DIY Electric Car Forums banner

EV High Voltage, Turning it on and off:

402K views 128 replies 49 participants last post by  Frank 
#1 · (Edited by Moderator)
I am starting this thread so I will stop hijacking other peoples conversations in order to sell my step-start delay:rolleyes:.

There are a lot of opinions on how to apply power to the controller. There are safety considerations for both the equipment and people. If we can work out some generally agreed upon ideas here then I will consolidate them and put them in the WIKI.

Some of the questions I have seen asked are:
How many contactors to use and where to put them.
Precharge, what is it, why do I need it, how do I do it.
How do I apply HV to the controller?
Do I need an emergency disconnect?
Fuses, Circuit Breakers or both?

I think I have the answers to these questions, but am open to having my ideas adjusted by other's opinions and experience.
 
#3 ·
Done. ;)

I think it's a great idea to gather as much info about these things as possible and update/add to the wiki where necessary. As such, I'm also going to make this a sticky.
 
#4 ·
Precharge, what is it, why do I need it, how do I do it.

The PWM motor controllers common in EVs have a sizable bank of capacitors on their input. When you apply a Voltage across a capacitor it initially appears to be a short-circuit, that is, the Voltage across the capacitor is zero. If there is very little resistance in the circuit, e.g. a closing contactor with no precharge, then the current will be very high. Nearly all of the traction pack voltage will be across the closing contacts. The large Voltage difference and sudden high current (known as an inrush current) can cause damage to, and in extreme cases, welding of the relay contacts. Also of concern to some is the stress on the controllers electrical components caused by the inrush current.


This can all be prevented by the use of a precharge resistor across the contacts of the main power relay. The precharge resistor allows the capacitors in the controller to slowly charge BEFORE the contactor closes. This means that there is less voltage across the closing contacts and little or no inrush current.


MORE TO COME.
 
#5 · (Edited)
Precharge, what is it, why do I need it, how do I do it.

The problem with having a precharge resistor across the contactor is, there is high Voltage on the controller terminals even when the car is turned off. This is because the capacitors remain charged all of the time.
I've heard it argued that keeping the caps charged all of the time keeps them 'fully formed' and thus, extends their life. While this is technically true, it is not really an issue with modern capacitors. Unless you plan on putting your controller in storage for years, the capacitors will likely outlast their associated active components (transistors and diodes) whether you keep them fully formed or not.

Many DIY'ers add some sort of power switch, circuit breaker or disconnect to remove the high Voltage from the controller when the car is parked.

This solves the 'high Voltage on the controller' problem BUT introduces a new wrinkle. You must now turn things on in the correct order or you will defeat the purpose of the precharge resistor.
For example, if you first turn on the contactor and then close the power switch there will be no precharge. You will have reintroduced the high Voltage/large inrush current problem.
In this case, you must first close the power switch, wait an appropriate precharge delay period, then close the contactor.

MORE TO COME
 
#6 ·
Precharge, what is it, why do I need it, how do I do it.

If a precharge switch is added in series with the precharge resistor it can be used to turn the high Voltage on without switching a large current flow, as is done with the contactor or power switch.

In this configuration the power switch becomes an emergency disconnect that is normally left on. The precharge switch is turned on first and then, after a delay, the contactor closes.
This is different than the previous design because now the "on switch" (the precharge switch) can be a relatively small relay and the turn-on sequence can be easily automated to avoid closing the contactor before precharge.

MORE TO COME
 
#7 ·
Precharge, what is it, why do I need it, how do I do it.

Here is how I did it. I have a Step-Start device that turns on the precharge relay when the start signal is received (the ignition key is turned to the START position). After a time delay the contactor is turned on.



There are additional safety and convenience features of the Step-Start Device, but the basic function is to make sure that the precharge relay is always turned on BEFORE the contactor and that at least some minimum amount of time passes between the two events.

End of Precharge stuff.

So, how about some feedback, comments, flames, dissenting opinions or support from all you other experienced and knowledgeable DIY'ers out there.
 
#8 ·
Precharge, what is it, why do I need it, how do I do it.

Here is how I did it. I have a Step-Start device that turns on the precharge relay when the start signal is received (the ignition key is turned to the START position). After a time delay the contactor is turned on.



There are additional safety and convenience features of the Step-Start Device, but the basic function is to make sure that the precharge relay is always turned on BEFORE the contactor and that at least some minimum amount of time passes between the two events.

End of Precharge stuff.

So, how about some feedback, comments, flames, dissenting opinions or support from all you other experienced and knowledgeable DIY'ers out there.
Thanks RF! I think it is very helpful. Can you describe what the step-start device is comprised of and how it works?

Thank you.

Gary
 
#10 · (Edited)
Thank RF! Very nice. I'm in the middle of some basic mechanical stuff, but will be getting to the electrical controls through the winter. I have thought briefly about this timing requirement.... but due to current tasks, haven't pursued it. Are you building any of these devices for sale by chance?

I was thinking.... what might be nice is an output to drive an "idiot light" (pardon the expression...but it works for me lol) when the precharge time is satisfied.

I assume you are monitoring the vacuum pump to ensure that the brakes are fully functional before starting? Could, perhaps, the verification of this be a signal from a vacuum switch verifying adequate vacuum presence?

Regards,
Gary
 
#16 ·
RF - please forgive my ignorance, but what is the drawback to leaving the pre-charge resistor on all of the time? Is it bad for the controller? Or dangerous? Initially I was thinking that it would drain your pack, but a capacitor is essentially an open circuit to DC current once it's charged, so that part should be negligible.
 
#17 ·
That is an excellent question.

Many DIY'ers use a high-Voltage cutoff of some type, a circuit breaker or Anderson plug for example, and normally turn off their HV when the car is parked. The step-start prevents them from powering up in the wrong order.

For systems that always have HV applied to the contactor (such as my EV) it can be a shock hazard since it leaves high Voltage on the controller terminals even when the car is turned off.

There is nothing wrong with leaving it on all the time as long as your controller terminals are well guarded.

Joe
 
#24 ·
You want an correctly rated opto to be in parallel with the precharge resister.Most opto switches are high impedance devices it would take all day to charge your caps. Better is to use your regular precharge resistor and have any sensing placed in parallel and rated for full pack voltage. They can be set to switch at anywhere from 50-30%, (caps are at 50-70%full).
 
#25 ·
How about an incandescent bulb as you described earlier with a photocell attached to it and connected in series to a digital input? When the light was on, the cell would conduct and the input pulled low. As the light dimmed, the cell would stop conducting and the input would float high.

Could the same setup be used for a 12V relay?

David
 
#26 · (Edited)
Wouldn't it be easier to use the "RUN" position on the ign switch to control the 'pre-charge' stage and the "START" position to close and latch the main contactor? Here is what I'm talking about using ladder logic.
The IGN switch would be place in 'run' until an "idiot light" turned off letting the driver know the caps are charged, then rotated to the 'start' position to turn off the 'pre-charge' and turn on the main contactor. (Think diesel engine with glow plugs)

 
#27 ·
Wouldn't it be easier to use the "RUN" position on the ign switch to control the 'pre-charge' stage and the "START" position to close and latch the main contactor? Here is what I'm talking about using ladder logic.
The IGN switch would be place in 'run' until an "idiot light" turned off letting the driver know the caps are charged, then rotated to the 'start' position to turn off the 'pre-charge' and turn on the main contactor. (Think diesel engine with glow plugs)
#What happens if the driver doesn't wait until the run light comes on?
 
#31 ·
Just curious with your electrical knowledge why you wouldn't just use like an Omron 12V on delay timer. A timer and mounting block would only be about $65 or so with shipping.

You turn on the ignition which starts the countdown and energizes the resistor relay. Once it times out, it energizes the contactor. I'm sure you have already thought of this I'm just wondering what your reasoning is to "reinvent the wheel" so to speak.

One relay I found.
http://www.newark.com/omron-industrial-automation/h3-4-12dc-30s/timer-on-delay-12vdc-30s/dp/07WX9125
 
#33 ·
Those relay contacts are rated for 30VDC max. You will have a very hard time finding an affordable relay with contacts rated over this. Since the current is so low, most people just use these relays anyway for KSI, with no problem, switching pack voltage. Another option that works with certain controllers (Curtis 1231, for example) is the tap 24V for use with the KSI signal, to stay within the relay ratings. This would not work with a Logisystems, though, so check the documentation for your controller.
 
#34 ·
RFEngineer. Interesting article, and I think your device is a good solution to the problem of switching on the contactor without enough precharge for people who want the HV completely isolated when parked up.

My 2pence worth is that I think people are worrying far too much about leaving the precharge on all the time. The current drawn from the traction pack is negligible, the available power is tiny, therefore in a fault condition, the vehicle is not going to go anywhere. And even if you do disconnect the precharge, High voltage remains on the controller input terminals for many hours after disconnect, so extreme care needs to be taken to safely discharge the caps before maintenance even with the HV disconnected.

Dont get me wrong, I think a power switch / kill switch is an extremely good idea for maintenance / emergencies, I just dont think they need to be used every time you park up. Just my opinion as an EE who designs drive controllers (But not in the EV power requirements)
 
#36 ·
... And even if you do disconnect the precharge, High voltage remains on the controller input terminals for many hours after disconnect, so extreme care needs to be taken to safely discharge the caps before maintenance even with the HV disconnected.
When I shut off my contactor and precharge relay the Voltage on the controller side drops from ~120VDC to less than 50VDC in under 30 seconds. That is with nothing connected to the input of the controller (other than a 250mA DC-DC converter running the digital Voltmeter) and just the motor on the output.
 
#41 ·
What I do is use a relay to close the resistor circuit in the key start position, then a time delay relay set up as latching-turn the key to start for 1-2 seconds, then the time delay relay closes after that 1-2 sec, latching itself and powering the main contactor and an indicator light. If the key is just turned to ignition the contactor will not close- it has to be first put in the start position for 1-2 seconds-if released to soon, power will not latch the time delay relay and the timer will start over. you must hold that start position 1-2 seconds.
This way here the precharge is only in the circuit for 1-2 seconds (enough to charge) and so power is not always at the controller.
 
#42 ·
Hi I am new to the forum, but I have been following this topic around the internet for quite some time. Currently my setup is not high voltage, but I would like to begin to think about wiring in a system as I gradually upgrade. I have a few questions/comments.

Precharging to my understanding is essentially done to keep your controller happy by not having a huge inrush current that could damage it. It also to my understanding helps eliminate some of the arcing on the contactor. Would that then suggest that if you had a (10W in my case) pre-charge resistor to get the controller up to 80-90% on the caps, you could then have a 2nd larger pre-charge resistor (say 500W) that would help reduce further/elimate arcing at the contactor. Is my thinking correct here?
 
#43 ·
Hi I am new to the forum, but I have been following this topic around the internet for quite some time. Currently my setup is not high voltage, but I would like to begin to think about wiring in a system as I gradually upgrade. I have a few questions/comments.

Precharging to my understanding is essentially done to keep your controller happy by not having a huge inrush current that could damage it.
The primary reason is to reduce the stress on the contactor when closing. The smaller the difference of potential across the contacts the less they will arc and the longer they will last. Also, since they are not pitted by arcing, the contacts are less likely to weld together when you try to break them under high current (emergency) conditions.

Inrush damage to capacitors et. al. should not be an issue with a properly designed controller.

It also to my understanding helps eliminate some of the arcing on the contactor. Would that then suggest that if you had a (10W in my case) pre-charge resistor to get the controller up to 80-90% on the caps, you could then have a 2nd larger pre-charge resistor (say 500W) that would help reduce further/elimate arcing at the contactor. Is my thinking correct here?
The peak current through the precharge resistor happens early in the process. As the caps in the controller charge up the voltage across the resistor drops and the current decreases. I don't see why you would need a second, larger, precharge resistor if the one that does the initial current limiting is large enough.

Joe
 
#44 ·
I am in agreement with Joe and would just like to add that the controller charge up happens very fast-less than a second (according to a curtis engineer), I say just use the recommended resistor and you will be fine. As far as removing it from the circuit after the controller is charged, that seems to be a personal choice and well discussed in this thread.
Mike
 
#47 ·
You can (carefully) connect a voltmeter across your precharge resistor and then watch the meter as you turn on the high-Voltage. You should see the meter immediately jump to your full pack-Voltage and then gradually drop. Note the time it takes for the Voltage to drop to an acceptably low level. This is how long you should wait before you close your contactor.

For example, My battery is about 130-Volts when fully charged. It takes about 15-seconds for the Voltage across my precharge resistor to drop below 60. I try not to close my contactor before this point.
 
#48 ·
RFEngineers, I like your circuit, but have one question. Are you actually checking that the HV traction pack voltage is present before you start your process? I ask this because often there is a circuit breaker or emergency off switch connected before your circuit. If these are off and your circuit goes through the processes... and then the HV is applied, no precharge will occur - Danger!
Regards Ric.
 
#49 ·
Ric,
You are correct, that would be a problem, which is why I never turn off my emergency disconnect. In fact, if I pull the emergency cable under my dash, I can't reapply power without opening the trunk and manually reattaching the Anderson connector.

To answer your question no, my current design does not check for the presence of HV, but that may be an option in the future.
 
#50 ·
A couple of questions:

Precharge resistor. Is this some kind of special resistor, or just an ordinary resistor? How many ohms?

Also won't the resistor decrease the voltage to the controller which may damage it? Or if it limits the amperes to much, the microprocessor in the controller might not start properly?
 
#51 ·
A couple of questions:

Precharge resistor. Is this some kind of special resistor, or just an ordinary resistor? How many ohms?
It's usually a power resistor. Mine is 1000-Ohms at 40-Watts. Ask your controller manufacturer for their recommendations as to Ohmage and power rating.

Also won't the resistor decrease the voltage to the controller which may damage it? Or if it limits the amperes to much, the microprocessor in the controller might not start properly?
In a word, no.
 
#53 ·
This Thread has been a wealth of info, thanks to everyone, but there are two questions left:

Do I need an emergency disconnect?
-The answer is a definite YES, but there are many methods, what are the recommendations on this?

Fuses, Circuit Breakers or both?
-Fuses are consumable, and I'd prefer to have something that can simply be reset, does anyone have any recommendations on what could be used for a breaker in a high current (1000+) application?
 
#54 · (Edited)
Precharging is certainly an important safety topic.
Leaving the controller energized can be a safety issue, as well as a fire hazard in case there is a failure when away from the car. I have heard of more than one Curtis catching on fire while charging (granted, the charging voltage was too high) when they had the precharge resistor connected all the time.
The other important reason for precharge is to insure that the controller has not failed "full on" while cooling off from being overheated (relatively common). If that happens and the precharge circuit turns the controller on despite it, then the car takes off uncontrolled.

Fifteen years ago I made a simple precharge controller for use with Curtis style controllers. You can see a schematic of it here:
http://cafeelectric.com/downloads/precharger_1.1.JPG

This circuit is very old, but it is where I started when I developed the integrated precharge for the Zilla so it is simple and lacks things.
For one, it has no way of insuring that your main disconnect is on before turning on the main contactor. Most people respond to that situation like this:
Hmm, car did not go but the contactor turned on...
Oh! Breaker is off!
Turns on breaker.
Bang! Damage is done to breaker contacts, possibly to controller as well.

If the breaker was tripped due to a controller failure you may find yourself dislocating the garage wall. A friend had the unfortunate luck of doing that when his Auburn controller blew on startup, I won't name him since I think he has still not told his wife what really happened.

Marginal controllers can be damaged when turned on without precharge, a Curtis engineer once told me it can damage their controllers, I've never seen it damage one of mine but since Y2K my designs won't start without precharge.

For those reasons it would be good to have a circuit that keeps the contactor from engaging if the HV pack is disconnected.

If anyone wants to play with this circuit, beware that I provide it only for entertainment use and needs many changes before it is safe. I never produced it since when I started to try to account for all the possible safety issues, I decided it needed to be integrated into the controller with the microcontrollers that watch each other for safety.

Also, as suggested by Lee Hart on the EVDL, put a 10 ohm 1/2 W Carbon comp resistor in series with the 400V cap. It helps protect the contacts at precharge initiation while still being low enough to allow the cap to snub 150V turn off.

As to what the circuit does:
HV inputs 1 and 2 are on the high voltage terminals of the contactor. As long as the voltage on those is over about 6 Volts (indicating the controller is not precharged, and sensed by the series circuit of the bridge, current source Jfet and optocoupler) then the fet that drives the contactor is held off.

Turning on the Key input turns on the precharge relay, which turns on the precharge resistor which is a PTC fuse. This allows for a fast precharge while also avoiding a blown precharge resistor in case of a controller fault.

If the controller has a fault, the PTC goes high resistance and the voltage stays on the main contactor contacts keeping the main contactor disabled.

D4 is just a contactor coil snubber diode.

This post is long enough, for my views on contactor safety you could visit my FAQ page at http://cafeelectric.com/Sfaq.php and click on the link titled "How do I wire two contactors for higher voltage use?". Sorry for the funky use of a blog for the FAQ, it seemed the easy way at the time.
________
marijuana vaporizers
 
#88 ·
Precharging is certainly an important safety topic...
Fifteen years ago I made a simple precharge controller for use with Curtis style controllers. You can see a schematic of it here:
http://cafeelectric.com/downloads/precharger_1.1.JPG
It's GREAT to see Otmar posting on this topic. He's the Grande Master of controller design! Read his post carefully, folks.

I like his circuit. It's not fail-safe, but it's pretty darned good. It properly senses voltage, and won't engage the main contactor if the controller fails to precharge. It is protected against reversed polarity (miswiring), is not sensitive to pack voltage, and doesn't need a relay to control the main contactor coil. It doesn't need a micro, and doesn't depend on assumptions about precharge times. Using a thermistor as the precharge resistor keeps it from burning up if the controller fails to precharge (as does a light bulb).

Improvements:

  • I'd use a better relay than the Radio Shack one at K5 (like the Schrack PT52B012B I mentioned in an earlier comment).
  • Add a 10 ohm resistor in series with C1.
  • I prefer a light bulb over the thermistor.
  • I'd connect the left lead of bridge D3 to the top of R4 instead of J1. This prevents current from flowing from pack to controller through D3-T1-D1 and U1 with the key off.
  • Move the upper lead of C2 from J3 to the collector (pin 4) of optocoupler U1. C1 should be about 0.2uF. This keeps the main contactor from pulling in for at least 20msec, to allow time for K5 to pull in and the precharge circuit to begin functioning.
  • Use a bidirectional 24v zener in place of D4 (1.5KE24CA etc). It lets the main contactor turn off much faster.
 
#55 ·
So let me get this all straight in my mind...
..
You are essentially saying is is better to have two separate contactor circuits (one to energize the second contactor, the pre-charge resiistor being part of the second contactor circuit)?
..
The first contactor would allow you to energize and de energize the circuit without all the inrush current issues with a cold Power-up. The second contactor, would then be closed to supply actual "go power".
..
In your schematics you show the contactors as switches (which is what they are, granted). But in your discussions you imply that a switch can be used ahead of the pre charge enabled contactor. Now correct me if I am wrong (I have been many times over the past half century) - but would not just about any DC rated switch vaporize when the power contactor closed.
..
This is why I am asking for (and offering ) clarification on this point.
..
We are talking about two physical contacors - right?
..
Dave Mason
 
#56 ·
So let me get this all straight in my mind...
..
You are essentially saying is is better to have two separate contactor circuits (one to energize the second contactor, the pre-charge resiistor being part of the second contactor circuit)?
If by 'contactor' you mean a large high current solenoid, no. See the explanation below for a clarification.
..
The first contactor would allow you to energize and de energize the circuit without all the inrush current issues with a cold Power-up. The second contactor, would then be closed to supply actual "go power".
The precharge relay would need to be rated for your traction pack voltage. The current requirements, on the other hand, would be minimal. The precharge current is limited by the precharge resistor. In fact, I have been using a 12-Volt automotive relay for this purpose for almost a year in my 120-Volt EV with out any problems. I do take the relay out every few months and inspect the contacts. They look fine.
..
In your schematics you show the contactors as switches (which is what they are, granted). But in your discussions you imply that a switch can be used ahead of the pre charge enabled contactor. Now correct me if I am wrong (I have been many times over the past half century) - but would not just about any DC rated switch vaporize when the power contactor closed.
A contactor is just a big relay with high current contacts. In automotive circles it is often referred to as a 'solenoid'. It all cases, it is just a switch controlled by a voltage.
Many people put a high current switch or circuit breaker in series with their traction pack so that they can turn the power off. Of course, that switch must be rated for the current and Voltage applied or it will, as you point out, vaporize.
We are talking about two physical contacors - right?
..
We are talking about one (relatively) low current relay to shunt the precharge current around one (relatively) high current contactor.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top