10S custom skate ESC: testers wanted!

hey
- connected the VESC to a 6S
- no motor connected only USB to laptop - ubuntu 14.04
- followed all the instructions and running february BLDC release (even here if it doesn't matter as I cannot get the VESC detected by the OS)

1432546009-img-2714.jpg


error in dmesg :) not sure what to do, I have tried other USB slots in this laptop.

Code:
[  348.284156] usb usb3-port1: unable to enumerate USB device
[  354.416224] usb 3-2: new full-speed USB device number 22 using uhci_hcd
[  354.536221] usb 3-2: device descriptor read/64, error -71
[  354.760126] usb 3-2: device descriptor read/64, error -71
[  354.976232] usb 3-2: new full-speed USB device number 23 using uhci_hcd
[  355.096214] usb 3-2: device descriptor read/64, error -71
[  355.320102] usb 3-2: device descriptor read/64, error -71
[  355.536149] usb 3-2: new full-speed USB device number 24 using uhci_hcd
[  355.944174] usb 3-2: device not accepting address 24, error -71
[  356.056110] usb 3-2: new full-speed USB device number 25 using uhci_hcd
[  356.468171] usb 3-2: device not accepting address 25, error -71
[  356.468248] usb usb3-port2: unable to enumerate USB device

=> THIS HAS BEEN FIXED BY CHANGING THE USB CABLE. BELIEVE IT OR NOT ! :)
 
thank you torque

weird... does it mean that to configure my motor I need a specific device ... I thought that the BETA was kind of standalone, just configure my VESC with USB to my PC and that's it. I don't want to upgrade firmware, just to detect my motor and go ride! Not sure having a STLINK v2 was a pre requisite for the beta ! otherwise it could have been good to include it (or at least to specify that it was needed !)

from VESC conf instructions:
Connect the ESC to the USB port of your computer and click “Connect” in BLDC Tool. The lower right corner should now say “Connected”. If you have gotten this far, you should be ready to connect a motor and configure the ESC from BLDC Tool.

I may have missed something because it is not stated to have any STLINK here

is this OK :

http://www.ebay.fr/itm/ST-Link-V2-Mini-Metal-Shell-STM8-STM32-Programmer-Emulator-Downloader-/221773345364?pt=LH_DefaultDomain_186&hash=item33a2b8b254
 
torqueboards said:
You'll need the Stlinkv2 and/or Stlink board to go with your VESC. You cannot connect VESC by usb only.

VESC needs Stlink, USB + Power.

As far as I know, the VESCs from onloops beta have old firmware preloaded, so an old version of BLDC Tool should work without using a SWD cable to upgrade the firmware. This is a link to that old version:
http://home.vedder.se/public/VESC_Beta_Jacob/
 
thanks Benjamin. that would be so good if people that are going through the VESC Onloop beta have all the same info ! (having a Onloop VESC BETA wiki would definitely help and that will allow to have BETA lessons learnt and everything updated on a single page !)

my problem is that I cannot get my VESC detected by the ubuntu, the BLDC tool is more at the application layer that us the /dev detected by the OS
 
STLink v2 USB Program dongle
Here are images of the STLink v2. You can use any STLink v2 on eBay. Just as long as the pins line up. I'm pretty sure. Ben would have to verify.
IodptMK.png

QPHQxt6.png

The white JST connector is a 5 pin JST connection 2.0mm spacing.

Dual ESC Can Bus Connector
4 pin JST 2.0mm - http://www.ebay.com/sch/i.html?_odkw=stlink+v2+jst&_osacat=0&_from=R40&_trksid=p2045573.m570.l1313.TR0.TRC0.H0.Xjst+2.0+connector.TRS0&_nkw=jst+2.0+connector&_sacat=0
2 males on both side.
Only need CANH and CANL connected.
One esc must be set at primary and other as secondary.
Use 3 pin wire for RC receiver.
 
Huge thanks to Benjamin. Got it working. No need of STLINK at this time to configure the VESC.

I will post a full detailed walkthrough but I want to share with you a short video of two VESC running a dual diagonal setup on my existing board. I will switch fully to VESC but I need to modify my enclosure and be sure that the capacitors used on the onloop beta are approved by Vedder :) The test you are going to see have two VESC, one master, one slave using CAN bus communication. This thing is SO smooth. Hardly believable.

You will also see bandaro DIY TX enclosure ... best of both worlds.

One thing is that on one of my two VESC, the configuration was totally crazy with extreme weird values. I used a SK xml as the baseline, did auto detect motor (PPM has to be off for this to work) and then put the two values manually, then save. Slave VESC don't need PPM - only CAN communication has to be enabled.

Enjoy. It's running 6S for my desk test...

[youtube]Kl8NUMXMLHU[/youtube]

@onloop : what are the specs of the capacitors used ?
 
lizard said:
Has anybody done some tests with higher currents or in RC cars yet?

I'm planning to put the VESC in my RC car. I don't have exact numbers, but from what I have seen in comparable setups it draws somewhere between 100 and 200A during acceleration (with RC ESC without current control). But that is only for short periods of time, maybe 3-4 seconds. Average current consumption is somewhere around 40A, runtime is about 10 minutes on a 6s 5000mAh pack. Car weighs around 3.7kg with Castle 1717 1650kv motor and is geared for 160kph.

I have read that maximum current limit is about 150A because of the shunts.

Can I just set it to 150A and try it out? I am not sure how fast the heat builts up in the FETs, I assume it will not get too hot if I'm a little careful and check temps often? Or will the FETs overheat in a second or so (faster than the temp sensor can probably sense because it's not inside the FETs) when I run 150amps through them?

Vedder have the VESC on a RC car
http://vedder.se/2014/05/.

The specs says 240 A for a couple seconds. About 50 A continuous.
http://vedder.se/2015/01/vesc-open-source-esc/.
 
Barndom said:
lizard said:
Has anybody done some tests with higher currents or in RC cars yet?

I'm planning to put the VESC in my RC car. I don't have exact numbers, but from what I have seen in comparable setups it draws somewhere between 100 and 200A during acceleration (with RC ESC without current control). But that is only for short periods of time, maybe 3-4 seconds. Average current consumption is somewhere around 40A, runtime is about 10 minutes on a 6s 5000mAh pack. Car weighs around 3.7kg with Castle 1717 1650kv motor and is geared for 160kph.

I have read that maximum current limit is about 150A because of the shunts.

Can I just set it to 150A and try it out? I am not sure how fast the heat builts up in the FETs, I assume it will not get too hot if I'm a little careful and check temps often? Or will the FETs overheat in a second or so (faster than the temp sensor can probably sense because it's not inside the FETs) when I run 150amps through them?

Vedder have the VESC on a RC car
http://vedder.se/2014/05/.

The specs says 240 A for a couple seconds. About 50 A continuous.
http://vedder.se/2015/01/vesc-open-source-esc/.

With the 1mOhm shunts, about 150A is correct. For a 6s RC car that needs that much current for a couple of seconds, I recommend using irfs7430 FETs and 0.5 mOhm shunts. Then you can set the current limit to 200A. I haven't tried it myself, but I'm quite sure that the temperature sensor will be able to keep up if it is set to 100 degC like default. The FETs should be able to handle 200 degC for a fraction of a second, and the sensor should not lag behind that much.
 
okp said:
@onloop : what are the specs of the capacitors used ?

Not sure. But I'll try to get the details.

as far as I am aware (based on a discussion between jacob & ben) ben checked this before we proceeded with the final assembly. And he confirmed these parts where suitable & As far as i know He also has one on hand.

Also there are more reports of these "DRV Fault code" issues on other boards that people have built that where not from our BETA batch..

To complicate matters There are also people who are loading new firmware onto there BETA VESC and using incompatible/old BLDC tool..

So for now i think it is best to take a global approach & focus attention on the VESC as a whole and then discuss the various, specific, issues that are being detected.. perhaps a new thread could be started for each specific issue that arises.

So no, I dont think splitting info into a specific wiki just for the 50 people that got units from our beta run will be beneficial in the long run. Let's keep it all together.

I'd suggest if ANYONE Is asking ben questions or looking for technical support from him & haven't donated to him yet maybe a small donate of $5 to $10 is worth considering. Or more.
 
@onloop. I will donate for sure. No doubt.

I just want to be sure that Ben agrees on the capacitor specs used for the BETA VESC, as he's questioning it few post earlier; which is why I am also asking. I plan to run it on 12S and don't want issues to fry my VESC or cause damage to other components.

As of now, everything goes smoothly, just need to build the new enclosure and make this fit inside.

I plan also to add, as advised by Ben, two 40A car fuse in series to avoid any issues if motors get in short circuit. This is just prevention purpose.

I think the BETA VESC should be managed as a specific program for the buyers/testers, if a firmware upgrade is needed, it should be specified somewhere to ensure everyone is aware on the BETA VESC as we have all the same VESC in the enertion batch. Of course if people want to upgrade firmware by themselves for whatever reasons or for the fun of it, or just fry their ESC by doing wrong stuff... it's their matter; but at least there should be a common VESC BETA support thread or wiki with reliable information and clear installation guidelines. At this time, only reliable information seems to be provided by Benjamin or partially by isolated posts; so this is why I am asking him for his kind support :) Mixing BETA VESC with other is just mixing everyone and making this a little mess from my opinion.

A bit off topic but when you SPACE batteries will be ready ?! :D
 
On another topic.

I am constantly being asked about VESC Availability.

So i am considering a 2nd run of 100 BETA VESC units.

This will be with a different supplier. The factory we used to make them in the first batch claimed they lost money on the 50 we produced and they have substantially increased the price... not to mention they took longer than they promised And swapped stuff around.

Note: my dual vesc setup (same as all 50 beta vesc) is working flawlessly.

So anyway now I am using 4 new specialist factories/suppliers

1. The PCB maker.
2. The Components supplier
3. The Micro Device Assembler
4. The firmware loader / tester / Co-ordinator.

》There will be no parts substitution.
》The latest firmware (at time of assembly) will be loaded.
》the capacitor on the power leads will be modified and instead of using one I will use 3 in parallel which makes the unit thin/flat.. they will also be closer to the fets.


Anyone interested?
 
Technically speaking the Enertion BETA VESC should not be any different from any other vesc.

As This is an open source project and the instructions to make this device are published openly All devices should be identical.

so we should all be able to help each other No matter the origins of the device.

so the source of the device should not necessarily be the primary method of how the information should be organized. Perhaps the date it was created is the best method to sort the user feedback.

Is there some option in git hub for user feedback?

However as one if the organizers of the beta program I did recognize that a single source of information would be useful.

I also created a thread. Here: http://endless-sphere.com/forums/viewtopic.php?f=31&t=66958&hilit=vesc#p1007172

But nobody used it.... probably because the moderators moved it to the sales section
 
Something to think about. Fake components are used very heavily in china. Any heavily used component like the arm chip or drv8302 for sure has counterfeits made in china. A drv8302 is $15 for authentic, a chinese company that is using fakes probably gets them for under a buck. From the pictures of the beta boards the printing on the both of the pinned chips do not match up with the chips I have purchased earlier this year and do not match up with vedders board pictured on his site which is a much older revision of his so so the chips used on the beta's could be a much older version or could very well be fake. If they are fake then who knows how different they are from the authentic chips Ben buys directly from mouser and does all testing with. Seeing as how the components even when purchased in bulk are $70 to $80 USD plus the cost of the board finding a company that will use authentic components and not fakes and build the board for a low price in china does not seem logical. There are three people so far on this thread that had stated they did not work after a short while so I am just saying this. I am sure you know I am selling these as well, I really do not care if you sell them or at what price but keep in mind you would be selling Ben's board with possibly counterfeit components. The way to tell is the printing on the chip, authentic chips will have very high quality ink that will not scratch off and all letters are exactly the same shape/size down even looking thru a magnifier like pictured. I would confirm with your manufacturers that they are using actual authentic chips and not counterfeits.


ti.png

Snapshot2015_0526_092353.jpg

Snapshot2015_0526_092716.jpg
 
silviasol said:
Something to think about. Fake components are used very heavily in china. Any heavily used component like the arm chip or drv8302 for sure has counterfeits made in china. A drv8302 is $15 for authentic, a chinese company that is using fakes probably gets them for under a buck. From the pictures of the beta boards the printing on the both of the pinned chips do not match up with the chips I have purchased earlier this year and do not match up with vedders board pictured on his site which is a much older revision of his so so the chips used on the beta's could be a much older version or could very well be fake. If they are fake then who knows how different they are from the authentic chips Ben buys directly from mouser and does all testing with. Seeing as how the components even when purchased in bulk are $70 to $80 USD plus the cost of the board finding a company that will use authentic components and not fakes and build the board for a low price in china does not seem logical. There are three people so far on this thread that had stated they did not work after a short while so I am just saying this. I am sure you know I am selling these as well, I really do not care if you sell them or at what price but keep in mind you would be selling Ben's board with possibly counterfeit components. The way to tell is the printing on the chip, authentic chips will have very high quality ink that will not scratch off and all letters are exactly the same shape/size down even looking thru a magnifier like pictured. I would confirm with your manufacturers that they are using actual authentic chips and not counterfeits.


ti.png

Snapshot2015_0526_092353.jpg

Snapshot2015_0526_092716.jpg

Quite frankly I think your being a d!€k

And you are also coming very close to breaking the forum rule below.

》DON'T post messages for the sole purpose of attacking a person or company or just to start an argument.


1. Please show me some legit article or documents that confirm "fake components are heavily used in china"
Otherwise it just sounds like you are being a sore looser

2. "Drv8xxx for sure has counterfeits made in china" so i assume you have proof of this statement? Otherwise your are basically making shit up to better serve your own interests.

3. "A china company probably gets them for under a buck" more made up crap! Do you have any real facts?

The real truth:
our china supplier was unable to source all the parts locally specifically they where unable to source the STM and Drv8xxx chip in china. They where bought from USA Then shipped to the factory for assembly.


.... so anyway based on this false info you are spewing towards us i conclude that 》All of Your Drv8xxx chip must be counterfeits, because on your own VESC you had a "drv fault code" occur. That seems like a far conclusion, right?

So please contact your supplier and get proof that your vesc you are selling, & making a HUGE profit on are not counterfeit....

also let's see proof of your donations to ben.


Thanks.


P.S. Just so you all are aware... We used a reputable chinese electronics factory. The name of the company is Flier. I am sure bruno from alien power system can vouch for them... all of he's ESC are made by them.

i
 
hey ! some of you are starting to know me - I am here to learn and share with community everything about long boarding and having fun on my boards.

I found on ES a great community of good and knowledgeable people. I am not affiliated to anyone, APS, Enertion, DIYElecSkats, ADS... I just buy stuff (I am then a client and not a competitor) from everyone and also share your company links with everyone who is interested in building an electric skateboard as I am spending hours and hours on my board riding in Paris.

At this time, and as a client (I got two VESCs from BETA), I think there's no need to see silviasol post (and I didn't read it or understood it this way) as a competitor post. He's just questioning some key points based on his knowledge. I am not saying that quality checks are not done (of course theses are) but for example, I had evidence that my BETA VESC has a total different configuration installed when I read the configuration. In fact both of my VESC where not "the same" in terms of configuration. Well no matter because I have contacted Ben who told me to have both conf the same, based on a xml baseline file, and just change the two values after auto detect.

From a client perspective (I ordered the VESCs built), no need to argue - keep it easy guys ! I am just asking :
- if onloop can check with his supplier that by looking on some transistors or electronic pieces, it could be questionnable (or not) that some may be counterfeit (it can happen). Just a question.
- if onloop can provide me the capacitor specs, this has been also requested by Ben as even if it works as onloop says on 12S with no problem; Ben questioned it.

Quite easy, no competitor stuff. Let's all go through this nice BETA VESC run, post videos, installation feedbacks and continue to congrats + donate to VESC (I already did it partially when ordering my two BETA VESC, but I will continue to do it for sure) and welcome onloop on this VESC BETA for his energy on moving the scene forward.
 
As for the shunts yes we where made aware of this by Ben with our fist photo we recived from the supplyer. Ben pointed this out and told me that he needed the specs of the shunts, so I gave them to him! He pointed out that the way they where soldered would not work and suggested to mout a better way which is how they are know!

Onloop and I where not happy about them using a different shunt and I was going to put an order in with mouser that day to get them with in 2 weeks so a 2 week delay. After talking to Ben, he told me that the shunts should be fine and would perform the same as the fets until cooling was applied to the fets and then the shunts would be the weak link.

We decided to not make the changes for the shunt and offer the shunt at a lower cost to customer if they wanted to upgrade it. I currently have the shunts on order! 100 of them!

As for the big capacitor as I have had a customer that has is dented the brand a capacitor is 63V2200uF the capacitor Nichicon. I will be sending my customer a replacment capacitor.

As for the small capacitor. We asked our supplyer to you the exact parts that where on the BOM but they did not.
I have asked for the specs of the capacitor on C1 and other. I will update you all very soon.

As for our service. We both will try are best to fix any issues that any body has regarding coding and software. I have got and OS X version working and have been working very hard to deploy this into a DMG file for easy install. This will be updated when the original updated are made available and all support for this will be run through me not ben to keep ben on development not tech support.

As for hardware if you have an issue with your ESC contact us and we will work to fix it. Even if that is sending a replacmebt part with in reason we will provide a solution of some sort.

These esc's where sold cheaper then any one els at the time and made it accessable not just to our buyer but also to my self and onloop. Onloop has seen the issues with our supplyer and has organised to have the production of his next Batch moved to a better company who will order all parts exactly as the BOM says.

As for firmware I have set up software to remote access your USB port on your computor to program your ESC if it is needed or update firmware and even tech support.

Firmware can not be updated from what was shipped to current with out a stlinkv2 but if you do not feel confident doing this part but have the stlinkv2 I can use the remote software to perfom this on my computor. You will not even need to have linux installed but we recommend at this point you do install Ubuntu.

I have a VM image that I will be putting online of ubuntu with BLDC tool already configured so all that you have to do it open it in VMware virtualbox or parrellel and yes the image is compatible with all 3. This will be ass easy as downloading and clicking on the image and clicking on BLDC tool.

I am also looking into windows application as we speak!
 
if my understanding is correct, I am learning that :
- I need shunts for my 12S dual diagonal installation to be safe (not sure I have been made aware of that, or I probably missed a post)
- that the specs of all Capacitors are still not available (I assume Ben wanted to have the specs confirmed).

I don't really care about the accessibility of the tools to configure the VESC; as a newbie in Linux I succeeded. All DIY people should be working the same way. It's always good to learn. Focusing on the electronics and being sure that Ben validates that the installations done are fully safe is much more important for me.

As a participant of the BETA, having two VESCs that I will be using intensively soon on the streets of Paris, I need to get my open points fully confirmed by onloop + Ben; As a personal thought, I think it's always good to have the feedback of the BETA before going into a new batch; making sure all buyers have it fully working in the first batch before going on the next move.
 
Back
Top