DIY Electric Car Forums banner

1 - 20 of 54 Posts

·
Registered
Joined
·
730 Posts
Discussion Starter #1
When I registered here, I set myself the goal to start my first conversion build in 2018.
And a couple of months ago I decided to go for it after I saw the first results of €1000 DC conversion here.

To do a build on a shoestring and in a short period of time, the electronics have to be simple (no uC) and inexpensive.

No build can do without a precharge circuit for the caps of the power stages. Motor controller in the first place.
So it's no surprise that I have seen quite a few posts with questions about precharging.
And it's the reason for starting a seperate thread on it.
Precharging hasn't been high on my list of priorities until now, because there's nothing exciting about slow charging of caps.

But actually, once I started to design a circuit, it turned out to be very pleasant experience.
With a strong 70s feel to it. Yeah, discretes! I love it. Paint it orange and purple.

So I ended up with 14 discrete components to get the caps charged and the main contactor switched on automatically once the charging is done.
Now we have to put it to the test.
Parts bill (retail): €10 incl. 1x1" PCB. Of course excl. case, connectors, cables, contactors.

Inputs:
- battery + from precharge contactor, min 40V, max 1000V.
- power stage caps -
- power stage caps +

Outputs (isolated):
- GND (12V system)
- MAIN CONTACTOR COIL -, other coil lead connected to +12V rail (12V system).

Or can I just copy a €10 circuit with the same functionality that has been posted here before?
 

·
Registered
Joined
·
952 Posts
I used 150R power resistor with TL431 reference, high (680k/xxx) resistor divider, 1W 5V DCDC converter with PC817 optocoupler and DPDT relay. Of course there is one 1A NPN transistor there too. And it works! While voltage is under divider setting TL431 is off and opto is keeping relay ON using pullup on signal line. If voltage goes higher than divider setting TL431 gets lit and PC817 pulls signal to GND. Relay goes OFF and precharge ends. NC contact wiring on relay enables DC contactor line to trigger only when "precharge complete" is signaled. Like a double interlock safety. If you need you can also wire a mushroom button into precharge line to release precharge relay GND and disable car precharge, DC contactor and everything.
 

·
Registered
Joined
·
952 Posts
If you're at it...could you also incorporate an economizer into the contactor driver? Most EV contactors don't like 12v continuously...
You should use 24V AC coil it holds at 12Vdc and doesnt heat-up as much. Or just get used Tyco EV200 contactor type...

https://www.ebay.co.uk/itm/Tyco-Electronics-Kilovac-Relay-EV200H1ANA-12-24VDC/163086719206?epid=16020218871&hash=item25f8b9d8e6:g:71AAAOSw8nxbGWO5
https://www.ebay.co.uk/itm/KILOVAC-Relay-EV200-Contactor-500-Amps-12-900VDC-9-36VDC-EV200HAANA01-12-24VDC/131397160850?epid=1676098530&hash=item1e97e18f92:g:g~AAAOSwMpZUq-fC
 

·
Registered
Joined
·
1,122 Posts
For as often as you precharge (once per ride), I would use a resistor and a pushbutton and be done.

Push and hold for a few seconds, she'll be precharged enough. Solved problem.

Next.
 

·
Registered
Joined
·
565 Posts
You should use 24V AC coil it holds at 12Vdc and doesnt heat-up as much. Or just get used Tyco EV200 contactor type...
But....the 12V ones are very cheap coming out of used car batteries...

If you want to go cheap: use the unit from a used battery: Fuse, + and - contacters, precharge resistor + precharge contactor, all in a neat package...(usability depends on brand / type of car)

Combine this with Tony's precharge controller... (or a bms with drivers)
 

Attachments

·
Registered
Joined
·
730 Posts
Discussion Starter #10 (Edited)
About €6. :)

The controller can handle the high current of a DC contactor continuously. I have to do the math and measurements as a check, but I estimate 25A peak (200msec), 5A continous @ 50 degree C.

The 400V 300A Panasonic AEP-XX-300-12 draws up to 3.3A at turn-on, only 0.33A after 100 msec.
 

·
Registered
Joined
·
730 Posts
Discussion Starter #11
@ Arber333

I don't use voltage as a trigger. The output to the main contactor is triggered by the charging current dropping below a threshold (now set at 1mA).

So it's independent of the SoC of the battery.

No current is drawn from the caps to be charged.

The circuit draws about 30 microAmps @ 100V from the battery but only during precharge.
 

·
Registered
Joined
·
2,229 Posts
...and because it's triggered by precharge current instead of voltage it's independent of the pack voltage:) I like it.
 

·
Registered
Joined
·
2,229 Posts
Is the blue component in the bottom left a cap?
 

·
Registered
Joined
·
730 Posts
Discussion Starter #16 (Edited)
No, it's a the toroid core of the pulse current transformer.


Not done yet with the final checks of the design. then a few measurements and finally a circuit diagram.


Oops, the toroid is in the upper right corner.


On the left there's a blue PTC thermistor.
 

·
Registered
Joined
·
730 Posts
Discussion Starter #17 (Edited)
The circuit diagram.

Done the stability analysis of the ac small signal model of the mosfet/optocoupler current sense and bypass circuit (second png).

A 1uF cap (C?) provides adequate phase margin at zero crossing points with varying precharge resistor values and Gfs of the mosfet.

Major changes: divider network ohmic value is compromise, mosfet (Q1) bigger package (power rating),
 

·
Registered
Joined
·
730 Posts
Discussion Starter #18
The circuit has been pushed forward and the measurements back and then you get this: the first mods.

Diode D2 was LL4148, now back to schottky SGL1-40 as shown in picture of prototype.
Resistor R5 will have a fixed value (10 to 15R probably).
The value of resistor R4 will be determined by measurements for maximum gate pulse duration.
Toroid transformer will be 16mm, 10:10 turns 0,4mm, to support longer duration of pulses.


Great fun, this!:D
 

·
Registered
Joined
·
730 Posts
Discussion Starter #19
OK, that's the discrete circuit. It can do with a 47R resistor in series with a 470u/35V elcap across the output thyristor, because the load has a high inductiance value. Helps with turn-on.

BUT NOW, bring in the ASICs.

Input (current sense):
First a isolated current switch with a open collector output: Honeywell CSDA1BA. https://www.digikey.com/products/en?keywords=csda1ba
Ten turns at the sense port will trigger the output low at 32 to 80 mA. Great.
But it does have a 6 to 16V power supply input. The discretes can do without.

Output: First pick is the smartmos switch BTS3110. https://www.digikey.com/products/en?keywords=bts3110. Built-in protection: ESD, overcurrent, temperature etc.
Can be driven with PWM for ECONOMIZER purposes. Package: SOT223.

Maybe a tiny PWM ASIC for the ECONOMIZER? Why not the LTC6992IS6? https://www.digikey.com/products/en/integrated-circuits-ics/clock-timing-programmable-timers-and-oscillators/689?k=ltc6992 It appears to be working OK in an inexpensive motor controller. :)


Anything else? Hmm, no, think that's all the ASICs it takes.
 
1 - 20 of 54 Posts
Top