[Tesseract]

Is it possible to have regen with a Kostov motor and Zilla HV Z1K controller?

Nope. The hardware is not capable of it.

 

[CroDriver]

Nope. The hardware is not capable of it.

Motor or controller? Or both? I read somewhere that the Kostov can do regen

 

[Tesseract]

hardware = controller.

The Zilla is not capable of regen, but that's not unusual as very few controllers for series dc motors are capable of regen. The only one I am aware of is the Zapi H2 (or H3?), and it seems to blow up on a regular basis if regen is actually enabled.

The main reason why is that the series dc motor has what is called an unstable transfer function in regen, prone to either locking up the motor shaft or blowing up the controller if attempted. The flipside of this is why the motor will self-destruct if power is applied either when the motor is unloaded or locked-up.

 

[Guest]

Non interpole motors are the ones with regen problems and the controllers must be able to handle that incoming rush of amps. In a light weight vehicle the H3 can and does do just fine with regen and an interpole motor but if you try with a heavy vehicle and a standard series with no interpoles and advanced it will most likely blow both the motor and controller. The best one to use for the heavier vehicles is the H2 but that one is only built for 120 volts or less. Still that would be fine. The interpole motor should not be advanced if you intend to do regen. I will be trying my Zapi with regen as soon as I finish my Kostov and get my programmer from Zapi.

Pete

Zapi controllers are still available but have a lead time of a few months. Many have forgotten about them but they are still around. Mostly because they did not do regen properly but it was not the controller but the person building the vehicle and the type of motor. It was learned the hard way about standard series motors that were advanced for higher voltages and regen. They do not mix as when the regen function is active the motor is actually over advanced and can arc out. You want neutral timing for regen. Much easier on the motor and your motor will also have better torque but not as much on the high end. Use a transmission.

Pete

 

[esoneson]

CroDriver: There's more than one way to skin a cat. (american idiom)

http://www.northrim.net/wyanders/ev/regeneration.html

Eric

 

[Guest]

Very very nice. If you have the room this would be an excellent way to go. Nice skin-o-the cat.

Pete

 

[CroDriver]

Thanks for the great info guys!!!

 

[Bowser330]

Non interpole motors are the ones with regen problems and the controllers must be able to handle that incoming rush of amps. In a light weight vehicle the H3 can and does do just fine with regen and an interpole motor but if you try with a heavy vehicle and a standard series with no interpoles and advanced it will most likely blow both the motor and controller. The best one to use for the heavier vehicles is the H2 but that one is only built for 120 volts or less. Still that would be fine. The interpole motor should not be advanced if you intend to do regen. I will be trying my Zapi with regen as soon as I finish my Kostov and get my programmer from Zapi.

Pete

Zapi controllers are still available but have a lead time of a few months. Many have forgotten about them but they are still around. Mostly because they did not do regen properly but it was not the controller but the person building the vehicle and the type of motor. It was learned the hard way about standard series motors that were advanced for higher voltages and regen. They do not mix as when the regen function is active the motor is actually over advanced and can arc out. You want neutral timing for regen. Much easier on the motor and your motor will also have better torque but not as much on the high end. Use a transmission.

Pete

So the interpoled DC motor can work well with regen if the controller can handle it....what we need is a 400+V controller that can handle 2000A AND have the ability to Regen...

This would really close the gap between the AC & DC EV setups...

 

[Qer]

So the interpoled DC motor can work well with regen if the controller can handle it....what we need is a 400+V controller that can handle 2000A AND have the ability to Regen...

You're not asking for the impossible, just pretty damn close to it...

Pumping up the voltage from, say, 50-100 Volt to 400+ Volt won't be easy. To even have the slightest chance to pull it off you'd have to have a series/parallel-setup that switch to series in regen so you AT LEAST start at 100 Volt AND have a motor setup that can handle that insane pack voltage (in regen motor peak voltage = pack voltage) without blowing up.

On top of that, regen in series gets extremely complicated the more motor and battery voltage differs, so pushing the voltage that much will probably take you dangerously close to the point when the whole setup go totally wild and the current shoots up exponentially, most likely burning something up in the process (like the motor(s)...).

I think regen with a pack voltage of 300 might work with two motors, but it's a big might because there's simply too much that can go to hell.

 

[Bowser330]

You're not asking for the impossible, just pretty damn close to it...

Pumping up the voltage from, say, 50-100 Volt to 400+ Volt won't be easy. To even have the slightest chance to pull it off you'd have to have a series/parallel-setup that switch to series in regen so you AT LEAST start at 100 Volt AND have a motor setup that can handle that insane pack voltage (in regen motor peak voltage = pack voltage) without blowing up.

On top of that, regen in series gets extremely complicated the more motor and battery voltage differs, so pushing the voltage that much will probably take you dangerously close to the point when the whole setup go totally wild and the current shoots up exponentially, most likely burning something up in the process (like the motor(s)...).

I think regen with a pack voltage of 300 might work with two motors, but it's a big might because there's simply too much that can go to hell.

Understood...how much does regen help range anyway...everyone says about 10% but thats sounds like a pretty rough estimate since a stop-n-go traffic driver would see a lot more regen than a freeway cruiser with no traffic driver...

 

[Tesseract]

So the interpoled DC motor can work well with regen if the controller can handle it....what we need is a 400+V controller that can handle 2000A AND have the ability to Regen...

This would really close the gap between the AC & DC EV setups...

What gap? You mean how a series (or sepex) dc motor is capable of much higher torque at very low RPMs compared to even the most sophisticated VFD and high-slip induction motor? *That* gap???

Anyway, how valuable do you folks perceive regen to be? 'Cause the facts are not in regen's favor EXCEPT in the ac system where you get it for free. I mean, the power has to go through all of these steps:

1. switchmode charger efficiency: 94%
2. charging efficiency of LFP batteries: ~95%
3. discharge efficiency of LFP batteries: ~95%
4. controller efficiency in motoring mode: 98%
5. motor efficiency in motoring mode: 85%
6. motor efficiency in regenerating mode: 85%
7. controller efficiency in regenerating mode: 98%
8. charging efficiency of LFP batteries: ~95%

Multiply all of that together and you get an overall efficiency for regen of 56%, and that's using LFPs and a PFC charger. Lead-acid batteries have a charge/discharge efficiency of ~70%, which causes regen efficiency to plummet to approximately 22%. Oh, and then you multiply that efficiency number by the total amount of time spent in regeneration to get the net amount of power recovered. Say in heavy traffic city driving you spend 50% of the time braking, well, that's a stunning 11% of power recovered with Pb batts or 28% with LFP and we all know I am being overly generous with that time spent braking there... Realistically, the off-the-cuff assessment regen gets you 5% more range is probably about right.

Now, when I said that the transfer function for a series motor is unstable in regen, that applied to neutrally timed motors with or without interpoles, too. Also, regen tends to suck for dc motors in general (sepex included) because back EMF is proportional to shaft rpm, so as the motor slows more and more current needs to be drawn to feed the boost converter so that the output voltage is higher than the pack voltage. Drawing more amps means braking torque goes up. Worse, the instability in a series dc motor tends to compound this behavior, which is probably why those controllers that have managed regen with series dc motors behave somewhat "jerky".

Gottdi, you've heard the Zapi's work fine in regen whereas I have heard nothing but the opposite and yes, this is with neutrally timed motors. Pretty much everyone says DON'T enable regen with a Zapi. However, I wholeheartedly encourage you to do so and then report the results back to us.

 

[Guest]

Do not enable Regen with the Zapi H3 on a large vehicle going 65 mph. The Zapi can limit the feed back amps to limit the power generated and help keep every thing alive. The H2 on the other hand CAN handle regen but you really need a neutral timed and interpoled motor to do that. Other wise you risk damage to both the motor and controller. The benefit of REGEN is not just some back into pack but to help slow the heavy vehicle down. I think it is worth the benefit. AC still needs to limit regen feedback to the batteries too.

Yes there is loads of information that says steer clear of REGEN in series motors and I agree if you are using an advanced non interpoled motor and have no way to limit your regen. There is also lots of rampant speculation that says steer clear too and none with any validity except that someone heard that someones controller blew because of regen and has no clue that the problem most likely was due to an improper setup vs just REGEN.

I listen but I do research too. The main reason most do not do regen is because very few controllers provide for that function. That is the reason and some excuses must be generated to show they are better so over time REGEN became a bad thing and NEEDING MASSIVE BRAKES to slow down your very heavy vehicle has become very good. I agree that regen is not always needed but I for one would sure like the benefit of the stopping abilities of REGEN. I'd also like the little I can dump back in from time to time too. Many of us actually do drive in stop and go traffic and must drive like most always drive. Zip to the stop and then stop. What a waste but a fact of life. Regen can help in that situation too. 10% is better than 0%

Pete

 

[Qer]

The main reason most do not do regen is because very few controllers provide for that function. That is the reason and some excuses must be generated to show they are better so over time REGEN became a bad thing and NEEDING MASSIVE BRAKES to slow down your very heavy vehicle has become very good.

You're reading Tesseracts comment as the devil reads the bible.

The thing with regen is that unless you're doing a lot of stop and go or live in a hilly environment, there's very little point in using regen at all. It's not an excuse, it's just a fact that regen very seldom makes that much of a difference for the range.

In an AC-system regen is more or less free since you "only" have to move timing a bit. In a PM or SepEx motor the magnetic field is (or can be) rather static so it's simple to control the amount of current you get from the motor by controlling PWM, even though you need a second transistor to do it. At high RPM the duty cycle can be kept low, at low RPM the duty cycle has to be increased. No rocket science. It will suck at low RPM's due to back-EMF, but it's still just physical facts and no excuse.

Now, series wound motors are a bit special. If there's no magnetic field, there's no current and thus there's no magnetic field, repeat, rinse. This differs from PM and SepEx so to start regeneration you need to introduce a magnetic field. When you get a magnetic field, you get current, which gives you a magnetic field etc. If you want more current (and thus adjust the PWM) you get more magnetic field as well and suddenly you don't just get more current, you get MORE CURRENT and if you're not careful you get MORE CURRENT!!! OMG!! WTF!!1!11!??! *BOOM*

That's why regen in series wound motors is hard. The controller has to be very good at reacting fast and accurate, otherwise it'll lock up the motor, burn itself to death or simply lose the magnetic field because the reaction is not linear (as in other motors), it's exponential and that's why so few controllers can handle it. Still no excuse, still a physical fact.

If you so desperately want regen a series wound motor is not the optimal choice, simple as that.

 

[caramelzappa]

I'd just like to pipe in and say that range isn't the only reason to want regen. Batteries add a considerable weight to the car, and it's a lot easier on the brakes if you're using regen to slow down as much as possible.

 

[Tesseract]

I'd just like to pipe in and say that range isn't the only reason to want regen. Batteries add a considerable weight to the car, and it's a lot easier on the brakes if you're using regen to slow down as much as possible.

Yeah, but it's a lot harder on the motor so what would you rather wear down faster - the brake pads or the commutator and brushes??? I'll take brake pads any day of the week.

Otherwise, the merits/demerits of regen are covered in the Wiki:

http://www.diyelectriccar.com/forums/showthread.php?t=8848

This is an oft-repeated discussion. We should instead try to concentrate on HOW to implement regen in a series wound motor rather than go off on purely subjective tangents like whether it is worth it or not.

We - that is, evnetics - are going to attempt regen with our controller. We bought both 11" and 9" Kostov's and the hardware is intrinsically capable of flipping from Buck to Boost conversion modes (hopefully before the field collapses, otherwise we'll need to "tickle" the field externally).

At this point, though, it is being approached strictly as an experiment and even if it does work we are EXTREMELY reluctant to make regen a user-programmable feature as it will rapidly destroy an advanced brush timing motor like the most popular WarP and ADC brands.

gottdi - AFAIK, the Zapis are either impossible (or, at least, very difficult) to obtain in the US, so it's kind of a moot point how well or poorly they implement regen with the series motor, n'est-ce pas?

 

[Guest]

EXTREMELY reluctant to make regen a user-programmable feature as it will rapidly destroy an advanced brush timing motor like the most popular WarP and ADC brands.

I think I covered this. DO NOT USE MOTORS WITH ADVANCED BRUSHES PERIOD. Neutral timed AND interpoled if you plan on using a series motor. Kostov is a good choice. They were designed from the beginning for regen. (Forklifts and Industrial Vehicles)

I am very aware the Zapi is hard to come by but so is the Zilla now hard to get. Zapi has an outlet back east and the lead time is like 6 to 8 weeks or so. Not unlike some we get here. You just have to know where to go. The controllers that do regen WILL limit the field current so not to create a cascade effect.

Since you can still get new Zapi controllers it is not a MOOT point. Just not as practical as lets say an underpowered. H2 will do 600 amps solid and the H3 will do 800 and modified H3 will do 144 volts and 1000 amps. Mine has been modified. I plan on using it as a straight controller when I get the programmer.

I will be very interested in the results of your testing and implementation of regen in your controller. Let me know when you have that up and running. Since my Kostov is not in running mode and I don't have my programmer yet I can not do this my self. I am trying to secure an H2 and Shunt motor for Regen testing in my Ghia. If successful at getting these I will get it going right away. I am very wiling to give it a spin in a real car on the road. I know it can be done. Gaining power back for my pack is not the reason and I will not use a motor with advanced series brushes.

Pete

PS. Yes you need to set things up correctly first.
PSS. Since I do not have or willing to spend $10K on a simple AC motor and Controller just to have regen I would like to do it with DC. However impractical all here say it is.

 

[Bowser330]

I think I covered this. DO NOT USE MOTORS WITH ADVANCED BRUSHES PERIOD. Neutral timed AND interpoled if you plan on using a series motor. Kostov is a good choice. They were designed from the beginning for regen. (Forklifts and Industrial Vehicles)

I took Tess's comment to mean he didnt want to make it user-programmable, because so many people ask the forums if they can do regen with just about any motor they find...and time after time the knowlegable ones have to reply and help them understand...

Gottdi, you know as well as i do that there will be several people who will buy the Evnetics controller, buy a motor that does not apply, incorrectly turn on Regen (if its user programmable), blow it up and then come back and try to get thier money back or whatever.

I hope the research produces great results and that they can include regen, and make it locked and only able to turn on if the customer signs some type of agreement that states that if Evnetics can prove that the failure of your controller was due to activating regen on an unqualified motor, that the warranty is cancelled. If they cannot then they have to repair or replace, etc, etc. whatever the normal warranty jargon is...just a thought to protect both parties...

 

[kittydog42]

I don't really understand why anyone would manufacture a series motor controller with regeneration. It is asking for trouble. They will be judged by the failure rate and how they react to it, not by whose fault that the failures are. The fact is that motors with interpoles are harder to come by means that people are more likely to try regeneration with neutrally timed WarP and ADC motors, and the end result would probably not be good. Also, like others have said, the main benefit to regeneration is really the natural braking, as the actual return of energy is very negligable.

DC conversions are a cost-effective was to get into a conversion. AC conversions exist for those who want the fanciness. In my opinion, regeneration with series DC is not a good allocation of resources. Maybe I will be proven wrong by someone who really wants to do it, and that would be great - I like the braking feel of regeneration.

 

[Tesseract]

I think I covered this....

I thought I did too... why are you yelling at me?!?

I took Tess's comment to mean he didnt want to make it user-programmable..

Gottdi, you know as well as i do that there will be several people who will buy the Evnetics controller, buy a motor that does not apply, incorrectly turn on Regen (if its user programmable), blow it up and then come back and try to get thier money back or whatever.

Well, at least somebody is paying attention here...

And that is pretty much the exact scenario we wish to avoid, except we expect it will be the motor that blows up, not the controller.

Also, like others have said, the main benefit to regeneration is really the natural braking, as the actual return of energy is very negligable.

Yeah, I would be one of those people that has publicly expressed both sentiments...

Just in case it wasn't clear the first time I wrote it... we are going to *try* to get regen to work with a motor that has neutral brush timing and interpoles. We are mainly curious to see if it is possible to get regen to work with a series motor without resorting to a Rube Goldberg like contraption, not so much because we want to inflict heaps of misery and tech support requests on ourselves in the future.

IF we get regen to work AND we decide to release a version of the code with it activated (which, yes, you will have to contractually agree to hold us harmless for any damages to your motor, etc..) THEN the motor current will be strictly limited to prevent excessive braking torque and provide a more or less constant "engine braking" like feel.