Sunday, December 28, 2008

Building the heater

Money has been tight after ordering the batteries and the holidays so progress has been rather slow lately. This is technically the second heater but I will only fully document the final design.

I first built a water heater utilizing the existing heater core. It used a 1500 watt engine block heater and pump to circulate the water. I decided not to use this setup for a few reasons. First and least important was the noise I could hear from the pump circulating the water. A few people recommended a pump that was very quiet but after further thought I didn't try it out. The second reason was the amount of heat I noticed on the outside of the heater unit itself. I figured if all that heat was on the outside it was wasted energy especially while driving if air was passing over the heater. The hoses also get very hot on the outside which again means energy lose. Last I figured this gave me a fluid that could leak, and therefore something that needed to be checked from time to time.

So, as many others have done, I went for the ceramic heater setup which replaces the water heater core with an electric version. Mine, however, well be quite different from the typical basic ceramic install as you'll see later.

The first part in building a ceramic heater is removing the heater core and this is not an easy task, especially with newer vehicles. As I've mentioned before, get the service manual for your vehicle as it will really save you time during your build.

I won't go into boring details of the dash removal but figured I'd still post a few pictures to give you an idea.

Here is a pile of plastic trim, air bags, steering column, etc which is well over a foot tall. There are a few things in the picture to give some perspective. Of course also not shown is the full dash itself.


Here is a picture after removing the dash. I thought I was home free at this point. I was wrong. From the middle to the passenger side you can see the heater box, blower assembly and condenser unit. The blower and condenser needed to also be removed to get to the heater box. This really wasn't that much more work but meant the refrigerant needed to be evacuated. I was hoping to avoid that.



A slightly closer picture with the heater box finally out! Look at all those wires! Most of them are attached to the dash so this is really nothing.



At last the heater core itself is removed.



Here is a shot of the heater core itself. It's a lot skinnier than I thought it would be.



OK on to building the new heater core. I purchased this ceramic heater from Kmart for $30. It was the cheapest ceramic heater they had. First I purchased a $20 heater which was not ceramic and just used five coils of wire. I doubt these would hold up well with the vibrations of a car. It was so cold in the garage I just kept it for a shop heater :)


Here I am testing the temperature coming out of the heater. 231 Fahrenheit, hopefully that will do the trick on the cold mornings. It rarely gets below 40 where I live so this should do just fine.



Now for a test to see how much power this heater really draws. It bounced around quite a bit while running but it's drawing right around 1500 watts as advertised.



Here is the ceramic heater element itself.



Here is the first place my heater will be different than most others. Instead of destroying my original heater core which I hope to sell to get some cash back I'll build my own. I'm using aluminum 1/8" thick for all the pieces. The main frame is 1" square tubing.


Below I've cut out a recess for the heater to set in with my plasma cutter. It was cut slightly higher and angled in to allow clearance for the outer terminals.



I messed around to see if I could weld this assembly but aluminum is a huge pain to weld with a MIG welder. To be honest I was suppose to use 100% argon gas but was only trying it with my 75% argon 25% CO2 mix so it failed horribly. I couldn't afford to get another cylinder for the argon gas so fell back on self tapping screws to hold it together.


Here you can see on the left a piece of flat laid across the side and end pieces using the self tapping screws to hold it together.



Here is one side assembled. The screws go into the side pieces and the end pieces.



I drilled a 1/2" hole on the top for the wires and put a couple of grommets in there to protect the wires from the rough metal holes.



Now I need to make some plates to fill in the gaps where the ceramic heater doesn't cover. I used a piece of metal clamped down to the plating I was cutting to act as a guide for my plasma cutter.



Here is the bottom plate installed.



It's hard to see here but I've cut some slits into the side pieces to allow me to bend them in. This will give a solid piece on the sides a good distance away from the ceramic heater element to allow about a 1/4" of silicone.



I originally cut out small pieces of plastic (from a coolant reservoir I used on the water heater setup) to keep the ceramic heater from touching any metal. This is very important not to short out the heater element to the frame. However, the plastic couldn't withstand the high temperatures if the airflow was too low so I removed these and replaced the gap with more silicone.


I used a high temp (650 Fahrenheit) silicone gasket sealant putting about 1/4" on both sides and bottom. I then cut and installed the top plate. Both of these plates will force all the air to pass through the heater element and not be able to bypass it.



Here is a side-by-side of the water and electric heater cores.



Next I added a foam stripping to all sides and the top. This not only keeps the element from being able to move a little inside the heater box but adds a little more sealing to keep air moving through the element and not around it.



Here is the completed heater being installed back into the heater box. Now that the new core was complete I also did a quick continuity check with my multimeter to make sure that none of the elements were grounded to the frame.

Here is a shot in the housing with a wiring harness built for it. There are five seperate 12 gauge wires inside the harness coming from each section of the element.
Later I will show the electronics and wiring side of the heater. The five wires actually allows for more than the two heater settings (low/high) that they give you and I will be taking advantage of this to allow for more levels of heating. This will not be done, however, through any switches. I will use the existing hot/cold knob and build a circuit to detect changes to this and convert that into a desired temperature setting. Then depending on the temperature differential I will determine how much heat to output. This should end up a lot better than a simple on/off switch on the dash and will use less power when only a little heat is needed without completing shutting down which would then be shooting cooler air out the vents.

I reinstalled the dash and did a quick AC test with the heater. My garage was around 47 F and so was the inside of the car when I started. I was able to get the car up to 80 F. I also measured the vent temperature and I was getting 115 F. I compared this to my truck after it heated up which was putting out 140 F. I'm not sure what my original temp would have been in the car but figured this was at least some reference point. In addition I'll be running the heater on 144v DC so it will actually get hotter than the AC test.

Sunday, November 23, 2008

Batteries ordered

After months of debating which batteries to buy the deed is done and there is no going back. I originally started this build with the idea of using the typical LA(Lead Acid) battery as the power source. After doing the math and seeing what other EVs range seemed to be it didn't seem LA could really get the range I was looking for, at least not without making my car weigh in a 4,000 lbs. I ordered 45 of the ThunderSky LFP-160Ah modules at $264 a pop, not cheap. On top of that the charger and BMS system added another $3,000. The full setup was ordered from Elite Power Solutions.

The amount of money up front for this setup is scary but after doing the math the investment should be well worth it in the long run. First off I can get about four times the capacity out of them as LA pound for pound. The setup I ordered will add 550 lbs. of lithium to the car which should keep the overall weight of the build at or maybe slightly below the original weight. This compared to most EV conversions needing the suspension beefed up to support the extra weight. Lighter will also mean faster acceleration than with the LA and also less power required to accelerate.

A standard LA battery last around 200 charges. Some of the higher end cells can go as much as 400 cycles. Cycle life of the lithium cells start at 1000 and go as high as 3000 cycles for 70% DOD. If I used flooded LA and got 200 cycles out of them I would still break even if I only got 800 cycles out of the lithium cells so I shouldn't have any trouble getting my moneys worth if the cells hold up....we'll see.

Hurry up and wait. So now that I finally committed to this purchase I get to wait for 5 to 9 weeks for them to arrive. UGH. I'll take this time to work on the electric systems of the car and hopefully have things ready to go when the cells arrive.

Sunday, November 16, 2008

12v Test Video

It's just too cool to see the car moving that words can't describe. I got my neighbor to video as I backed it out of my driveway up the street a bit and then back in. I edited the video to shorten it up since at 12volts it wasn't moving very fast if there was much of an incline.


video

Sunday, November 9, 2008

Installing the motor

After taking a nice vacation to Oregon for some quad riding in the dunes and getting over a nasty cold I was finally able to get back to the project. First I needed to hoist the motor and transmission down into the car so I could start measuring things and figuring out how to build the motor mount. I knew this would be a two man job even with the hoist so I grabbed my neighbor again for some help. The top center bolt of the adapter plate to transmission turned out to be perfectly balanced for the setup.

Here is another closer shot to show what we have to work with. You can see the original two motor mounts that we need to eventually tie into. Still plenty of room in here for other things later such as batteries.

Here is a top down view just to show another angle. My neighbor had a good idea of using this strap to support the motor weight so the car could easily be moved around if needed.

A quick note about the rotation of the motor. It turns out that all Honda engines do not rotate counter clockwise like I read. At least the S2000 rotates clockwise. This effects where the front motor mount holes need to be so make sure your motor is setup correctly before making your motor mount.

Let the work begin. Here I am making my first cut of the project.

Here is one of the brackets made to set against the adapter plate and utilize those bolts to build a frame around the motor. A plasma cutter makes these custom corners a breeze. The metal used here was 3/16" x 3" flat bar.

That same mount shown in place.
Next I welded 1/4" x 1 1/2" angle to the plate.
I made another plate for the other side of the motor. These couldn't set at the same level on this side and are slightly higher overall because the closest bolt to the center is below the angle iron in this shot and can't be seen.
Here is the finished support frame. The front bracket was also 3/16" x 3" flat. The center was notched out for the secondary output shaft of the motor. The two ears were added to bolt to the front of the motor. We'll see how all this fits a bit later.
Finally on the frame assembly two holes have been drilled on each side to mount the brackets that attach to the old motor mounts. These must be bolted on and NOT welded because the motor mount bolts are at an angle and the motor will not drop in properly if it's all welded together.
Here is one of the two motor mount adapters. It's hard to see here but there is a bend to metal to transmission from the angle of the motor mounts to the level frame support. The larger center hole on the left is a 1/2" hole and existing motor mount bolt goes here. The slightly offset hole from there is an alignment hole where the motor mount has a tab sticking up. The two smaller holes are for the mounting to the frame support. I used 3/8" bolts here and drilled the holes one size up from there. The most important feature of these adapters is the small strips of metal on each side. They create a triangular shape and add a great amount of support against the weight that will be pushing down on them.
Here is a quick shot of the whole assembly before we take it out for paint. It's much easier to the angle of the mount adapters in this shot. At this point I've probably installed and removed pieces of this a dozen times as the measurements were made and it was all built.

Here is the whole assembly again after a primer, two coats of semi flat black and one coat of clear.

Here is a shot from the back side of the assembly.

Here are those two ears I was talking about earlier. These are the only two spots on the motor. Depending on how you build your plate and fasten it to the motor these two bolts could end up anywhere. You do have the option of rotating them in 90 degree increments though by rotating the motor on the adapter plate. I didn't feel like taking everything back apart and didn't see much benefit from a rotation so I worked with what I had.
Here we have the completed and installed motor mount from the right side of the car. The largest bolt on the plate connecting to the adapter plate originally called for 47 foot pounds and that was reused here. The motor mount nuts were torqued to original spec of 40 lb/ft and the newly added 3/8" bolts were torqued to 35 lb/ft.

Here is a shot from the right side of the car. You can see here that bolt is just below the angle iron and a tight fit. Make sure you give yourself enough clearance to get a socket in there. Also I mentioned earlier this side had to set slightly higher. You can't really see it here but I crafted two 1/4" x 1" x 3" spacers to set between the frame and the mount adapters to make up the 1/2" difference.
Here is a shot of the entire assembly installed.
With this done I finished installing a few last things on the transmission (shifter, console, etc). Next I jacked up the rear of the car and put the transmission into 1st gear. I then tried the old 12 volt test again and the back wheels started spinning forward. Make sure to keep an ear out for any strange noises or vibrations. Things should be fairly quiet or something is wrong.
So then I couldn't resist. I lowered the rear of the car and ran the motor again. I was amazed at the torque from only 12v. The car immediately lunged forward and begin moving fairly quickly. I'm pretty excited at this point as things are slowly falling into place.

Friday, October 17, 2008

12 volt test

I started off by removing the bolt that holds the coupler and putting a drop of loctite to make sure it doesn't come out. Even if it did the coupler had to be pressed on and I can't imagine it will ever move again, at least not without using a puller.

Then I attached the spacers and adapter plate to the motor. Again using some loctite and torqued these bolts to 40 ft lbs. With these bolts, and despite them being recessed were still very close the flywheel so I shortened the heads by .07" each using a bench grinder. I didn't want to risk them rubbing after perhaps the motor and transmission warmed up.

Next the flywheel was bolted on. The original specs called for 90 ft lbs of torque but I dropped this down to 70 with a small drop of loctite. The original crankshaft this was on was a hardened steel. The steel we are using is considered soft and I didn't want to risk pushing the material too hard and strip out a hole on something that took so long to make.

We can now attach the clutch and clutch housing. This part can be tricky if you've never changed out a clutch. You'd think they'd design these different to avoid the problem but they don't. The problem is the clutch itself needs to be centered with the housing, and therefore the transmission input shaft. If you simply tighten all the bolts the clutch will probably be too low (gravity) and you'll never be able to push the transmission and motor together. Usually you can just buy an alignment tool that you slide in while you torque the bolts. I didn't have one so made one out of a 17mm socket and some electric tape. It wasn't perfect but allowed me to slightly move the socket on an extension until I could see it was aligned before tightening the bolts. These bolts called for 19 ft lbs and again I used just a small drop of loctite.

Next came the hard part. The two guide pin holes we drilled are very slightly off and it requires some elbow grease to get them started each time. After I finally got it together I realized the clutch arm had fallen out and I had to start over. This is a reverse clutch and it actually pulls on the clutch instead of pushing into it to release. It makes hooking the clutch arm into place tricky and has to be done at just the right time while you're mating the motor and transmission.

Now the grand finale! I was so nervous to apply the 12v to the motor afraid I'd hear some rubbing, clanking, or just see the whole unit vibrating badly. All of this of course unfounded since I witnessed all the machining work. So I wired everything up and then touch the final wire to the battery. Luckily the transmission cross brace was still attached because that's what kept the whole unit from falling over as the initial torque kicked in. The motor always instantly reached its top speed for 12v and output shaft of the transmission was spinning happily along. Very quiet, very smooth...woohoo it works!

Of course the initial test wasn't enough as I was just too excited. I had to go get my wife, bang on my neighbors doors (he helped me strip the car down originally), call my uncle, try out a few of the gears just to see the output shaft spin at different speeds.

Here is a quick shot of the completed assembly after the 12v test.


Another shot so you can see another angle.


Next comes hoisting the motor/transmission into place in the car and taking measurements for the motor mount that needs to be built next.

Wednesday, October 15, 2008

Adapter plate and coupler complete

I finally got some time from my uncle to machine the parts and it took longer than we anticipated. Engineering as you go and a lot of measuring twice before you cut really added to the time but I think we'll end up with a reliable setup.

I loaded up the truck with everything I could think I needed and headed to my Uncle's house.


First was to create the coupler. The coupler needed to be 3.125" OD with a 1.125" ID with a .25" key. Here is a picture of the raw metal before we began to work with it.



A shot of the metal slowly getting turned down to the diameter we need.


It's starting to take shape. You can see the end is a little less than 2" wide and about .200" deep if I recall. We did this so the coupler could sit flush against the motor bearing and wouldn't be able to work its way in.


Here we are cutting the coupler slightly longer than we want it. This was the only step of the lathing process that had enough friction to require a lubricant.


Now we are slowly shaving off the end to bring it down to the exact length we need.

After the basic coupler was shaped out we broached out the key way using a 40 ton press. This was so cool to watch I forgot to take pictures.

Next we needed to drill and tap the holes. We used the flywheel as the template and a drill bit that was the same size as the hole to make an initial mark before moving down to the correct bit. To use the existing bolts you'll need an 11mm drill bit and a 12mm - 1mm pitch tap. This is not at all easy to find. My Uncle had to order it from a supplier he uses. You won't find this at Napa or Ace as it's a very uncommon pitch for that diameter.


Here is a picture of the final coupler installed on the motor. A couple things I didn't capture on film was the recessed washer and bolt which screws in from the front. We had to recess these to give clearance for the transmission input shaft. We also are using a beveled hex bolt and we beveled the washer on the lathe to match and recess the bolt further. A quick coat of paint to help with rust and we are done. Notice the black garbage bag duct tapped around the motor. This is quick and cheap to do and will prevent anything getting in the motor during the build.


Here is the simple yet invaluable alignment tool. Basically the adapter plate has an existing 4" ID and the coupler is 3.125" OD. So we machine this tool to slide over the coupler and then the adapter plate slides over that allowing us to perfectly align the plate with the center of the transmission.


Here is the shot making sure the alignment tool works. It's a tight fit, just like we want. The outer darker metal ring is the steel alignment tool.


There were quite a few steps again here that I couldn't take pictures of. We needed more hands than we had.

The first thing you'll do is use any alignment pins on the transmission and mark those across first. Simply slide your plate onto the alignment tool and find a good position that will make sure no part of the transmission is sticking out past the edges. Then using a rubber mallet give the plate a whack over the pins to mark out their location.

In order to start marking your holes you'll need a couple of things. First is a good set of transfer punches. This can be placed in different sized holes and mark the center where we'll need to drill. There were four holes on the tranny that were threaded and we couldn't use a transfer punch. We bought extra bolts and cut the heads off and turned a perfectly centered point in the lathe. We could then use our mallet again to mark these locations and drill them all out.

After we had all of our holes drilled out we butted the transmision to the adapter plate and secured it with a few bolts and marked the outline which you can barely see as a scratch in the metal.
Now comes the cutting. In our case the most precision cutting tool we had was a plasma cutter.
It's not a perfect cut but only took about two minutes. I later took a grinder to the sides to clean it up a little bit but it doesn't need to be perfect as this edge is only cosmetic.
So next we realized the flywheel didn't have enough clearance and was rubbing on the plate. We then had to bevel the inside ring of the plate and recess the four motor bolts. I believe we put those in about .150".
Here we finally have the transmission mounted to the motor! The clutch and flywheel at this point were not actually attached to the coupler. My uncle and cousin held the assembly in place while I snapped a quick picture.
The next step will be to tear this back down and reassemble it with proper torques specs, locking compound, etc and then finally give it the 12v test!
I also had about 80k miles on this setup so I'm going to spend the money and replace the pilot and throw out bearings while I'm here. I'm thinking of just replacing the clutch too as it does show some decent wear.