Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[willbrunner]

Put in an order for the J1772 inlet and Anderson connector, so looking forward to the complete system.  I am wishing now I had ordered a second Eltek unit, because as nice as 2kW will be, 4kW would be better...

[togo]

Post your 3-D printable cover, Will?

[willbrunner]

I'm getting some 10 & 12 AWG cable on my way home today, and I want to make sure it all fits inside that cover before I post it.  I am also thinking of adding some tabs to help hold it on to the power supply case.

[originalspacerob]

Hello Everyone,

    I have been trying to set my voltage between yesterday and today and have not been having much luck.  I can load the sketch onto the Leonardo and it will send to the Rectifier but I keep getting the yellow light.  I have done it about 40 times and the same results.  I have the CanH and CanL going to the right place.  If I don't I do not get a response. I tested the board with a BLINK sketch and that works ok so I know the board is working.  For the Can connections I used the attached picture for reference for connection locations on the rectifier.
     I wait until the rectifier powers up and then I plug in the Leonardo.  I have the shorting plug on the board so that it uses the terminating resistors.  I tried two of my three rectifiers and all do the same thing.  I am attaching the sketch that I used which is the sketch that was posted on this thread.  I changed the serial number and set the voltage to 114 volts (114x100) for setting the voltage.  Any thoughts or ideas would be greatly appreciated.  I have a feeling I am missing something just can't put my finger on it.   

Serial Number:112071117284
Voltage Set: 114V   114x100= 11400  Hex: 2C88 (reversed hex to add to sketch 882C

[willbrunner]

A couple of things stand out:

The voltage setting message ID in Remmie's post was 0x05009C00, not the 0x05009C02 you have.

I waited 1 second between logging in and sending the voltage set.  You also have to make sure to wait 30 seconds after the voltage setting before powering off the rectifier.

Can you read any CAN frames coming over from the rectifier?

[originalspacerob]

Thank you for taking a look willbrunner. I tried the ID both ways and neither worked. The 0x05009C02 will give me a yellow light after 10 seconds and 0x05009C00 will give me no yellow light and no voltage change after even 1 minute.
I have not tried reading any CAN frames as of yet.
Could my CS pin settings be wrong?  I noticed that when remmie posted the sketch he used CS pin 10. But on this post the sketch attachment shows pin 17. That is the pin I have been using. I have my board wired with the screw terminals and not the pins. I'm also using the leonardo CANBUS from hobbytronics.

Updating the picture so show better detail how the Leonardo was hooked up during programming.  The picture does not show the power and meter hooked up yet.  But they are to test while programming. 





Sent from my iPhone using Tapatalk

[Skidz]

Just chipping in, having programmed two 1800W units before

First thing I notice is the wire on the 0v (ground)... CAN is a differential signal so you shouldn't connect ground. Some folks tinkering with the 1800W Eltek's blew up their Leonardo's by connecting the ground.
Second, did you make sure the terminating resistor is there? If it isn't, the bus is unbalanced and 'the bits fall out of the cable'
Third, I also had to modify the sketches to make sure the right pins were used. Since it's longer than 3 weeks ago I'd have to scavenge my archives for what i've changed exactly but I remember pin assignment...

Good luck programming the unit(s)!

[originalspacerob]

Thank you Skidz,

I updated my picture on my previous post.  I never used the ground on the CAN at all.  Just the CAN H and CAN L.  That picture was confusing with the wires laying underneath.  I think I may have a pin issue also.  Not sure if anyone else used the Db9 connection or if they used screw type connections like me.  Depending on which sketch you look at one shows use CS pin 10 and the other CS pin 17.
   I looked at the Hobbytronics CAN sketches and they appear to use CS pin 17 for output.

[willbrunner]

Yeah, the hobbytronics schematic shows ~CS connected to digital pin 17.  http://www.hobbytronics.co.uk/datasheets/ht/leonardo-can-bus-100.pdf

I notice on the standard Arduino Leonardo, there is no digital pin 17- that pin is the RX LED.  Are you using a version of Arduino that lets you control that pin?

Check avr/variants/leonardo/pins_arduino.h line 198 and make sure it says // SS      D17      PB0               RXLED,SS/PCINT0

[originalspacerob]

Yes.  I am actually using the Leonardo CANBUS from Hobbytronics.  That one appears to use pin 17 like the sketch that was posted.  I tried adjusting the voltage again today at lunch at work and no luck.  Used the same sketch as I posted above.  There were a few things I changed.  I changed the delay time from (100) to (1000) so it would change the lead in to 1 second.  That didn't work.  I also changed the 0x05009C02 to 0x05009C00 and had no luck. The board is definitely communicating with the rectifier because I get the yellow light like they said could happen and I may have to load a it a few times.  I loaded the base sketch in abut 30 times and no luck.  Waited like instructed.  I wonder if maybe my voltage is wrong, I did multiply it by 100 and then converted to hex, and then proceeded to flip the hex number like instructed.   I'll try again tomorrow.  I left everything at work. 

[originalspacerob]

I finally got it figured out.  I was able to change the voltage as needed to 115 volts.  All my flatpack are high by about .5 volts compared to what I set but hold steady.  So I set the set point to 114.5 and it sits right at 114.9-115.  I will have to check once I put a load on them.  The problem was not the pin anymore.  It is Pin 17 on the Hobbytronics Leonardo CANBUS for the proper CS pin.  The problem was that the Sketch had  "unsigned char setdefaultvolt[5] = {0x29, 0x15, 0x00, 0xBA, 0x2C};"  the 0x29 was 0x2B in the sketch that was shared.  I read some of the older posts on endless sphere comparing sketches and besides the CS pin for some other boards the 0x2B I had on my sketch was supposed to be 0X29.  After that change the voltage changed to what I wanted.  I was even able to play around and set the voltage to 104VDC even though they say it can only go to 110.  I didn't try any lower. 
   What I also found is that you don't have to turn the rectifier on and off each time.  Once it is running, you can make a change on the sketch, upload it and the rectifier will change in 30 seconds.  The yellow light comes on no matter what.  Once it comes on about 2 seconds after the voltage changes.  The yellow light will stay on forever until the rectifier is powered off and back on.  The next time the rectifier starts the yellow light will not be on and the voltage will be permanent like others have found.  So in my case the yellow light did not show and error, only that the sketch was finished sending and it showed the rectifier received it. 

Thank you to everyone that replied and helped work on this.  Very much appreciated.


Sent from my iPhone using Tapatalk

[ZEROCOOL]

Good to see you got it figured out. my 2 arrived today, purchased the hobbytronics arduino leonardo now i just need get everything else for these.  Got them for my 17 FXS 6.5 non modular
wanted 3 but read somewhere in 1 of the several eltek flatpack threads 3 plus the onboard might be to much for the FXS 6.5 so i have 2.

[originalspacerob]

Great news. I read the same thing when I was reading on the old threads. I am the same as you and need to get the other items also to hook everything up to the bike. 


Sent from my iPhone using Tapatalk

[Lenny]

Good to see you got it figured out. my 2 arrived today, purchased the hobbytronics arduino leonardo now i just need get everything else for these.  Got them for my 17 FXS 6.5 non modular
wanted 3 but read somewhere in 1 of the several eltek flatpack threads 3 plus the onboard might be to much for the FXS 6.5 so i have 2.

Your charging current limit will be around 52 A. If I looked at the right spec sheet, each unit is limited to 16.8 A of current (16.8 A * 120VDC nominal voltage = 2016W). 3*16.8 A equals 50.4 A, so it should fit perfectly using three units without onboard. At voltages below 120V the units are current limited to 16.8A, so the output power will be lower than 2000W.

Just give it a try, if you actually surpass the limit the red triangle will start flashing and the contactor will open a few seconds later. However, don't do that too often, because the contactor will wear.

If it turns out to be too much, which I don't believe, you could still regulate one unit to lower current using the hobbytronics board.

[Doctorbass]

Good to see you got it figured out. my 2 arrived today, purchased the hobbytronics arduino leonardo now i just need get everything else for these.  Got them for my 17 FXS 6.5 non modular
wanted 3 but read somewhere in 1 of the several eltek flatpack threads 3 plus the onboard might be to much for the FXS 6.5 so i have 2.

Your charging current limit will be around 52 A. If I looked at the right spec sheet, each unit is limited to 16.8 A of current (16.8 A * 120VDC nominal voltage = 2016W). 3*16.8 A equals 50.4 A, so it should fit perfectly using three units without onboard. At voltages below 120V the units are current limited to 16.8A, so the output power will be lower than 2000W.

Just give it a try, if you actually surpass the limit the red triangle will start flashing and the contactor will open a few seconds later. However, don't do that too often, because the contactor will wear.

If it turns out to be too much, which I don't believe, you could still regulate one unit to lower current using the hobbytronics board.

You are totally right!.. thanks for adding these explanations. That is saving some work from me