Serial Hybrid Project

[JCG]

Hi all, after posting for a while in the ebike area regarding a through-the-road parallel electric bike with an ultracapacitor storage buffer (http://www.endless-sphere.com/forums/viewtopic.php?f=6&t=7511), it's time to start floating information and calling on advice and ideas regarding a longer-term project that I hope to kick into high gear this summer.

I've written a bit about this in other places, but a few years before I took my current position, my University was donated an EV-1 electric vehicle by General Motors. We have been trying very hard to get a place to work on the car, and may finally have caught a break for starting work this summer. The project I'm overseeing has a limited amount of funding, but it centers on converting the EV-1 into a serial hybrid with an ultracapacitor bank as its electric storage buffer. The main (average level) power source will be a genset powered by a to-be-determined fuel - likely biodiesel, which we are going to make by recovering waste vegetable oil from campus dining services. I am also considering ammonia fuel. In general, the energy flow plan is:

Liquid fuel --> Generator set --> Rectifier --> Ultracap Bank (389 V, 20.6 F) <--> AC motor controller <--> 3-ph, 103 kW motor

In order to get students interested in the project while our garage facility is prepared, I have used existing supplies (purchased and scavenged) to make a model of the proposed series drivetrain. Here is an introduction to the scaled setup.

The genset, when in operation, will provide either 1 or 3 phase AC power which must be rectified to DC. I have saved a 220V, 1.7 kVA variac from the junk heap to serve as this source of AC. It will deliver single phase AC in a voltage range of 0 to 220 VAC, RMS (the outlet's voltage level). I have connected its outputs (white and orange) to a 20 A rectifier, mounted on a large aluminum plate (heat sink). It provides DC output, which will give up to about 311 VDC when connected to a capacitive load.

Here is the capacitor bank. Six 48.6 V, 165 F modules connected in series. I have only shocked myself three times.

Here is the motor controller. It is a Hitachi X200 series, which will power a 1 hp, three phase motor.

A closeup of the controller's input and output terminals shows why I selected this controller. From left to right, you have: 1) ground, 2) output phase 1, 3) output phase 2, 4) output phase 3, 5) - DC bus access, 6) + DC bus access.

Accessing the DC bus allows me to skip the controller's inverter and supply the controller with direct DC power. The ports are there for attaching a dynamic (resistor) brake, but I will use it to both power the motor and allow regen energy to flow back in to the capacitor bank. Here is the 1 hp (746 W) motor, with a 14 inch diameter cast iron v-belt pulley as a flywheel to store rotational energy:

The motor controller lets me use a potentiometer to control the motor's speed by adjusting the input frequency. 60 Hz provides an axle rotation rate of 1725 rpm, according to the nameplate. I charge the caps up to 260 V, which is a voltage in the operating range of the controller (190 to 395 VDC):

Here is a video of spinup to full speed (manual) and braking at about 15 Hz/s (programmed into the controller).

[youtube]FTfGhJ6LOXU[/youtube]

The table is not connected perfectly, so I get some shaking and hamonic vibration at about 25 Hz. The motor current for the same operation:

[youtube]NiM4e3-nMoU[/youtube]

The negative currents signify power heading back into the cap bank. I'm also using a ceramic heater element (cracked but still operational) to discharge the caps at the end of the day. It's rated at 1000 W.

I'm mostly interested in finding out more about gensets... the car will need one of 25-35 kWe rating to fit in the trunk. I'd appreciate any ideas or feedback, and of course I'd welcome any questions or comments.

 

[kent1956]

Great idea. I'd love to repeat it here, if I could get parts donated to do the project. I know the variac and ultracaps are not cheap!

For now I will follow your ongoing project.

 

[Jeremy Harris]

The genset power you've quoted seems pretty high to me. Most saloon cars have an average power requirement of only around 12 to 20hp (9 to 13.5kW). If your capacitor bank, controller and motor was sized to provide all of the peak power, then the genset only needs to provide the average power.

One of the problems with current hybrids (in my view, as an ex-owner of one), is that they don't have enough storage capacity, so need bigger IC engines. Ideally, the IC engine should be as small as possible, as that way it can be run at high load most of the time, where it's specific fuel consumption (and hence efficiency) will be at it's best. Part throttle operation always hits efficiency hard.

FWIW, my Prius managed to run for 60,000 miles using an average power from the IC engine of just 10.5kW. My wife's small Citroen diesel is averaging out at around 9kW so far. My guess is that the EV1 may be a little higher than these average figures, but not by much.

Jeremy

 

[JCG]

The genset power you've quoted seems pretty high to me. Most saloon cars have an average power requirement of only around 12 to 20hp (9 to 13.5kW). If your capacitor bank, controller and motor was sized to provide all of the peak power, then the genset only needs to provide the average power.

Hi Jeremy, thanks for those power numbers - I'm trying to collect that kind of data for cars already on the road. Data sheets I've reviewed for the EV-1 show an average power draw of less than 10 kWe when cruising at 97 kph (60 mph) at zero grade. A previous series conversion used a small but powerful gas turbine to produce 40 kW: http://www.autoworld.com/news/GMC/Series_Hybrid.htm You'll of course want to provide enough total power for cruising or moderate acceleration while charging the caps, which would put the practical genset range up to at least 20 kW - just in case a hill needs to be climbed (maybe even a long one on the highway).

Some progress is being made on the generator front - I'm working with a colleague of mine to allow the vehicle to be one of the "test beds" for a 2.5 L, 2-cylinder spark ignition engine he's developing to run on direct ammonia. It'll be coupled to either a variable speed alternator or starter-alternator. The target is 30 kWe delivered from 2-cylinder operation, and about 13 kWe from one cylinder (if the other is shut down by the engine controller). He's getting high compression ratios (20:1) which brings it up to diesel efficiency levels. An eventual target with ammonia is 30:1, but we'll see about that. I like the idea of running on one cylinder most of the time, and kicking in the second if power demands rise. Switching cylinders on and off definitely hurts efficiency less than throttling.

With any luck, the total vehicle weight would be a little less than when it was filled with lead-acid batteries... I hope that the trunk is big enough for the ICE (only 0.3 cubic meters / 10 cubic feet).

 

[Xing]

I love your project and read your lips since the first posting. And I am very jealous about your impressive capacitor pack.

I'm also using a ceramic heater element (cracked but still operational) to discharge the caps at the end of the day.

Come on and let the beast out. Never felt tempted to put a t-beam over the connectors to make a firework? Put THAT on youtube.com.

 

[paultrafalgar]

....
- I'm working with a colleague of mine to allow the vehicle to be one of the "test beds" for a 2.5 L, 2-cylinder spark ignition engine he's developing to run on direct ammonia.)

I wonder if this connection may be relevant to your work - a system for storing ammonia:
http://www.amminex.net/

 

[JCG]

I wonder if this connection may be relevant to your work - a system for storing ammonia:
http://www.amminex.net/

Hey Paul - yeah, their CTO (Tue Johannessen) and I have been at a couple of ammonia fuel conferences together. The storage medium is a magnesium chloride salt that soaks up ammonia, you only need to heat it to get the ammonia back out.

For now, the cheapest (and in my opinion best) way to store it is still as a compressed liquid in a tank at about 150 psi. We'll likely use propane-type tanks for gaseous NH3 to the combustion engine. The quote for the genset is on its way, and this should be a fun summer working with the EV-1.

 

[DISH]

Come on and let the beast out. Never felt tempted to put a t-beam over the connectors to make a firework? Put THAT on youtube.com.

Buss bar across bank of 2600F Ultra-Cap that I've been playing with.

 

[paultrafalgar]

I wonder if this connection may be relevant to your work - a system for storing ammonia:
http://www.amminex.net/

Hey Paul - yeah, their CTO (Tue Johannessen) and I have been at a couple of ammonia fuel conferences together. The storage medium is a magnesium chloride salt that soaks up ammonia, you only need to heat it to get the ammonia back out.

For now, the cheapest (and in my opinion best) way to store it is still as a compressed liquid in a tank at about 150 psi. We'll likely use propane-type tanks for gaseous NH3 to the combustion engine. The quote for the genset is on its way, and this should be a fun summer working with the EV-1.

I wonder if Calcium Chloride would work because iirc it's an industrial waste product (from which process i don't recall).