Please describe pinouts. I need UART..
you mentioned in #43 that wave soldering for the AL pcb , is this true ? or do you mean reflow oven ..
and if any of these methods is used , Does it use normal profile or a special profile ?
Normal profile seems to work for solder paste with low temp melting point.
Vally & Eva are waiting your controller )
Are those whale fishing poles? If so, I would understand the desire for 20kw or maybe more. Otherwise, what are the specific requirement to use so much power on a spool?
This looks like a fun guessing game
Are Vally and Eva used for towing up paraglider and/or hang glider pilots?
For HG we need up to 90kg tug-force, which would explain the 20kW.
This is paragliding towing. Complicated thing. I did many tests on 75300 (https://www.youtube.com/user/eptveptv123) for one seater wings. And calculated that I need 30kw controller for double seater (tandems).
I need 150Kg of continues force without overheating at all. And controller must handle high volts when paraglider is flying back. About 20Ms for safety.
Also I need relatively small 1-2Kg (7-10A) forces to keep rope tension when paraglider is flying back from the winch.
Your controller seems very promising for this application.
Could you inform about good DC-DC battery to 12v 5A?
I connected internal fans to 75300 AUX port. And I found that it is impossible to start fans when motor is working. It seems like 75300 12v used to control fets.
Also, I wrote application for safely control (https://www.youtube.com/watch?v=-R5f-PkWRyo) towing. It can be found under branch 'skypuff':
The 75300 can only supply 1A total on the 12v, since the 12v is also used to generate the 5v and 3.3v you probably can't use more than 0.8A before you have a problem.
Those delta fans can draw like 2-4A each depending on the model so best to use something external to the vesc in this case.
Here are some dcdc that might work
So these work as servos?
Having fans on the stator would make sense if the heat can escape the hub. I saw some guys pumping air through the hubs. That seemed to help with cooling.
Not like servos, but I count steps to control position. Force is applied and removed in a smooth way. Automatic take off sequence and wind up.
Fans make cooling speed about 5 times more:
Do you have news about controller?
We are adding a lot of improvements to the design for reliability. The hope is to start making a few prototypes in early May. Virus permitting.
Saw the "antisex dampening" in your code. It makes sense now after watching the video.
I am unhappy that I have to add additional DC-DC and circuit to run fans. Of course there should be a direct firmware controlled 12V 5A output to fans with powerful controllers..
I am unhappy about a few things but I am happy about a lot more other things. The net sum of this is that I am on the positive spectrum of happiness. And I think that's what matters.
As for DC/DC, 12 at 5A amounts for 60w of power. It would take some space in the controller and take some engineering, testing, etc.
Yes, it can be done as a part of the controller and I think I'll do it at least as an option, but for now, it's not the focus because it can be easily added off the shelf.
Seal controller's shell- enclosure hermetically and add in/out PC water cooling quick change connector's on both end's ! and get a radiator for oil cooling , and use it with transformer oil , You will need a water cooling pump tho . D5 Pump's are the best choice , very strong and reliable . Don't know how well if it would handle oil tho but . Cooling would be hell of a lot more efficien't that with air . Just need to tap in-to main battery https://www.overclockers.co.uk/xspc-d5-vario-body-screwring-wc-141-xs.html There are also passive cooling way's like thermall encapsulating potty for better heat transfer . Usually used in encapsulating electric motor windings - recuces air gap and yield's better thermal https://electrolube.com/product/er2183thermally-conductive-epoxy/tranfer t outer case to radiate out
Hi . I there any progress with your custom made Foc controller ? It must be difficult to get anything going because of the Covid -19 ! I'm cheching this page almost everyday LoL
Anything new? When is the next batch going to be produced? I want to buy a controller.
Yes, working hard to improve the design. The big change is that the controller will have now three boards: power stage on the Al substrate, then DC link board with capacitor banks sandwiched to it, and a control board with digital and analog circuitry on top. The focus is to maximize the performance and reliability. The plan is it get a small batch made in May.
Great work! I love the use of Fairchild FETs they are seriously good and well worth the extra money.
The Design has undergone some major revision. It's a three board design now:
1. Aluminum substrate power stage at the bottom providing the best thermal interface with the heatsink
2. DC link + Driver + Sensing board on top
3. Control board plugged into the DC Link board away from high voltage current tracks
Among other improvements, increased isolation barrier for signal integrity, embedded contactor circuit with precharge/arc handling, on board BLE, WiFi connectivity for easy firmware updates.
Optional but easy to add: SD card for data logging, GPS/GSM module for alarm, tracking, etc.
Boards are being built now, I should have a few working prototypes pretty soon.
@Vadicus I am interested in ordering one of these units. Anyway to get in to the pre-order for May still?
MX E-Bike Enthusiast
I am too, happy to test out the new hardware.
If you need testers I am sure you will find enough people in this thread
This is looking good and in a small form factor . Your 3 board setup is the way to go which is standard in high performance controllers ( cirtus for example) .
They use custom made busbars to move current from AluPCB to DC-Link PCB. Check the circled Items in the image . However for the phase they use those circular non-standard current sensors to save them from going into the DC-link PCB. just an idea if you havenot seen it before. In your design it is clear how are u moving current from buttom PCB to middle PCB ?
One more thing , infineon introduced lately this package PG-HSOG-8-1: https://www.infineon.com/cms/en/product/packages/PG-HSOG/PG-HSOG-8-1/
This is similar to PG-HSOF-8-1 . However it has pins like the traditional packages. they recommend this because is more suitable for aluminum pcbs because of the different thermal expansion , this can protect the pins from cracking , however I cannot find the App note right now.
Thanks for the tip on the mosfet package. The power stage is designed to accept "legged" packages such as D2PAK if needed but I am yet to confirm there is a benefit using those. Legless mosfet package allows to decrease the parasitic inductance. Thermal expansion may not be an issue on Al substrate as the whole heat dissipation is at least 2x better then on FR4, so the controller will run pretty cool in the first place. The power roll back is also set at around 90C or so.
Regarding the current loop current sensors, I've evaluated those and decided against them early on due to cost, availability and accuracy.
Any chance this has an IMU?
There is a chance if you add it. It's like $10 or less.
Wireless/BT touch display is now an option for the 24F controller: https://vesc-project.com/node/1814
Telemetry dashboard is implemented. Now, working on adding MC and APP configs. It will be possible to set up new motors and adjust parameters from the display.
Getting closer to shipping out the first batch of pre-orders. Also, working on optimizing production to make them faster.
What's the time between ordering and shipping?
How can I order one? On the powervelocity website it always says there are not enough in stock, even for preorder? Also has the prive gone up to 599 from 499?