reason white backlighting might be the best compromise between ability to see the display and ability to dim it so it's not sticking out like a sore thumb in the cockpit at night.

And it's also the most difficult color to focus on at night.
 
It's interesting to note that cars commonly use white, green, red or blue dash lighting these days and do lots of human interface studies yet seem to have no clear concensus here.
 
Last edited:
Lets also keep firmly in mind that this is an ignition display, not a PFD.

Of all the sources of information on our panels, this display should be the last thing you focus on in a critical phase of flight. Thinking back through all the displays I've seen in aircraft panels, there's a whole rainbow of color: TruTrak is blue (was green), Garmin was amber (now white), King was red... As Ross points out, there does not seem to be a consensus.

IMHO, the human factors analysis should be limited to making sure it is not overtly distracting, and call it good.
 
Last edited:
I'm do that right now; the prop's off, and just got the SDS today!

I can attest to Chairman Maus having his SDS and tinkering with it as I saw it yesterday and I'm in the I want one camp for sure.

But, I would really like to see the new controller. Can we get some new controller head pics please.....? Even is just mock up casing? Please?
 
Panel controller should be remote from ignition modules.

I think the best setup would be a very compact panel display remoted from the ignition controller(s). It could parallel adjust two ignition control units without loss of redundancy, and could be designed to take very little panel surface space and almost no behind panel space. Ihope this is being considered.
 
Update on the CPI2. After looking at the space available on the rear of the single piece design for all the connectors required and how to route all those traces aft on a narrow board, we decided to divide the panel display and the main board into 2 separate units. This will allow for more flexible mounting and make our design job much easier. I know many people were in favor of that layout anyway.

At this time, the panel display face is an octagonal with the major dimensions at 2.8 X 2.8 inches. Mounting screws penetrate the short 45 degree corners.

The display is a 12 X 2 LCD and we've done some cold temp and sunlight readability testing with white characters on a blue display which seems quite good overall. We want to do further testing on that one before we go ahead with it though.

We've also designed the backup battery tray which will be prototyped soon.

We're working through various software challenges but the display and surrogate keypad is working at about 98% so far. A few more minor things to solve then we'll be on to a prototype PCB after that. A couple new features will likely soak up a few weeks there before that can be produced and start full testing on the bench and our shop O-200 engine.

Fortunately the FF hardware and actual coil control software is all pretty much the same as before so that won't take any extra time.

There's not much to show so far except for 2 big breadboards and a sub board with a jungle of wires emanating outwards, connecting them. I'll be sure to post photos and videos when we get into the proto and testing stages.
 
Good to know Ross. Thanks. I've been thinking of pinging you guys on this thread, but decided maybe best just to letcha-all work. I doubt I'm not the only one kinda biting their lower lip in anticipation.
 
Ross,

I like the concept of a separate display unit. Is it going to be possible to operate the new CPI2 in normal operation without having the display connected?

For those of us that don't plan to experiment in flight (or minimally experiment) it would be nice if the initial setup/configuration operations could be done with the display on an extender cable and eliminate the need to find a dedicated home for the display on the instrument panel.
 
Ross,

I like the concept of a separate display unit. Is it going to be possible to operate the new CPI2 in normal operation without having the display connected?

For those of us that don't plan to experiment in flight (or minimally experiment) it would be nice if the initial setup/configuration operations could be done with the display on an extender cable and eliminate the need to find a dedicated home for the display on the instrument panel.

At this time, all development has been done with a single DB9 serial cable connecting the panel display to the control board. Same cables we use on the EM-5 ECUs.

We will try to make a completely blind version as well for those who don't want or need the panel display. You would not have any information about backup power switching, faults etc. though.

Welcome any other input in that vein while we are still in the design and initial software stage.
 
Ross,

Can you tell us about your overvoltage protection circuit?

I will be starting my SDS CPI2 firewall forward install during my condition inspection. Looking forward to the release of the CPI2 modules.

Thank you
 
In basic terms, at very high voltages, the OV circuit will pop a device mounted fuse from main bus power. The device will have already switched over automatically to the SDS backup battery at around 15.5 to 16V, wherever we decide that threshold should be to avoid nuisance trips.
 
Work last week revolved around getting board 1 and board 2 to query each other and communicate with the keypad properly. Lots of software work on that and some new hardware too.

We should have 2 simulation boards talking to the programmer board early this week if things go well.
 
At this time, all development has been done with a single DB9 serial cable connecting the panel display to the control board. Same cables we use on the EM-5 ECUs.

We will try to make a completely blind version as well for those who don't want or need the panel display. You would not have any information about backup power switching, faults etc. though.

Welcome any other input in that vein while we are still in the design and initial software stage.

Probably late to comment on this, but I like this idea very much - blind unit that you can panel mount the control head if youd like.....OR velcro mount it during initial stages, then disconnect and stow it in the aircraft kit box for later diagnostics when/if you get some funnyness...or wanna fool with the timing again.
 
Probably late to comment on this, but I like this idea very much - blind unit that you can panel mount the control head if youd like.....OR velcro mount it during initial stages, then disconnect and stow it in the aircraft kit box for later diagnostics when/if you get some funnyness...or wanna fool with the timing again.

Yes, you'll have the option to plug in the keypad when the module is blind mounted to access the modules for programming or diagnostics. A number of people, especially with tandem seat RVs simply don't have the panel space so it seemed this was a win-win situation for both groups and it is easier for us to design as well
 
Yes, you'll have the option to plug in the keypad when the module is blind mounted to access the modules for programming or diagnostics. A number of people, especially with tandem seat RVs simply don't have the panel space so it seemed this was a win-win situation for both groups and it is easier for us to design as well

Maybe a small suggestion.....perhaps dont have the control head connector come out dead center of the back of that unit. Rather out the back but on the bottom edge would be best I'd think (again thinking a velcro install here vs panel mount). I dont think out a side would work for panel mount tho, so I hesitate to suggest that......

Anyway, some thought on the location of the connector wrt mounting options should be considered if possible.
 
Maybe a small suggestion.....perhaps dont have the control head connector come out dead center of the back of that unit. Rather out the back but on the bottom edge would be best I'd think (again thinking a velcro install here vs panel mount). I dont think out a side would work for panel mount tho, so I hesitate to suggest that......

Anyway, some thought on the location of the connector wrt mounting options should be considered if possible.

The connector will exit the back of the panel display but it hasn't been placed on the board in CAD yet. I think your idea is good- it would allow some more options to mount it on the lower edge of the panel temporarily or otherwise. We'll try to incorporate that. Appreciate the feedback.
 
I've been asked if we'll still sell the original CPI after the CPI2 comes out. It looks like we will. A number of people plan to blind mount and just want a safe, canned advance curve. The present CPI will do that fine, with less space taken up and less cost.

If you need to take up minimal panel space and want to have in-flight programmable timing control and diagnostics plus auto backup power capability, probably best to wait for the CPI2 if you can.

Design of the first proto board has begun but I expect a second one to be required to complete full testing before we go into production. There's a bunch of new stuff happening and it's taken longer than expected to make the 2 boards play together.

The breadboard was running a coil and most of the switching a diagnostic software is working well. Still have to write software for the alert screens but that usually goes quickly.

I have a hard time seeing how production CPI2s will be ready for sale before March.
 
CPI2 Battery ?

Ross, it might be helpful to explain what the back-up battery does for you if you already have two batteries as usual for for an E dependant engine.
 
Ross, it might be helpful to explain what the back-up battery does for you if you already have two batteries as usual for for an E dependant engine.

With all the debate about alternator failures and lithium batteries we decided to do our own thing to keep the ignition running no matter what happens to the aircraft electrical systems. So if you have an OV issue, we protect the CPI2 from frying and switch to our own, small AGM battery. Same scenario with low bus voltage.

It seems everyone wants to do something different on their electrical system so we felt many people would want an assured solution to keep the engine running no matter what.
 
With all the debate about alternator failures and lithium batteries we decided to do our own thing to keep the ignition running no matter what happens to the aircraft electrical systems. So if you have an OV issue, we protect the CPI2 from frying and switch to our own, small AGM battery. Same scenario with low bus voltage.

It seems everyone wants to do something different on their electrical system so we felt many people would want an assured solution to keep the engine running no matter what.

....you mean the back up batt as a purchase "option"........? Also, I believe I rem you saying long ago that a 9v battery would power the system (or the control head?).....is the AGM to power the entire suite, control head and coils? How big is small, and where is said battery to reside......?
 
Ross,
You mentioned a "canned" timing curve, what would this canned curve be for a O-360 with standard 8.5 compression?
What benefits/detriments would this canned curve have for ROP high power, LOP below 65% power and auto fuel?
Thanks!
 
....you mean the back up batt as a purchase "option"........? Also, I believe I rem you saying long ago that a 9v battery would power the system (or the control head?).....is the AGM to power the entire suite, control head and coils? How big is small, and where is said battery to reside......?

A 9V battery can run the module to test it but not the coils.

At this time, we intend to offer the backup battery as an option. Some may feel they don't need triple redundancy and just want to power the CPI2 backup wires with their own backup battery.

Yes, the AGM is 2.9 amp hr. and will run the module and both coils on a 4 cylinder for about 45 minutes, one coil for about 1.5 hours. This is just to get you back on the ground safely to sort things out.

We've designed a bracket/tray to hold the battery which can bolt into 2 planes- say a near the floor and on an adjacent bulkhead at 90 degrees. Could possibly be mounted on an upper front panel bulkhead with some suitable doublers. Total weight of the tray and battery is around 3 pounds.
 
Ross,
You mentioned a "canned" timing curve, what would this canned curve be for a O-360 with standard 8.5 compression?
What benefits/detriments would this canned curve have for ROP high power, LOP below 65% power and auto fuel?
Thanks!

I would say little to no power gain over mags running ROP.

For auto gas we need to be conservative, limiting max timing to about 22 degrees above 25 inches. Would probably not run over 28-29 degrees on auto gas even at very low MAP LOP.
 
I would say little to no power gain over mags running ROP.

For auto gas we need to be conservative, limiting max timing to about 22 degrees above 25 inches. Would probably not run over 28-29 degrees on auto gas even at very low MAP LOP.

Dont sell yourself short Ross. After the flight testing I've done I have no problem with saying the "benefit" of your ignition is "increased detonation margin". Regardless of fuel used, fixed magneto timing is a compromise fighting the vastly different ignition requirements when high and lean vs. low and fat. By defaulting to the middle ground, magnetos are inadequate when LOP and too aggressive when at 100%.

That said, backing off timing at takeoff power results in lower CHT and more detonation margin without a loss of power, regardless of the fuel used. Said another way, the 100LL is buying you chemical detonation margin - in response to a magneto lighting the fire too early.

The ability for your ignition to pull timing out at high MP is not being "conservative"... It's simply a better way to operate the engine.
 
Last edited:
Back at work today. We were able to make some decent progress on the panel portion circuit board over the holidays. It's about 90% done now except for a second power plane. Proto board design will go out this week for manufacture and we should have that powered up before Jan. 7th hopefully.

Nice to see things coming together more quickly now as more tasks are scratched off the list.

We unfortunately were not able to move the DB connector down very far on the board without a lot more work due to other more pressing design issues. That's the way these things go sometimes, especially when there are time constraints involved.
 
We unfortunately were not able to move the DB connector down very far on the board without a lot more work due to other more pressing design issues. That's the way these things go sometimes, especially when there are time constraints involved.

........bummer........
 
The proto board will be done tomorrow. Unfortunately there was a mix up with file transfer which delayed us a couple days. Hopefully loaded by the weekend and running.

A lot of stuff packed into a 2.3 X 2.3 inch PCB. It was challenging.

We're excited to see it run on the bench soon.
 
Any new information on the backup/aux battery option(s)?

Will there be an audio and/or visual low/high voltage alert?
 
We'll provide the optional backup battery tray which will fit a Powersonic PS-1227 AGM battery and ATO fuse holder. The battery can be easily purchased online from various sources.

Yes, visual alert via LED and and screen message. Considering an aural alert as well.
 
Back
Top