Another Engine Control and Regulator project - DC Alternator

[thomasonw]

For the easily technically confused, you might want to change your typo of "booth" to "boost".  I understood.  

Bruce, Thanks - and glad to see you doing better! 

I tried to go back and edit my typo you noted, but it seems the system only allows me to make changes for so long after a post is made..  (I can no longer see the 'Edit' button for that post.....)  Is there some way to go back and edit older posts?

Been busy drafting out the the code for the stand-alone regulator.  The core regulation function was easily lifted from the controller project, but will tell you - HUI (Human User Interface) is just a PITA!!!   It is why I always (well, when I was programming for a living) preferred embedded controls or system programming (I liked to write languages / compilers, OSes, and the thing that does something when you push down a key - like decide WHAT key you pressed down - but never 'applications'  Awk.....  )    BUT:  I have persevered, made a simple ASCII string based command set to allow for updating of things like charge parameters and timings to be stored in the EEPROM of the Atmel CPU.  Status is sent out via an ASCII  stream with the idea someone I am working with make some nice looking apps for PCs and perhaps PDAs to see the status and/or make changes without needing to recompile the source..

Parts are accumulating at our mail-drop - I will be picking them up at the end of the month.  Then  more and more fun!  I am really interested to see how this new FET driver ckt works, really like how it looks on paper.

And on the full controller, it continues to work well - though our actual use of the Generator is declining with increased sunny days (and the solar panels).  Of late my efforts have been with the oneWire temp sensors - dealing with restarts when they fail, and also adding in validation code to look for wako data.  ala, our last generator run faulted out with Battery Temp = 185f - well over the 140f trap.  And no, that was NOT the true battery temp!!  (185f, or more correctly 85c seems to be a defined error state of the DS18x20 onewire temp senders...)

[BruceM]

Glad you're making such good progress on your standalone universal alternator regulator, Thomason.

Your HUI for setting programmable parameters is a very nice feature.  I've avoided this labor by using a three pin header to allow code/parameter updates via laptop reprogramming.  Your approach is way more professional and should make for a more marketable product.  I predict it will be very popular.

Best Wishes,
Bruce

[thomasonw]

Glad you're making such good progress on your standalone universal alternator regulator, Thomason.

Your HUI for setting programmable parameters is a very nice feature.  I've avoided this labor by using a three pin header to allow code/parameter updates via laptop reprogramming.  Your approach is way more professional and should make for a more marketable product.  I predict it will be very popular.

Best Wishes,
Bruce

I also intend that a PC will be used to do configuration changes via the HUI - communicating over either the Service Serial port or Bluetooth via  to-be-written fancy point-n-click application(s - idea is could support PCs, mac, iPhones, Androids, NMEA-2000??, etc..  just a small matter of programing - and not my me, some one is offering to do that part ).  But I elected to use clear ASCII as opposed to Binary data packets for communication to and from the external platform.  This was to allow for the use of those different external platforms without getting into issue associated with sending raw binary data, ala: how a given computer chooses to format a floating point number. . . .  (It also  makes a really hard to read sudo-HUI for those who like CLIs - Command Line Interfaces!!!)

And FYI:   I just posted up the draft of the Source (as bench-testing on a UNO, hardware will not be here till the end of the month), and a .wrd document covering the ASCII communications.  https://docs.google.com/folder/d/0B5GiaoeXCQ3vQU5ncXhETjQ3OGM/edit?docId=0B5GiaoeXCQ3vSWpZTGVPMDlpRlk

Should also point out there is an 8-position DIP switch on the regulator that can be used to select different predefined charge profiles and system sizes.  One need only go to the ASCII stuff (and/or the nice to-be-written PC/Andrid/... apps) if they want something other than those pre-defined profiles.  (Or, one could just simply modify the source code and recompile things!!!).  Serial port is also used if one wants to see lots of numbers come out of their regulator (Vbat, Amps being produced, Bat and Alt temp, Engine RPMs, etc..). 

But in its simplest use none of the above is needed.  All one needs to do is use the DIP switches to select the charge profile for their batteries,  indicate the capacity of their batteries (again via the DIP switches) and hook up the regulator.  Before starting anything I do a raw VBat read via the Atmel A/D converter and decide what system voltage we are connected to (12v, 24v, 36v, or 48v) then make needed adjustments to the charge profiles to match.   That is it!

And one special feature (because I am a BOAT guy) is a 'sync' port:  If you have say a twin-engine boat charging the SAME battery bank, connecting a telephone cable to the two SYNC ports will allow the regulators to share information and better coordinate their activities.  Maybe this could extend to land-based as well?  - in the case of someone having two DC generators on site:  a LARGE one and a SMALL one..  Idea is to prevent them fighting each other while charging and share info to make better charge state transitions decisions.

AS this moves forward, I likely will look to migrate some of these ideas back into the fully integrated engine control and regulator.  But for now will work to proof them on the stand-alone alternator regulator.

[BruceM]

Very advanced and capable software features...I think it will be popular for off grid battery charging too. I'm jealous, my sophisticated software days are over.

Best Wishes,
Bruce

[vdubnut62]

You two make this sound so easy! And I understand just as much as a goldfish understands nuclear fission.
But please keep up the good work. With luck I can buy one and just maybe figure out how to hook it up.
As long as the instructions are picture heavy
Ron

[thomasonw]

You two make this sound so easy! And I understand just as much as a goldfish understands nuclear fission.
But please keep up the good work. With luck I can buy one and just maybe figure out how to hook it up.
As long as the instructions are picture heavy
Ron

Ya, it does seem like Bruce and I tend to get into Full Propeller Mode rather easy 

I do hope to make this not only flexible; and being open-sourced and in public domain - very flexible -  for anyone who wants to work from this design.  But am also hoping to make the standalone regulator (the non-integrated one w/o any of the engine monitoring, starting, throttle controller, etc)  with the same easy-of-use as standard external regulators  -->  select the options you want via the 8 DIP switches and connect it up.  All the 'geek-dum' is needed only if you want to do things like:  Use your iPhone (yours, I do not have one...) to see what is happening with your system.

A question does come up about 'manufacturing'.  At this time I do not really have a vision of going into 'production', this is more of a hobby for me - plus I get a really great regulator!  But perhaps later this fall (or winter) can consider a build of a few units for folks who are interested.  But want to debug the 1st units before anything like that . . . . .

[thomasonw]

As long as the instructions are picture heavy

Ron

ok Ron   

Today I made up this picture:



Had planned to anyway and am in kind of a holding pattern until I can get to our mail drop and pick up the PCB and parts.   But this shows how I intend the external regulator would be connected to an alternator.  Of special note are the the A, B, C, and D terminals.  These are used to allow the regulator to support P-type (High Drive) or N-Type (Low Drive) alternators w/o any other changes.  As shown (with the Jumper between A and B) the regulator is configured for a P-Type (High Drive) field. The Field is then connected to C/D (or in this case just C as the other end of the field is connected internally in the alternator to ground.)  To support a N-type, one would jumper C&D, and connect the Field to A/B.  (Or just B if the alternator has connected the other end of the Field to BAT+ internally).

The DIP switch is used to select built in charging profiles, while the Bluetooth (or via the Service Port) can be used to enter custom profiles.

The Battery + and Battery - connections should be made AT THE BATTERY, while the Alt + and Alt- can be made directly to the alternator.  (The reason I separated the BAT and ALT wires is to allow support for Alternators which might have a different field voltage then the charging battery.  e.g., using a converter 12v alternator to charge a 48v battery where the field is still 12v).

[vdubnut62]

So simple even a Cave Man can do it. 
I don't have an Iphone either, I want a removable battery phone in case I want to go dark.
Thanks!!
Ron.

[thomasonw]

Made an update to the 'picture'.  Here is the revised one (I had the Current shunt + and - backwards).     Wish there was a way to go back and edit my posts - seems like that option is taken away after a few days.....

[glort]

I would to have something to control a F&P 240V generator or an induction motor Generator.

I have really been trying to follow this interesting thread but shamefully, so much of it is over my head. Please excuse my EXTREME ignorance, but I'm wondering if it would be possible to use a PC with the right connections and software to do this?
I can get any amount of desktops, Laptops and servers and would have a lot more hope setting one of those up than these other little machines I just have no knowledge of.

If that's not possible, are there any commercially made Controllers that would be OK for an induction motor as a generator?

[thomasonw]

I would to have something to control a F&P 240V generator or an induction motor Generator.

I have really been trying to follow this interesting thread but shamefully, so much of it is over my head. Please excuse my EXTREME ignorance, but I'm wondering if it would be possible to use a PC with the right connections and software to do this?
I can get any amount of desktops, Laptops and servers and would have a lot more hope setting one of those up than these other little machines I just have no knowledge of.

If that's not possible, are there any commercially made Controllers that would be OK for an induction motor as a generator?

Good morning, and please I would say no shame.  I spent +/-20 years BUILDING PCs (IBM, Epson, Intel, etc.), but just run away when I have to do much more than simple 192.168.xx.xx things. . . .

This project is really oriented towards a DC 'generator', not the AC one you are describing.  Although some features (ala start / stop engine, monitor temps and oil pressure) could be reused things like engine speed control and such might not have the precision and reaction time to allow for stable 60hz (50hz) AC power generation.  (Am thinking about the response times of the throttle motor control I am salvaging from auto cruise control units).   I just do not know. . . . . 

It MIGHT be possible to modify this project towards an AC generator head, and if someone has more experience in that area then I and wants to work on it - am always open to some more fun!   But as it is today, I would perhaps think it is not currently the best choice...

As to other products, will leave that to folks who have more expense in those areas then I.   And again, if someone can see some value in a modified version of this project - leveraging the engine control portion and making needed changes on the AC side? -  and wants to perhaps start something let me know!

-al-

[Ronmar]

Well an excited field generator is prettymuch an excited field generator.  A little code and a scaler for sampling 240VAC, and I could see this thing driving a ST head fairly easilly...

There is no real way to regulate an induction motor to deal with varying loads.  An induction might work OK making bulk DC that is regulated much like a solar system regulates whatever it gets from the panels...

[BruceM]

It's possible to do dynamic switching (say via Triacs) of capacitors to regulate an induction motor, but I haven't seen such a beast commercially, and I think it seems a whole lot easier and more practical to use a generator head.

I have a 4 capacitor, zero cross switching circuit and board I can provide to someone who is determined to have a regulated induction motor generator, but it was designed for rectified AC so the DC mosfets and drivers need to be switched to Triacs.
It was used for my 120VDC battery charger MK I version, which I abandoned due to severe power factor related problems with my AVR when generator fired.  I switched to a dead nuts simple motorized Variac design (controlled by my Picaxe battery bank monitor), which has worked very well with great power factor since I use a 1100 watt toroidal transformer primary as a choke for DC filtering. I could have added a microcontroller chip (I'm a PIC fan) to my ST-3's homebrew AVR and done my AC battery charge regulation there, but I wanted to be able to use some AC while charging the batteries.

+1 Ronmar's comment - No reason Thomasonw's lovely state of the art open source design couldn't be slightly modified for any AC or DC generator head.

[thomasonw]

I know for my work with the EPA and the ENERGY STAR programs, measuring AC can be a little different then DC, but there are also nice integrated chip solutions out there that could take the place of the INA-220's currently being used.  Driving the Field for Voltage Regulator would be doable I suppose, and perhaps the simple cruise control throttle control units would be responsive enough to regulate RPMS and frequency?   With all that, could leverage the whole soft start, cool-down, alarm, remote, and instrumentation capability of the controller..

If folks are interested in this as a 'next step' (once I finish the stand-alone alternator regulator project (http://www.microcogen.info/index.php?topic=3116.0) I have been having fun!   Woudl need someone to work with me who actualy has an AC generator and would be able to use it for a test-bed, my system is all DC..   

But if there is interest in a AC verion of this integrated controller / regulator project, and someone who could assist me . . . . . .

-al-

[BruceM]

I did try out one of the analog RMS AC voltage chips for my AVR.  Alas, it was too slow in conversion so my flicker compensation suffered badly and I could not tolerate my shop heat/light lamps. So I reverted back to my simplistic analog RC integrator for AC voltage measurement.  It can't handle big loads with poor PF, so I stopped trying to do that.

I think you are best off doing your own RMS calculation via the processor, after AC rectification and resistor scaling, as the RMS voltage can be computed every half cycle.  An analog sample every 500usecs would be dandy.