Open Source DC controller

AC, DC, amps, volts and kilowatt. It's all discussed in here
User avatar
jstan
Groupie
Posts: 41
Joined: Thu, 21 May 2009, 19:57
Real Name: John Stanton
Location: Swan Reach Vict

Open Source DC controller

Post by jstan » Fri, 17 Jul 2009, 03:00

I have just finished a new parts list, I can't work out how to post it to here however here is a summary, the complete list will be posted on Paul's forum.
John


Mouser Total        US$33.29 AU$58.90
Plus Shipping of US$30.00=AU$37.50                         AU$96.40
If four people combine the shipping is a flat rate so four into AU$37.50 = AU$9.38 plus AU$58.90 =AU$68.28

Digi-Key Total     US$200.37AU$251.00
Plus Shipping AU$90.00 =                                     AU$341.00
If four people combine the shipping is a flat rate so four into AU$90.00 = AU$22.50plus AU$251.00=AU$273.50
               Based on conversion at 9 30pm 16/7/2009                    

Total cost does not include boards
Indivindual orderMouser AU$58.90Shipping AU$37.50Total AU$96.40
          Digi-Key AU$251.00Shipping AU$90.00Total AU$341.00
               Total-----------AU$437.00

Four people combine Mouser AU$58.90Shipping AU$9.38Total AU$68.28
          Digi-Key AU$251.00Shipping AU$22.50Total AU$273.50
               Total-------------AU$341.78

Mouser has free shipping to Aus. For orders over $200 if you order in AU$ however their prices for buying in AU$ are quite a bit higher than buying in US$.
Shipping parcels should be kept under $1000.00 , over this attracts Customs
JohnStan

User avatar
jstan
Groupie
Posts: 41
Joined: Thu, 21 May 2009, 19:57
Real Name: John Stanton
Location: Swan Reach Vict

Open Source DC controller

Post by jstan » Sat, 18 Jul 2009, 02:28

I can blame it on age or rusty brains however I got it wrong last time, I mixed up the prices Mouser charge for AU$ orders free shipping and their price if you elect to pay US$ . Here is my revised list prices. Of course the more people to share the cost of shipping the cheaper per person.
John

Mouser Total        US$33.29 AU$41.43
Plus Shipping of US$30.AU$37.50                    AU$78.93
If four people combine the shipping is a flat rate
so four into AU$37.50 = AU$9.38 plus AU$41.43          AU$50.81
For Mouser prices AU$ with free shipping AU$58.90
if four or more people order                            AU$58.90
Digi-Key Total     US$200.37 AU$251.00
Plus Shipping AU$90.00 =                      AU$341.00
If four people combine the shipping is a flat rate
so four into AU$90.00 = AU$22.50 plus AU$251.00     AU$273.50
               Based on conversion at 9 30pm 16/7/2009                    AU$308.50

Total cost does not include boards
Individual order     Mouser AU$41.43          
Shipping AU$37.50     Total AU$78.93
                Digi-Key AU$251.00          
Shipping AU$90.00     Total AU$341.00
                         Total-----------AU$419.93

Four people combine   Mouser AU$41.43          
Shipping AU$9.38      Total AU$50.80
Four people combine Digi-Key AU$251.00          
Shipping AU$22.50     Total AU$273.50
                         Total-------------AU$324.30

Mouser has free shipping to Aus. For orders over $200 if you order in AU$ however their prices for buying in AU$ are quite a bit higher than buying in US$.
Shipping parcels should be kept under $1000.00 , over this attracts Customs



JohnStan

Yogibear
Noobie
Posts: 13
Joined: Sun, 12 Jul 2009, 18:18
Real Name: James Williams
Location: Corinda

Open Source DC controller

Post by Yogibear » Wed, 29 Jul 2009, 22:15

Trying to get a hold of some recent circuits to build my own controller I would appreciate it if you could send me something
Jimbo

User avatar
jstan
Groupie
Posts: 41
Joined: Thu, 21 May 2009, 19:57
Real Name: John Stanton
Location: Swan Reach Vict

Open Source DC controller

Post by jstan » Wed, 29 Jul 2009, 23:54

Jimbo.
the only controller circuit available that I know of is the one we are following on the forum at http://ecomodder.com/forum/showthread.p ... 4-211.html . They are testing their controller at present, you can find the schematic at their web site, the parts list that I have posted there is probably the final however we are waiting on the present tests. Don from Brisbane Has offered to have the Power board made, he will have to order them in lots of 50 to get the price down to $12.00 plus postage. The logic board should be no dearer and will be organised as soon as Paul is satisfied that the tests are OK and no alterations are required.
John
If you want me to send you the schematic and parts list please let me have your Email address
Last edited by jstan on Wed, 29 Jul 2009, 13:59, edited 1 time in total.
JohnStan

Yogibear
Noobie
Posts: 13
Joined: Sun, 12 Jul 2009, 18:18
Real Name: James Williams
Location: Corinda

Open Source DC controller

Post by Yogibear » Thu, 30 Jul 2009, 15:40

Im getting to like this group of people, everyone seems to be genuinly helpfull, Im retired, but had an electronic business in Sydney, when scr's where the new thing running DC motors on the mains and they finally got a darlington up to 10A at 400v. I think I still have an old trax program that might run with vista. My real interest will be in trying to come up with a better lighter battery system, high voltage and using AC. I dont want to close down the back seat of my Barina. I'll be in on the 2 boards so count me in. My email is jswills67@hotmail.com. please anything that you can send to help me would be trully appreciated. I have been through the inventing stages and it is only a field that lets the rich get richer. I like this open forum (help each other) Thanks for your reply
Yogibear Jim
Jimbo

User avatar
jstan
Groupie
Posts: 41
Joined: Thu, 21 May 2009, 19:57
Real Name: John Stanton
Location: Swan Reach Vict

Open Source DC controller

Post by jstan » Thu, 30 Jul 2009, 23:00

Jim
Have a good look at the size of the Li-Ion batteries, they are fairly small and quite light compared to Lead acid. You may not need to lose the back seat of your Barina.
John
JohnStan

User avatar
avolt
Noobie
Posts: 19
Joined: Thu, 22 May 2008, 23:29
Real Name: Andrew
Location: N.S.W Australia

Open Source DC controller

Post by avolt » Fri, 31 Jul 2009, 03:09

Hi John
you can count me and a friend in too,(x2) when the time comes,and paul's happy with the controller,hope it test out ok
they are doing such a good job!
thanks and Cheers
Andrew

Yogibear
Noobie
Posts: 13
Joined: Sun, 12 Jul 2009, 18:18
Real Name: James Williams
Location: Corinda

Open Source DC controller

Post by Yogibear » Fri, 31 Jul 2009, 14:10

Thanks John I will look into the li-ion I think the Preus uses Nmh
Jimbo

Yogibear
Noobie
Posts: 13
Joined: Sun, 12 Jul 2009, 18:18
Real Name: James Williams
Location: Corinda

Open Source DC controller

Post by Yogibear » Fri, 07 Aug 2009, 15:32

Thanks John I would like to see them, does any one in Brisbane have them and has anything eventuated with the controler as I,d like to put it together. For it to work 120 volt the power devices would probably be 1200 Volt
Jimbo

User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Open Source DC controller

Post by Johny » Fri, 07 Aug 2009, 16:02

Yogibear wrote:... For it to work 120 volt the power devices would probably be 1200 Volt
They generally use 200 V MOSFETS for 144 VDC systems. The fast recovery diode(s) take care of back EMF from the motor.
FETS get significantly more expensive over 200 V, the high current (120A) ones especially.

The AC Inverters we are using from 600VDC battery packs use 1200V IGBT devices.

User avatar
jstan
Groupie
Posts: 41
Joined: Thu, 21 May 2009, 19:57
Real Name: John Stanton
Location: Swan Reach Vict

Open Source DC controller

Post by jstan » Fri, 07 Aug 2009, 16:20

Jim
I don't know of anyone in Brisbane with Li-ion batteries. If you contact Ian of ZEVA in W.A. they seem to be the main importers for Australia.
We are still working on the cheap controller, I will email everyone who has enquired when it is up to the stage of ordering. I have found someone who will fabricate aluminium cases for $20.00, we are working on the heat spreaders and bus bars.
John
JohnStan

vetinavan
Noobie
Posts: 3
Joined: Wed, 27 May 2009, 01:35

Open Source DC controller

Post by vetinavan » Sun, 09 Aug 2009, 18:18

Count me in too when the time comes.
Boards & Digikey/Mouser parts

User avatar
jstan
Groupie
Posts: 41
Joined: Thu, 21 May 2009, 19:57
Real Name: John Stanton
Location: Swan Reach Vict

Open Source DC controller

Post by jstan » Mon, 10 Aug 2009, 17:29

We are ordering copper for Bus Bars and heat spreaders as well as Digi-Key parts this week for a Australian Bulk Buy, anyone wanting to join please message me.
John
JohnStan

Yogibear
Noobie
Posts: 13
Joined: Sun, 12 Jul 2009, 18:18
Real Name: James Williams
Location: Corinda

Open Source DC controller

Post by Yogibear » Tue, 11 Aug 2009, 14:33

Yes let me know how to send you money. I am going to build the controller.. Regards Jim (yogibear)
Jimbo

User avatar
mcudogs
Groupie
Posts: 122
Joined: Fri, 17 Apr 2009, 01:27
Real Name: Don Saxby
Location: Brisbane
Contact:

Open Source DC controller

Post by mcudogs » Sat, 29 Aug 2009, 15:12

Latest circuits for the controller are at

Control board

and

Power board

I should have the new board layouts completed in about a week.

New functions include a motor overspeed detector, battery voltage monitor, spare relay output driver, spare analog / digital input or output.

There will also be a software mod to make the controller fail safe if a MOSFET goes short circuit.

Squiggles
Senior Member
Posts: 742
Joined: Wed, 22 Apr 2009, 03:19
Real Name: Neil
Location: Newcastle NSW

Open Source DC controller

Post by Squiggles » Sat, 29 Aug 2009, 20:01

There are two things happening worth watching, maybe both by the same chap (Adam?). He is doing a very nice alternate power section arrangement and using multiple FET driver chips.
I think they are both likely to be significant improvements.

User avatar
mcudogs
Groupie
Posts: 122
Joined: Fri, 17 Apr 2009, 01:27
Real Name: Don Saxby
Location: Brisbane
Contact:

Open Source DC controller

Post by mcudogs » Sun, 30 Aug 2009, 01:11

Adam's design had the face of the mosfets against the heat sink not the metal tab. It would have lasted about 2uSecs under full load. He also hasn't done any thermal calculations. I don't think I will be waiting for any more innovations from that source.

Squiggles
Senior Member
Posts: 742
Joined: Wed, 22 Apr 2009, 03:19
Real Name: Neil
Location: Newcastle NSW

Open Source DC controller

Post by Squiggles » Sun, 30 Aug 2009, 03:23

Yes, you could be right, although I like the fact that he found a heatsink before designing the power board. Apparently he has redone it with the devices turned over.

Also he is using more FET drivers and he seems to have considered propagation delay in his driver board layout. So he must be getting some decent advice from somewhere.
Keep in mind Paul had no idea to start with and there are still a few items I believe can be improved and probably will be.

It is still early days in the evolution of the Cougar!

Yogibear
Noobie
Posts: 13
Joined: Sun, 12 Jul 2009, 18:18
Real Name: James Williams
Location: Corinda

Open Source DC controller

Post by Yogibear » Sun, 30 Aug 2009, 13:42

Having been involved in the development of new products, I really apreciate and enjoy the value of this site. Cougar is likely to be the envy of all when she is finished. Putting youth as a bonus on my pension.. Thanks..Looking forward to building Cougar.. This is also upgrading me as a few components weren't available when I was at the table
Jimbo

bga
Senior Member
Posts: 492
Joined: Mon, 01 Sep 2008, 19:27
Real Name: Bruce Armstrong
Location: Perth WA

Open Source DC controller

Post by bga » Sun, 30 Aug 2009, 20:44

Hi All,

Just looked at the schematics for the controller.

I have a few concerns:
a)
There is no gate protection on the transistors.
This is in two parts:
1 - It needs a 20 volt zener to limit Vgs to prevent
      overvoltage damage to the gate in case of static
      discharge or transients.
2 - It needs a pulldown resistor of, say 10K on the gates
      to prevent inadvertent turn on if the controller
      board or one of the gate drive lines are disconnected
      under bias.
I think that it would be better to move the gate resistors and driver IC to the power board so that there can be a single connection to the logic PCB and the in-built protection of the driver IC can function at all times (no undervolt lockout?).


b)
The flyback diodes are inadequate and will likely fail if the controller is used in direct to diff drive applications with largish motors;

The PWM controller operates as a buck type DC-DC coverter, supplying the motor with a reduced voltage for speed control.

At low speed, the motor current is high and the battery current is low. The 'on' duty cycle time of the transistors is also low, meaning that most of the time, the flyback diodes are conducting the motor current. This current will be at, or near to the controller's current limit of 750 amps and for approximately 90% of the time. The transistors are only rated for 30 amps each continuous, or about 300 amps in this system.

I had a look at Farnell for their selection of diodes, it's not very good in TO220s. It may be worthwhile investigating using the same type of FET as the low side driver with the gate shorted to source via a resistor. The intrinsic diode in the power FET is a lot more capable than the discrete diodes.

c)
The fast (hard) switching is likely to generate a lot of RF harmonic noise.
It may be necessary to limit the switch on/off rates to reduce RFI via an RC network on the gates. Diodes can be added to make the switchoff faster than switch on if needed.

[edit: grammar corrections to make sense]
Last edited by bga on Mon, 31 Aug 2009, 07:25, edited 1 time in total.

MPaulHolmes
Noobie
Posts: 7
Joined: Wed, 02 Sep 2009, 07:24

Open Source DC controller

Post by MPaulHolmes » Wed, 02 Sep 2009, 08:28

Hello! I think D6 protects against spikes. This is the datasheet:
http://www.diodes.com/datasheets/ds21502.pdf

Also, see R17, a 1k pulldown to make sure the driver output is pulled low. I learned the hard way to make sure to have one of those. The bike controller I made first didn't have that, and the mosfets shorted all at once. made me mad!   Image

It would be better to have the gate resistors/driver off the control board, absolutely! hehe. We will definitely do that.

There is undervolt lockout. The mosfet driver is disabled if either of the 12v or 5v supplies drops below a certain naughty point.

Each diode "package" is rated for 60 amps continuous. Each diode is really 2 diodes inside the to-247 case. The overcurrent trip is adjustable from about 490 to 775 amps, but it's only safe to adjust it to maybe 550 or 600 amps if you wanted to live really dangerously. It is very safe and functional set to 500 amps. But just add a couple more mosfets/diodes and you have a 600 amp controller instead.

At low throttle, the current is also low. The software uses a PI loop that can be tuned to any motor so that current tracks throttle very well, even with a locked rotor (that's how I tuned it), which is the worst case scenario for brushed DC motors.
hi!

Squiggles
Senior Member
Posts: 742
Joined: Wed, 22 Apr 2009, 03:19
Real Name: Neil
Location: Newcastle NSW

Open Source DC controller

Post by Squiggles » Wed, 02 Sep 2009, 13:55

bga wrote: Hi All,

Just looked at the schematics for the controller.

I have a few concerns:

b)
The flyback diodes are inadequate and will likely fail if the controller is used in direct to diff drive applications with largish motors;

The PWM controller operates as a buck type DC-DC coverter, supplying the motor with a reduced voltage for speed control.

At low speed, the motor current is high and the battery current is low. The 'on' duty cycle time of the transistors is also low, meaning that most of the time, the flyback diodes are conducting the motor current. This current will be at, or near to the controller's current limit of 750 amps and for approximately 90% of the time. The transistors are only rated for 30 amps each continuous, or about 300 amps in this system.

I had a look at Farnell for their selection of diodes, it's not very good in TO220s. It may be worthwhile investigating using the same type of FET as the low side driver with the gate shorted to source via a resistor. The intrinsic diode in the power FET is a lot more capable than the discrete diodes.


Are the intrinsic diodes in the power FETs to be ignored in this design?
Do they not work in parallel with the external flyback diodes?

Squiggles
Senior Member
Posts: 742
Joined: Wed, 22 Apr 2009, 03:19
Real Name: Neil
Location: Newcastle NSW

Open Source DC controller

Post by Squiggles » Wed, 02 Sep 2009, 13:56

MPaulHolmes wrote: Hello!

Nice to see you here Paul, welcome.

MPaulHolmes
Noobie
Posts: 7
Joined: Wed, 02 Sep 2009, 07:24

Open Source DC controller

Post by MPaulHolmes » Wed, 02 Sep 2009, 14:12

Hey Squiggles! The intrinsic diodes in the mosfets don't help. The freewheel diodes conduct from M- to B+, whereas the mosfet intrisic diodes go from B- to M-. Isn't that sad!?
hi!

Squiggles
Senior Member
Posts: 742
Joined: Wed, 22 Apr 2009, 03:19
Real Name: Neil
Location: Newcastle NSW

Open Source DC controller

Post by Squiggles » Wed, 02 Sep 2009, 21:16

Ah, yes, maybe I should have studied the circuit a bit closer!

Post Reply