Page 1 of 1

Precharge & PostDischarge

Posted: Tue, 02 Aug 2011, 01:46
by ibruinsma
Here is version 3 of the Precharge and Post Discharge circuit that a number of us have used so far.

This one does not use a rectifier, it switched the polarity of the Buzzer by using the second half of the relay.

I have also found that if you are using a DC circuit with a very large capacitor, you may want to lower the RC time constant for faster pre-charge/post discharge time.

Simply reduce the precharge load resistor value. Each time you half the value you halve the precharge time, and theoretically quadruple the power required. However the Power does not really quadruple as the time that resistor is dissipating heat is shorter (consider the resistors thermal time constant of about 30 Sec). So it should be OK to double the power for halving the resistance (within limits of course).

The resistor will be damaged (over heat) if you have a load after the precharge. The low power value means it can only take transient discharge pulses. The buzzer will sound when it is dissipating power, so if you hear it continually buzzing cut the main power immediately.

Happy Precharging,

Ian Bruinsma (Smartarse)

Three Images attached in three parts.
I have the full sheet in PDF but I can't upload that here..




Precharge & PostDischarge

Posted: Tue, 02 Aug 2011, 05:11
by evric
Hi Ian,
One problem I can see, is that it's very easy to switch the Ignition switch straight though to the "ON" position. I'd like to see a timeout period here to eliminate that possibility.

Precharge & PostDischarge

Posted: Tue, 02 Aug 2011, 14:02
by Tritium_James

Shouldn't you also be precharging the input caps on the DC/DC converter? That relay you have switching the HV DC to it won't last as long as you'd like, I suspect. Note that it's also not rated to switch DC above 30V.

Precharge & PostDischarge

Posted: Fri, 12 Aug 2011, 00:37
by polo-ev
This circuit as is shown using the Jaycar SY-4050 relays is potentially dangerous & will fail the NCOP14 Checklist Clause 4.3.
The components suggested are not rated for high DC voltage use.

These relays are rated for 30VDC 10A max only.

They are not safe to use to switch HAZV traction pack voltages at any current & must not be used in this application.
This circuit would also have -ve HAZV traction pack voltage present at the buzzer -ve terminal when the ignition switch is off.
This could also result in a very unsafe situation as most people would only expect 12V DC to be present at a buzzer.

Electrical safety is of utmost importance when building an EV.
All components used must be rated correctly for the DC voltage level present & expected current load within a circuit & all circuits must be designed on a fail-safe basis.

Precharge & PostDischarge

Posted: Fri, 12 Aug 2011, 00:41
by evric has some high voltage solid state relays which work well.Part number: SSRDC200V40A

Precharge & PostDischarge

Posted: Fri, 12 Aug 2011, 02:24
by polo-ev
Solid state relays are a great way to go.
They do need heat sinks if you want to switch more than about 5-10A continuous.
A couple of electronics guys I know also recommend flyback diodes on the control side of them just in case there are voltage spikes from relay coils & the like elsewhere in the circuit.
There is a simpler solution to the precharge.
Two contactors - 1 primary & 1 secondary.
Precharge resistor permanently across secondary contactor.
Turning IGN on energises primary contactor & starts precharge.
Simple delay circuit then turns on secondary contactor.
(Cougar controller has built-in pre-charge delay timer)
Postdischarge can be done by simply having a 10W 12K or 15K resistor permanently in circuit.
Yes, it wastes a little bit of power when the car is running but the amount is negligible.
Turn the car off, both contactors open & the resistor slowly bleeds the caps down. Takes about 15 minutes.

Precharge & PostDischarge

Posted: Sun, 21 Aug 2011, 03:30
by ibruinsma
Excellent stuff guys...

It is good to see there is some good discussion about safety here... I think there is not enough safety applied to any of this EV stuff...

Yes I agree with the safety hazards when using AC rated parts that are used as a DC circuit breakers. All the reading I have done so far suggests that an AC rated circuit breaker should be reduced by a factor of at least 5 times or more for DC rating.....
So yes the 30V DC part is kind of under rated...

I started out with this because I saw guys with fully charged caps at hundreds of volts not considering how to charge or discharge them. Working on things live is extremely dangerous... But I have seen it many times...

As a gut feeling only, the reliability will be OK because the currents are so small.. (But not insignificant!!!). So it is very unlikely that contacts will not fuse closed or lock open. However the open circuit DC arc may survive for a while if the discharge resistor is low enough to maintain plasma (arc). But the currents (arc) will only survive for the time it takes for the cap to charge or discharge (if it can be maintained at all with such a high resistance...). I would love someone to help here who has knowledge of arc distances when the DC current is limited by a large resistor.

But I am not keen to try it and touch the Controller end at any time (even if it has been discharged etc....)

There is enough current to kill you in single fault condition. SO DON'T TOUCH ANYTHING EVEN IF IT HAS SINGLE FAULT PROTEECTION APPLIED...

But much more importantly it does not meet regulatory requirements by first principal (e.g. correct DC voltage rating).

I think the easiest would be design for double fault protection level so that two faults must occur with a check for with fault showing failure...

I think some more careful consideration needs to occur here...

Keep up the good work guys...

Precharge & PostDischarge

Posted: Sun, 21 Aug 2011, 14:16
by ibruinsma
Oh Yea, Someone pointed out that the Relay part number was not changed from Ver 2.0 (SPDT: Jaycar SY4050) to Ver 3.0 (DPDT: Jaycar SY4052).

So if you still want to experiment with this circuit, then the DPDT relay (Jaycar SY4052), even though it doesn not have the "DC Volatge Rating".

"I don't make mistakes" - Oops!

Precharge & PostDischarge

Posted: Fri, 02 Sep 2011, 04:41
by ibruinsma
After studying up on AC and DC relays, and why the DC rating is lower than the AC rating, it ends up that:

1)     Both AC and DC form an arc when the contacts open, however for AC the arc is only sustainable until the AC has a zero crossing (approx 10mS worst case). A DC arc is sustained continually if/while the right conditions are met.

2)     The Arc is only sustained if there is enough DC current available to sustain it.

3)     The Arc (DC or AC) has a relatively low resistance (high current). Remember 10mA for 20mS is enough to kill you. There can be well over 10mA.

4)     There is sufficient series resistance so that there is not sufficient voltage accrss the contacts (or current) to maintain an arc.

5)     The AC voltage rating has sufficient contact (insulation) to with-stand 1.414 times the AC rating at DC before the isolation will break down. i.e. a 240V AC rated relay will have sufficient insulation to handle well above 273V DC as long as the contacts remain open (not at the closed to open state).

6)     In normal DC switch/relay, the switch will have DC arc rating at about 1/5th of the AC rms voltage rating at DC. Hence they are rated at much lower DC rating than AC rating (normally 1.5th). This is because an arc always forms when the relays opened if large currents are flowing and will be sustained at DC even when the contacts are fully open. This is not the situation in our case (read below).

7)     The DC rating is even more degraded if there is an inductive load. This circuit’s load is strongly resistive.

8)     The AC and DC current ratings are similar because it is mainly due to the amount of current that will flow during the arcing process, and during normal operation (contact resistance and closure). The AC and DC current rating are slightly different because of less arcing time that occurs at AC (10mS zero crossing).

9)     Both the AC voltage and DC voltage rating are related to a small extent by the load current, however the voltage peeks at a max of the basic isolation resistance (contact gap and coil isolation) of the relay.

10)     The peek operating current in this circuit is less than 1/15th of the relay rated current. That is, it operates at 0.7A for a 10A rated device at 240V. This current occurs at switch on (0.7A), but quickly fades away towards zero amps. This occurs both for charge and discharge condition.

11)     The initial current (of 0.7A) is not sufficient to maintain an arc at DC “brake over distance” for the rating of well above 273V. No arc will even be formed in this circut.

12)     The reliability of a relay/switch is mostly determined by amount of in-rush current (0.7A) and/or how long/much arcing occurs (amount of heat, plasma deposits, and pitting). Both of these are minimal and thus the reliability will be well beyond even the “main contactor” reliability in your car.

13)      Even under single fault condition (e.g. arcing, contacts damaged, main contactor faulty, or resistor burns out etc…), the controller caps will either charge/discharge or not-change/not-discharge. It will not become hazardous unless a second fault occurs (like the basic insulation for the traction voltage is compromised). Thus, there remains a double fault protection with a safe architecture.

14)     Note: your controller, motors, connections, DC- DC converter, current sense, main contactor, fuse, and battery terminal wiring is mostly not double-fault protected, so be aware of these much more hazardous items.

15)     A fault in this circuit will cause an immediate alarm (or lack of warning) that the pre/post discharge circuit is not normal. If you continue to use it at this faulty state, the “first fault” is the operator (you!!!).

16)     But most convincing is the simple observation that there is no voltage across the contacts at the time they are opened, either at pre-charge or post-discharge. But even it there was, and there was sufficient current to maintain an arc (which there isn’t), the arc could only last for the time it takes to charge/discharge the controller. This abnormal condition could only be maintained for a short time in any case.

I do however note that anyone using this above 250V DC should take extreme care and over rate the:

•     Relay,

•     Wiring,

•     Isolation, and

•     Level of fault protection

by many factors.

Treat it is as you would treat house wiring or industrial wiring in a potentially wet (splash hazardous) environment. If you do not have experience in dealing with the subtle faults that could occur at these voltages, then don’t even try experimenting – Get an education first!!!!!!.

There are good reasons why there are extremely strict regulations in any equipment above 50V in all commercial and industrial applications. Single fault condition must still remain safe (i.e. double fault protection), plus there should be a regular/automatic check system that can find single faults before they can become double faults and kill some one.

You (the designer) are responsible for any person who comes into danger by your negligence or lack or education. This includes on-lookers, other road users, mechanics, inspectors, and family etc…. It would be a sticky court case you would be involved in if your negligence was the cause of someone’s harm/death.

I state again: I do not take any responsibility for the safely of this circuit or its application. It must remain the responsibility of the end designer to design a sufficient level of inherent safety into these systems.

Only 10mA for 20mS is enough to kill you. Less than this can seriously injure you!!!!!!

So be safe guys!!!!!!

Precharge & PostDischarge

Posted: Tue, 01 May 2012, 17:02
by ibruinsma
I would like to add that this circuit still needs you own design work on the:

* 12V battery Charger connection and relay (it is wrong), and pre-charge (if possible) for the charger etc...

* Fusing everywhere to the appropriate limits on all signals, including the DC-DC converter circuit. Especially protecting the user (12V ignition system) against pre-charge relay catastrophic faults (100mA +12V and -12V isolation relays).

* "Start Key latching" regulatory requirements (and logical usage).

* Charging disable control, and charging main contactor switching etc..

I stress that this is only a technical note, not a whole design solution. You are the designer and you are responsible for your own design and safety requirements (verification etc…). I take no responsibility for this stuff..

Precharge & PostDischarge

Posted: Tue, 01 May 2012, 20:34
by Richo
Tell you the truth all I see is a complete lacking of proper eV controllers.
You would think after 5-10 years of controllers being built for eV's this would all be built in.

Precharge & PostDischarge

Posted: Tue, 01 May 2012, 23:32
by ibruinsma
Yea, I agree. Why can't all this be in controllers, including the main contactor, it does not cost much and it makes it all so simple...
I wish!!!!

One day – I will design a controller/charger/main contactor/BMA interface all in one box, now wouldn’t that be logical…

One day… When I am rich and famous!!!

Precharge & PostDischarge

Posted: Wed, 02 May 2012, 00:56
by Tritium_James
Richo, the reason it's not in the controllers (speaking as a controller designer!) is that you also need to be precharging your DC/DC, charger, A/C compressor inverter, etc, etc.

The real question is: why is that functionality not built into everyone's BMS! :)

Precharge & PostDischarge

Posted: Wed, 02 May 2012, 20:49
by Richo
I guess because the combined DC/DC and compresser inverter are unlikely to blow the traction fuse.
Why else would you bother with precharge in the first place?

Precharge & PostDischarge

Posted: Fri, 04 May 2012, 00:35
by Tritium_James
It's to protect your contactors. Not precharging = welding.

It's also to protect anything connected to the switched side of the contactors. Capacitance (in various things, motor controller, DC/DC, etc) with Inductance (batteries & cabling) gives an LC resonant circuit, and a maximum voltage on the devices of double the battery voltage. In the real world it's less than this because it's actually an RLC circuit, but it can be alarmingly high if you don't precharge, because the R is usually pretty low.

Precharge & PostDischarge

Posted: Fri, 04 May 2012, 20:22
by Richo
Then I guess it's not built into motor controllers already because no one cared about it before.
I haven't heard of any members having issues with thier contactors.
But this maybe due to most of the conversions are "low" voltage DC and have little capacitance anyway.

Precharge & PostDischarge

Posted: Fri, 04 May 2012, 20:34
by evric
People should care about it.
The ones that don't care can expect to have contactor contact points that become pitted, reducing the contact area and therefore increasing the resistance and causing excessive heat and could actually "weld" together.
This un-caring attitude basically reduces the life of an expensive item and the safety aspects of the EV conversion.
By the way, all of this: the precharge and the contactor are built into the Solaton controllers.