Adding regenerative braking to the Mirage
 

I have finally gotten the alternator delete done on my Mirage. It has been working consistently and perfectly ever since I tweaked a few things. It was a heck of a time getting it to work good, but I am pretty sure I got it all nailed down. Much more info in my thread here:

Help me kill my alternator - it is stubborn and won't die

https://mirageforum.com/forum/attach...3&d=1542415861

https://mirageforum.com/forum/attach...0&d=1565914673



To get this alternator delete to work, I had to create a circuit to trick the alternator into thinking the battery is fully charged. With the next phase of this mod, I plan to bypass this circuit when braking. This will give me some regenerative braking capability.

Here is the rough and not finalized diagram of the circuit I plan to use to do this. The 7805, capacitors, and dc-dc converter are the circuit that tell the alternator the battery is fully charged. The relay and switch are there to bypass the circuit to get the alternator to charge normally. This allows me to turn the alternator on whenever I want, or just when I am braking.

https://ecomodder.com/forum/attachme...1&d=1566397148



Finally, I know we've talked about this over the years, but I've never seen someone actually post about doing this. Have any of you guys actually done this? I'd love to hear more about how you did it and the results you had with it! :thumbup:

I've strongly considered doing something like this on my Insight, as I've seen about a 10%/10mpg improvement with my Meanwell pseudo-alternator turned off. Problem seems to be that the 12V battery can't hold any significant amount of regen, so one would need to drag the brakes for a long time to get ant usable regen - one couldn't just tap the brake once and throw in as many amps as would be necessary to charge the battery. I also rarely brake, so my braking isn't more than a quick tap every now and then anyway.

Supercapacitors sound like a possible solution, but I am not knowledgeable enough on them to know how they would work in a situation like this. If supercapacitors would work, and actual regen could be used to the 12V battery, that in combination with a small solar charging setup might be able to eliminate alternator need except on highway trips. The real goal would be to eliminate having to plug in the 12V battery at night, since saving fuel cost at the expense of additional electrical cost isn't a money savings.

What deep cycle battery did you go with?

Right now I am running a group 24 deep cycle 12V lead acid battery. Its rated for 80ish amp hours.

I agree, it'll be interesting to see how much it'll actually regen. I think I may pick up a 100a ammeter so I can see what is actually going on. I did some rough calculations that showed I could possibly regen roughly 2-3Ah per commute. My commute would use about 8Ah though (again guesses as I don't know my current draw).

The main reason I'm considering it is I have a few fair size hills on my commute. Both have stop signs/lights at the bottom. So, I have several opportunities to regen brake a bit.

I also just want to try it. :D It sounds cool to have a normal car with regenerative braking.

The max you are going to be able to push into a big AGM battery would be maybe 1hp worth of electricity.

That's about all the alternator will make anyway.

IIRC lithium can handle much greater amounts of regen than lead acid. I wonder if a tiny 6ah lithium battery could handle all the regen you could throw at it, and then use a DC-DC converter with a voltage cutoff (to prevent the lithium from discharging too far) to slowly charge the deep cycle lead acid battery from the lithium battery. Then you might be able to regen at a higher amperage - say 300 amps of regen versus just 50 or 100 (which might damage the lead acid battery). One would just need to do something to make it pull that 300 or so amps whenever braking. Obviously like in a hybrid any regen is still an overall efficiency loss compared to EOC, but if you have to brake, why not charge the 12v while you're doing it?

A small, 25w solar panel exposed to 4-5 hours of sunlight a day, running at average efficiency, theoretically should provide 5-6ah of charge per day: 20 watts * 4 hours = 80 watts / 14v = 5.7ah. Even with efficiency losses that's still about half of your electricity use.

The alternator on the Mirage is rated for 85 amps. Will it ever really pump out that kind of amperage with a battery that is anywhere near charged? I have no idea. I will have to get an ammeter setup to test that.

While stumbling through my process of killing the alternator, I did discover a way to get the alternator to pump out a solid 15V. This is higher than it's normal output of 14.4V. A small boost, but it will help pump in more amps. I do plan on testing the higher voltage to see what difference it makes.

Perhaps with a small solar panel and the regen I will not have to plug in. That would be pretty cool. But one step at a time for now.

Thanks for the ideas guys!

Alternator ratings tend to be optimistic.
To handle 85 amps you would need at least 6 gauge wire, to handle it continously it would need to be 4 gauge.
You will need at least 15 volts to over come wire resistance between the alt and batt if you want to run higher amps.

I have found that car alts typically have about 10 amps available for charging a battery at idle speed.
That alternator is only going to be able to make anywhere near 70 amps is spinning it 7,000rpm.
Even my modified DR44G welding alternator only makes about 25 amps at idle speed externally powering the armature with 13v. The DR44G is a pretty good size medium duty commercial truck alt.

Sounds like some testing is in order. I'll get an ammeter and shunt on order.

Someone I know "overclocked" their alternator with a different pulley size. Might be consideration if you don't plan on using the alternator for normal driving. I surmise there would be a definite amount of additional load from doing that, but you might be able to get a bit more "regen".

Also what about when your battery is full? Any idea on how to stop charge from going to the battery when braking if the battery is full to avoid cooking it?

No, NO, NO.

Spinning the alternator faster pretty much only produces more losses after a few thousand rpm. They're usually spinning at >2x crank speed, so you'll have most of it by 2000rpm at the engine. The limiting factor is the current capacity of the stator winding.

You can only increase output by increasing voltage or putting in a bigger alternator. With increased voltage you will need more rpm to let the alternator reach that voltage if the field coil is current limited.

The alternator will need to be run more than just when you brake unless you have a really huge battery that's being charged at home and depleted as you drive.

If it were me, I'd use about 30Ah @ 24V lithium titanate oxide batteries (~$500), a 24V regulator from a tractor alternator, and DC-DC converter (~$70), and a switch to cut the alternator out above a certain % throttle input (if the engine is at low load you might as well run the alt).

Using LiFePO4 is more troublesome since you'll need a lot more of them in parallel to accept a fast charge, like 70Ah worth (assuming 100A alternator - 30A ignition/ECU/fuel pump/lights) which gets quite pricey and bulky.

Yes, it mostly does just produce extra losses at higher RPM, but more power is available at lower RPM than otherwise, and for max regen braking you'd want to keep it in the highest gear possible down to 1000 RPM to avoid engine friction when slowing down. As I said this would be counterproductive if the alternator was going to be used normally, but if it's just going to be used for regen as the OP seemed to indicate, then it might be worth checking out at least.

Why 24V? The car runs on 12V....15V max if you want to keep things fully charged.

Cost is also a factor - $570 is a decent chunk of change. Once you start getting up there in price you might as well have a hybrid that has full regenerative braking and whatnot. Part of the economy of not having a hybrid is to avoid having expensive batteries in the car, whether those are HV batteries or expensive lithium 12v batteries.

super capacitors can take a charge almost instantly

A custom pulley is going to be a good chunk of change too, and the diameter on the alternator pulley is already small. Spinning the alternator faster for a tiny bit more regen doesn't sound like it's going to end up very economical to me if you end up having belt slip issues. Also the belt efficiency will drop, and the alternator fan will produce more losses.

24V because it instantly doubles the power output from the alternator. I'm not sure if stock rotor field windings can handle double the field current, but if they can there's no issue at all. If they can't, you might need to downshift one gear to get the best out of the alternator (you'd be increasing the regen to engine braking ratio anyways so it's fine).

Of course you can get away with less than 500 on the battery too, 30Ah is pretty generous. 20Ah would probably be totally fine, since not all the alternator current is going to the battery. It depends how often you want to cut the alternator and how long you want the battery to last.

Ultracapacitors to absorb the charge would let you use 12V LiFePO4 batteries to exclusively power the car, but the ultracapacitors also cost money, and you still need a DC-DC converter.

Obviously the cheapest solution is to just use the stock alternator with a 12V battery, but that barely saves any fuel.

It depends on how much you want to spend. On longer trips an alternator delete would get expensive quickly, and once the alternator has charged the battery to full, alternator load is pretty light. An alternator delete would have the most effect on shorter (<1 hour) drives.

My plan (down the road once my current one year old 12v dies) is to have two 20ah deep cycle batteries in parallel with a couple 25w solar panels wired up in parallel to keep them topped off, and a kill switch to manually control my alternator - which is really quite easy with how my Meanwell is setup. Estimated cost is $80-$100 for the batteries, and maybe another $100 for the solar.

All great ideas guys. Keep them coming! The more ideas we lay out the better chance we have of coming up with another great idea.

I just finished ordering a fancy ammeter for the Mirage. I was just going to use an analog meter with a shunt. But, the more I looked around, the more I found cool nerdy stuff that would just be nice and the price was quite reasonable. So, I ended up going with an ammeter that also measures voltage and amp hours used. You can also set a capacity, and it has a pretty battery gauge on it. This gauge should much better show just how much power I am able to regenerate with my regen braking setup. :thumbup:

DC Multifunction Battery Monitor Meter

https://ecomodder.com/forum/attachme...1&d=1566650573

Be advised that much of that fancy pretty interweb stuff is inaccurate as heck. I have had some in the ranger, and only the voltage was close according to my 1950's test equipment.

I'll check the digital fancy gizmo with my analog ammeter to see that its at least roughly the same. Beyond that, I'm not sure how I would go about testing it.

I moved the circuit into the cabin today. This should have been a short ordeal, but it really wasn't. For the life of me, I could not find an easy and quick ground for this circuit. I finally ended up finding a unused hole on the steering column and bolted the wire to it.

While I was moving the circuit into the cabin, I also added a toggle switch to it. Now I can turn my alternator on and off at will. I tested it and it works great!

https://www.youtube.com/watch?v=W6eP...ature=youtu.be



Here are some pictures of the process.

The fuse wire now runs through the firewall into the cabin.

https://mirageforum.com/forum/attach...9&d=1567046800



I stuffed the circuit in the glove box for the time being. I'll tidy it up after its all in there and working.

https://mirageforum.com/forum/attach...0&d=1567046875



Then, I had a heck of a time finding a ground... I really don't know what else to say. This really shouldn't be a hard thing to do, but it seemed to take me for forever.

The switch is currently just on a wire and I have to grab it and flick it on. I tried it on a quick drive tonight and it works beautifully. I will be adding the amp hour meter once I get it. But, in the mean time I think I'll hookup a 200A ammeter that I have kicking around. That should show me roughly how much power I'm getting with the regenerative braking.

https://mirageforum.com/forum/attach...1&d=1567047052

Interesting! I never noticed that the Scangauge gives inaccurate 12V battery voltage readings. I wonder if a cigarette lighter plug would work, of if you'd actually have to have a multimeter hooked up to read actual voltage. Looking good so far, definitely interested in seeing how much energy you can capture through braking. Have you hooked up your Amazon meter yet?

I do not have the Amazon meter yet. It should be here in day or two though. It will definitely help shed more light on how well this thing works.

That being said, there is a decent hill right before I turn into the industrial park at work. Of course the turn is right at the bottom of the hill, so it's often that I DFCO down it. Today I flicked the alternator on during the decent. Wow, I actually noticed the braking power. I was surprised that it was that noticeable. As we have mentioned, it is probably only providing about 1hp of braking power. I am not doubting that, but the fact that it's so easy to feel.

I'm imagining the braking power will likely be larger the lower your battery voltage is? A larger delta between alternator voltage and battery voltage?

I would guess that too. When I pick the alternator on, it kicks the voltage up to 15v. So the larger the voltage difference the more amps that will flow. The ammeter will show us just how much though. Looking forward to testing it out.

Alright, I got the meter. I'm trying to figure out where to put it. I think I'd like to 3d print a small bracket and slap it up under the HVAC controls.

https://ecomodder.com/forum/attachme...1&d=1567440813




Does anyone know of some good double sided tape that will hold up to summer heat? I know there has to be good stuff out there, I just don't know what to look for.

https://ecomodder.com/forum/attachme...1&d=1567440832

I got the new amp hour meter installed today. It seems to work very nicely. I have not tested it against my other ammeter yet.

However, I was quite disappointed that it does not handle the regenerative braking / charging the way I would have liked. It simply adds ANY current to the totals on the gauge. It never subtracts them. No matter what way the energy flows it adds.

It should still work for my purpose, just not as well as I had hoped. Here is a short video showing what I mean.

https://www.youtube.com/watch?v=XZNt...ature=youtu.be

I wonder if it's possible to hook up two of these (or two simpler units) to measure amps in vs amps out, by connecting them to your alternator switch - so one unit measures power draw when the battery is discharging, but then when the alternator is on the other unit is measuring power put back in the battery.

Was looking at an amp-hour meter today - https://www.amazon.com/AiLi-Battery-...dp/B07CTKYFTG/ - that might better indicate what's going on; you could compare the ah in your battery before and after a trip to see what your net gain/loss was. That in conjunction with your current instrumentation would provide a better idea of what is going on. It's a bit more pricey ($39) but less than comparable meters (the Victron one is $200 :eek:), and one could always return it if it didn't work.

If only I was back home and if only my current 12V wasn't only a year old; this thread is making me want to do an alternator-equivalent delete on my car!

I believe another of these gauges would work if you put one on the cable from the alternator. However, I would much rather have one gauge that works the way I would like. It seems from the description that the gauge you posted would do this.

I did make the commute with the new gauge installed this morning. I learned a bunch of interesting info:

1) Power usage:
- 2A with they key in the ON position
- 6.5A with the engine running
- another 2A for running lights
- more testing to come on this...

2) The alternator can indeed spit out it's rated power, at least briefly. The highest I saw was actually 91A for a short period. I'm not sure what kind of continuous power it can output, but that is of little concern for my setup. When things are more setup, I plan to get some video of things working. It's too slapped together right now to do so.

3) The Mirage is pretty darn efficient when it comes to power use IMO. I figured that it would use a lot more power with being a newer car. If I am using 7A on my 30 minute commute, that means I only have to regen/charge 7Ah a day. That is not that much energy. I highly doubt it can all be done by regen, but with a small solar panel I am sure I could get away without plugging in to charge under most circumstances.

Batteries aren't 100% efficient. You'll need about 10-12ah to recharge 7ah use. And this is on the main battery lead including the starter circuit?

Excellent point about recharging. :thumbup:

Yes, the ammeter shunt is right off of the negative battery terminal. All power use goes through the shunt. I even moved the charger wire to go through it after taking the picture.

https://ecomodder.com/forum/attachme...1&d=1567518175

I reset the amp hour meter before heading home today. I opted to not use any regen just to get an idea of how much power I actually use on my normal commute. I will say this is a very ideal case. I ran no accessories at all, no fan, no headlights, etc. This is how much electricity it takes to run the car, period. I also used P&G which reduced usage further.

The voltage is reading high because for whatever reason on the last little bump start kicked the alternator on for some reason beyond me at this time. That actually kicked it up from around 2.5Ah to 2.7Ah.

https://ecomodder.com/forum/attachme...1&d=1567549980

Not to be annoying, but that looks low for consumption. Given my stance on cheap chinese electronics, can you do comparisons with semi calibrated equipment?

The only other gauge I have is a regular analog 200A ammeter or 10A multimeter ammeter. I'm not sure how I can use either of those to verify accuracy. Ideas are welcome.

I agree that the usage seems very low. However, running rough numbers the gauge doesn't sound horribly far off. My average commute is about 25-30 minutes. So lets say:

The engine is on for 2/3 and off for 1/3.
Power usage is 6.5A with the engine on, engine off is 2A.

Engine on power usage is 6.5A * .33 hours = 2.15Ah
Engine off power usage is 2A * .17 hours = .33Ah
Total power usage = 2.48 Ah

Sounds reasonably close to me.

Today's commute in was quite idea. Lots of EOC. I did use the fan on low (3.5A draw) for about 2/3rds of the trip for some heat. I got to work with about 4Ah used. So my really rough guess is I regenerated about .5 Ah of that. This is really rough guestimation though.

The trip home has more opportunities to use regen because of the placement of stop lights and signs at bottoms of hills. I sure hope I can get more than .5 Ah out of regen hahaha. But, I will do my best to verify it.

https://ecomodder.com/forum/attachme...1&d=1567686674

If your 10 amp is fused and you have spares, use that. Your load is supposedly lower than 10 amps. Or you can use a known hot resistance from a vehicle lightbulb and measure what the current through that bulb is after going through the shunt.

I've been looking around and I found another gauge that I am going to pick up instead of the one I have. It doesn't have as many gauges, but I think it will work a lot better. It specifically says that it measures current in both directions and adds and subtracts from Ah used.

DROK DC 0-300V 200A STN LCD Display Digital Multimeter Voltage Ampere Power Energy Ammeter Voltmeter Battery Volt Amp Meter AH Monitor Panel with Hall Sensor

https://ecomodder.com/forum/attachme...1&d=1567805872

https://ecomodder.com/forum/attachme...1&d=1567805886

4Ah * 12v = 48 Wh

1 liter of fuel is 40 MJ = 11kwh
If we assume terrible IC and alternator efficiency that is 1kWh of electricity produced.

So you saved 0.048 liters of fuel.

In roughly 20 days (or a work month) you save a liter of fuel.

You have any mpg numbers yet?

I haven't done any mpg testing on the mod yet. I honestly haven't planned to do it either. A bunch of people have done it over the years. Check the wiki link below.

Car MPG Efficiency Modifications Main

Yep. I have already been over those many times. They say more numbers tests required :)