Junkyard motors parade

AC, DC, amps, volts and kilowatt. It's all discussed in here
electricty4cars
Noobie
Posts: 9
Joined: Wed, 04 Sep 2013, 22:54
Real Name: Electricity 4 Cars
Contact:

Junkyard motors parade

Post by electricty4cars »

Those are already pretty decent numbers considering that you are working under financial constraints. Some are even lower that what you will come up with. I am impressed with the way you handle your hobby. You make sure that you still work with it practically even if you are already addicted to it.
Last edited by electricty4cars on Mon, 09 Sep 2013, 21:42, edited 1 time in total.
Visit www.electricity4cars.com to Download EV Conversion Guide
User avatar
peskanov
Groupie
Posts: 153
Joined: Sat, 15 Oct 2011, 00:55

Junkyard motors parade

Post by peskanov »

BigMouse, that's a neat idea. Our motor is already well dried under the sun, but it could be a good idea to test on the next one (I guess there will be another one at some point, rewinding motors is nasty but also strangely addictive).

electricity4cars, thank for the encouragement.

Speaking about money constraints, I have been thinking about ways to power the motor in the dyno, and also have meditated about the price of the LiFePo pack.
It does not make sense to invest (in my case) so much money now, it would be wiser to save money for a year or so. But...without battery pack the fun with the motors and conversion must stop; so I finally went to a battery shop and asked for lead-acid prices.
I plan to buy a lead-acid pack, use to test the motor/motors and convert the car using it. Later I will sell the batts and put a decent LiFePo pack on it.
I got 3 offers, and I would like to hear opinion from the you people, as I don't know too much about lead-acid.
All the packs are 8 blocks of 12v, 150Ah.
- AGM generic brand, 8 12v batts, 360KG. 2300 aud.
- Tudor (local brand) flooded, 304 KG. 2080 aud.
- Fulmen (cheap brand) flooded, 1700 aud. That one is 140Ah.

I want to perform a continuous test (batt. 150A) and a peak power test (batt. 500A, few seconds).

Are those packs ok for peak power testing?
Which pack would give me better range, flooded or AGM?
Are AGM batts worth the extra weight?
T2
Groupie
Posts: 121
Joined: Sat, 08 May 2010, 04:51
Location: ON,CANADA

Junkyard motors parade

Post by T2 »

Since I may embark on the rewinding of a 5.5kw soon with plans to include an integral 4 :1 planetary reducer later on, I continue to follow your progress with interest.

I will be considering a stationary Back-to-Back test rig with facility for testing at nominal powers and perhaps greater.

The most economic way is to seek out a convenient location which has 3-phase power on tap. The test rig I have in mind will perform comprehensive load testing but at two speeds only. These speeds will be 1800rpm and 3600 rpm. Each of these test speeds will be facilitated by shaft coupling, in turn, one or other of a 2-pole or a 4-pole induction motor. These AC induction machines will be operated at their normal 460Vac ratings although primarily as induction generators. The other piece of equipment required is a 460Vac/100Vac 10Kw transformer to supply the Curtis controller. This will need the addition of 3 dual-diode devices mounted to a heatsink in order to provide 140Vdc rectified current.

The way I see it, your investment in a battery pack begins to degrade the moment the acid is added and the cells are given their initial charge. Of course breakers yards may give you a deal on used batteries that could be attractive. Then there is the fact that you still do not yet have a mechanical load nor a means to dissipate that power. My solution on the other hand delivers on all fronts.
T2
User avatar
peskanov
Groupie
Posts: 153
Joined: Sat, 15 Oct 2011, 00:55

Junkyard motors parade

Post by peskanov »

T2, it's great to see your a planing a high revs rewind. I hope you will share some part of the process with us if you go ahead.

I considered testing the motor in a way similar to the one you are proposing. In my case, I don't need a motor brake as I have access to an old electromagnetic dyno. Looks like this one:
Image
Therefore, I can test continuous and peak power, provided I have the power for the Curtis.
I have access to a 15 KW 3 phase line, but that would not let me check peak power. I don't like that because, having an "amateur" rewound motor, it would good to test it (an examining the stator afterwards) before fitting it in a car.

Also, transforming and rectifying the current will need some serious money for 10-15 KW (I guess it could be pretty cheap for the USA, were voltage is already 110V, but not here).
At the end, a lead-acid pack allows testing continuous or peak without problems. I considered renting 2 forklift battery packs for a weekend, from any small factory close to my working place; but in the end I see I have no money to buy LiFePo for my conversion, so I prefer to solve both problems at the same time and buy a lead-acid pack right now.
User avatar
peskanov
Groupie
Posts: 153
Joined: Sat, 15 Oct 2011, 00:55

Junkyard motors parade

Post by peskanov »

Dyno time!
I bought a set of batteries a few days ago, the AGM ones. I am pretty sure they are not adequate for an EV, but having AGM for the price of flooded made them too good to resist Image. Well, I hope they will last for 2 years and will give me time to save for LiFePo.
As the brand is unknown here (it is Indian), I bought just one to test it a bit. Here it is, and the specs:
http://www.aegan.co.in/upload/A12-150.pdf
It weights 45/46 KG, has 150Ah.
Image
If the specs are right, it can crank 2000A peak. It also has a good 1 hour capacity, I guess due to low Peukert.
I made an small capacity test using steel wire. I miscalculated, as I didn't anticipate that resistance grows a lot with heat. As I didn't want to rebuilt the thing, instead of testing capacity at 75A (my target) I had to test it at 25A. Well, at least I was able to check the battery was not total rubbish.
Image
My diy "rheostat":
Image
According to spescs, the battery had to discharge in 5 hours. Took 4 hours and 15 minutes. Not bad, as it's brand new and not cycled. Lead-acid takes a few dozens cycles to reach nominal capacity.

So, with all components ready we wen to the dyno. We built the battery pack there using a crimper and thick cable and terminals. That was quite fast, maybe took an hour or two.
Image
...and mounted the motor in the dyno. We built a base for the motor unsing 2 pieces of "U" shaped steel parts.
Image
Chaining the motor to the dyno:
Image
And here is the full setup; I think it took about 5 hours.
Image

The full test it still to be performed, but we made it run a bit.
1st problem: the Curtis 1238-6501 didn't like the 103V tension. All the info I have found says 105V max or 108V max, but it refused to work. We tried to find all relevant parameters (pack voltage, max overvoltage) but got no results.
Maybe the 105v limits is just for the 1238R version?
Well, just to test the setup we skipped one battery and got down to 90V. That made the controller to run fine and rolled the motor. Silent as a ninja, all we could hear was the chain.
We tested and compared the phases using the clamp ammeter, it seems all phase went along (still have to test the same with bigger loads).
As a final test, we ran the Curtis autotune. Bad thing; it ran the test, heated seriously the motor wires and messes with encoder parameters!?
We were not able to make it work again, as the controller displayed "wrong parameters" all the time. We fixed the incorrect parameters of the encoder, but some other param is also badly setup.

I hope the dyno will be ready soon (some parts have to be fixed/rebuilt, has not been used for years, the told me) and we will be able to run all the test. If anybody here has any interesting set to run, just tell me.
User avatar
BigMouse
Senior Member
Posts: 600
Joined: Thu, 28 Oct 2010, 02:39
Real Name: Vincent Tannahill
Location: Silicon Valley
Contact:

Junkyard motors parade

Post by BigMouse »

Awesome stuff! I can't wait to read/see how this goes. Please take video of a dyno run once you get it going.

What sort of data will you be logging on the dyno?

I had hopes of building a dyno of my own at one point, but decided I didn't want to invest as much money in it as it would have required. I'll use a locked-rotor with a load cell measuring torque to set my time-constant-dependent parameter. One value for a cold motor, another value with the motor at operating temperature, each time finding the value that gives the highest torque. Then interpolate between the two in the controller based on motor temperature. Not ideal, but hopefully good enough.
User avatar
peskanov
Groupie
Posts: 153
Joined: Sat, 15 Oct 2011, 00:55

Junkyard motors parade

Post by peskanov »

BigMouse,
here you have an small video of the dyno running! We are still setting things, so no results yet (btw, I fixed the voltage limit problem thanks to the advice posted in Ivan's garage).
http://www.youtube.com/watch?v=8-bEra5qP3A
The guys of Sport Devices SL, owners of the dyno (we are collaborating in a few projects) tell me this dyno was used with motors up to 170HP, but that was some years ago. Lately they have been using a kinetic dyno (unlike the one in the video, which has electromagnetic braking).
Running this thing is scary, the chain is getting some rough treatment. I know it does not look that way in the video, but being there is very different!

I would like to know:
- Max torque & max power (for this controller of course, as I am controller limited). I have my calculations, but surely real data will be different.
- Temperature and power developed at continuous 50A,75A,100A,150A,200A (at the battery). I want to know if the motor can sustain these values without overheating. I plan to drive at 50A-100A to get some range from this battery. And I want to know the efficiency and look for possible problems if the squirrel cage is damaged. So far, I have not seen any symptoms of damage, as phase unbalancing or excessive heating...
- Torque curve in the "field weakening" region, max rpm, check for vibrations,
- Behaviour running on star configuration (max torque etc).

Yep, I also played with the idea of building a dyno. I toyed with the idea of using a water pump as load, and later thought about using an old AC motor as generator (using 3 big capacitors) and resistive load.
These are not very practical ideas, I know Image.

Maybe you can do the same I did and try to find companies running a dyno in your area. Many people is interested in electric traction now, and they could offer you free use of the dyno just to know the EV technology a bit better.
Vectrix150V
Groupie
Posts: 73
Joined: Mon, 19 Aug 2013, 19:20
Real Name: Heath Young
Location: Newcastle NSW

Junkyard motors parade

Post by Vectrix150V »

Quite a few people have bad things to say about the autotune that the Curtis controllers do - did you note down the default values before autotune borked your settings?
User avatar
BigMouse
Senior Member
Posts: 600
Joined: Thu, 28 Oct 2010, 02:39
Real Name: Vincent Tannahill
Location: Silicon Valley
Contact:

Junkyard motors parade

Post by BigMouse »

Looks good. I'm glad to hear the rotor seems to be in reasonable shape. I'd also be interested to see if you notice a difference between star and delta connection. I expect you will due to being controller limited in both cases.

I'll ask a contact of mine at a local transformer winding place if he knows of any electromagnetic dynos. I know he has contacts at a motor winding shop.
User avatar
peskanov
Groupie
Posts: 153
Joined: Sat, 15 Oct 2011, 00:55

Junkyard motors parade

Post by peskanov »

Vectrix150V,
no, I didn't take any note about previous values. Autotune should not change many parameters; I think it changes slip gain, a few field weakening related values, and encoder settings
Slip gain moved to ~2, and it seems original value is <1.
It didn't change any FW value (these are disabled by default).
The motor runs really smooth, so I guess the slip gain is OK; of course I had to fix the encoder params myself as they were messed up.

BigMouse,
I think you will find dynos easily in the world of ICE; i.e. little shops supporting racing teams (motorbikes, cars).
Sport Devices, the company supporting me, has worked a lot with high end motorbikes. These kind of guys are quite approachable, and many are curious about EVs now.
User avatar
peskanov
Groupie
Posts: 153
Joined: Sat, 15 Oct 2011, 00:55

Junkyard motors parade

Post by peskanov »

Hi there, time for some updates.

Bad news first: I lost access to the dyno, and that was before I got skilled enough to get good results. Unfortunately, Sport Devices SL restructured and the dyno was moved to other place.
Anyway, here is a summary of what I learned.

Problems, glitches, enigmas:

- Contactor troubles. We have two of those (400A, they say):
http://kellycontroller.com/main-contact ... -1102.html
Both got soldered/stuck under working conditions!
The 1st time, I traced the problem to a faulty precharge resistor (we installed it poorly).
The 2nd time I was just speeding up the motor, it seems the 400A current was too much for the contactor and got soldered.
It's easy to fix (open, separate and grind a bit), but I am not sure I can use this contactor model for the car.

- Curtis controllers have a sensor pulse limit. This means you cannot run any encoder generating more than 600.000 pulses/second. As we had an 200 pulses/revolution encoder, the controller stopped speeding at 3000 rpm.
Thanks god, Chinese encoders are cheap and reliable so we got a 100 pulses one and now we can speed up to 6000 rpm.

- I got a (weak ) electric shock having one hand in the motor and the other in the aluminium fin of the controller. I measured it and got changing voltages up to 100V.
This puzzles me, as we measured insulation of our winding and got good numbers. I suspect this charge is produced by induction of the head coil in the font/end plates of the motor.
Should I connect the motor body with the controller fin? Would that fry the controller? If anybody has a clue, please tell me.

- Measuring a brake dyno it's not easy, as it needs some brake control regulated by a PID. Also, you need to adjust the motor controller PID as well. If not, both engines will fight and you will get a messy graph when accelerating. We were not able to adjust the PIDs correctly in the first 2 sessions with the dyno, and there was no dyno for a third try. Image

A few results I got (AC amps are just approximated) from those early testings. As you can see important data is missing (i.e. maximum power was not measured):

Test (Delta) at 1500 rpm

    100 Amps DC -> 200 Amps AC -> 8.5 kw 47 nm

Test (Delta) at 2500 rpm

    196 Amps DC -> 200 Amps AC -> 14 kw 45 nm
    390 Amps DC -> 440 Amps AC -> 27 kw 95 nm

Continuous test (5 minutes) WYE, 2800 rpm

    155 Amps DC, 150 Amps AC -> 12 KW

No load test

    WYE 4400 rpm; 10 Amps DC
    Delta 4800 rpm; 15 Amps DC, 48 Amps AC


At 4000 rpm, vibration was noticeable. An small machine (3 KG) resting on top of the dyno started dancing and moving a bit. My pal Vicent says this vibration is nothing (he worked in a factory for years, servicing electric motors among other things). I think I will drive under 4000 rpm just in case.

So, that's all I got. Torque numbers are a bit lower than I expected, but they are a still pretty close, so I am pretty happy with my rewound motor.
Now I am looking for the car and saving some money. And also looking the paperwork. Jesus, this part is going to be thought!
Last edited by peskanov on Mon, 13 Jan 2014, 20:13, edited 1 time in total.
User avatar
Johny
Senior Member
Posts: 3749
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Junkyard motors parade

Post by Johny »

peskanov wrote:- Contactor troubles. We have two of those (400A, they say):
http://kellycontroller.com/main-contact ... -1102.html
Both got soldered/stuck under working conditions!
They look like Albright copies. The EV200 contactors are still around on eBay (Eg. eBay item 171198856624) but they are up around US$110 now. They are a vacuum contactor but still need to be treated well to work properly.
Should I connect the motor body with the controller fin? Would that fry the controller? If anybody has a clue, please tell me.
Can you measure what the heatsink fins are connected to (-ve battery?)? I think the safest way is leave the motor grounded to the chassis and somehow prevent anyone topuching the fins. When you do the car the traction battery pack should "float".
Now I am looking for the car and saving some money. And also looking the paperwork. Jesus, this part is going to be thought!
Yes, you have indicated all along that red-tape was going to be a drag.
User avatar
peskanov
Groupie
Posts: 153
Joined: Sat, 15 Oct 2011, 00:55

Junkyard motors parade

Post by peskanov »

Hi Johny, this contactor looks good and has a decent price. But maybe I should check for a higher Amps one, maybe in a few years I will get a bigger controller.

I checked the conectivity of the Curtis. The cooling fin seems to be well insulated from all 5 terminals (B+,B-,U,V,W).
I can insulate the fin when installing, but we didn't when converting the Quad. Maybe connecting them is OK?

I mentioned the paperwork cause I read the Spanish manual for vehicle reforms, and everything is biased to exchanging an ICE for a another one! Image
User avatar
weber
Site Admin
Posts: 2936
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Junkyard motors parade

Post by weber »

peskanov wrote:I checked the conectivity of the Curtis. The cooling fin seems to be well insulated from all 5 terminals (B+,B-,U,V,W).
I can insulate the fin when installing, but we didn't when converting the Quad. Maybe connecting them is OK?

If it's insulated from hazardous voltage AC and DC sources (as it should be) then the shock is probably due to capacitive coupling of the higher harmonics of the PWM. This can be expected from motor body to chassis and inverter heatsink to chassis as well as inverter heatsink to motor body. To avoid this, all exposed metal should be "equipotential bonded". In the case of a typical EV, this means both inverter heatsink and motor body should be connected to chassis.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
User avatar
jonescg
Senior Member
Posts: 4721
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.
Contact:

Junkyard motors parade

Post by jonescg »

weber wrote: If it's insulated from hazardous voltage AC and DC sources (as it should be) then the shock is probably due to capacitive coupling of the higher harmonics of the PWM. This can be expected from motor body to chassis and inverter heatsink to chassis as well as inverter heatsink to motor body. To avoid this, all exposed metal should be "equipotential bonded". In the case of a typical EV, this means both inverter heatsink and motor body should be connected to chassis.


If you had shielded wires on your motor's phase wires, would you attach the shield to the metal housing of the inverter or to the motor? Both generate a bit of noise, but I wonder if it matters...
AEVA National President, WA branch director.
User avatar
weber
Site Admin
Posts: 2936
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Junkyard motors parade

Post by weber »

jonescg wrote:If you had shielded wires on your motor's phase wires, would you attach the shield to the metal housing of the inverter or to the motor? Both generate a bit of noise, but I wonder if it matters...
You raise a point I failed to mention about equipotential bonding. There should not be any loops. It should have a star or tree structure. If there's a loop it will act as the short-circuited secondary winding of a transformer, for any stray magnetic field that's going, including from nearby lightning, and large currents may flow around the loop.

The shield should be connected at the source, i.e. the inverter. You might say that during regen, the motor is the source, but the inverter is always the source of the PWM frequencies.

However, the recommended connection for a motor inverter (VFD) is to connect the shield at both the inverter and the motor, provided that the motor is otherwise floating relative to chassis and inverter housing (which are connected). i.e. the motor housing is connected to chassis only via the cable shield and inverter housing.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
User avatar
Johny
Senior Member
Posts: 3749
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Junkyard motors parade

Post by Johny »

peskanov wrote:I mentioned the paperwork cause I read the Spanish manual for vehicle reforms, and everything is biased to exchanging an ICE for a another one! Image
Fair enough. I should say that it was only marginally better here. No-one at the actual motor regulations department (VicRoads) knew anything about EVs - and the engineer simply followed the paperwork. Assuming that you follow a similar system over there - in that you get an engineer to examine and approve the conversion first, are you on any forums (or other contact) that could lead you to an engineer that has already done an EV?

With Spain on the leading edge of renewable energy generation and making all the right noises for commercial EV uptake (telephone booths to charging stations etc.), you would think that somewhere there is someone who could re-interpret Decree 736/1988 and help you through the process. Google soesn't help much though - Spain Electric Vehicle COnversion doesn't yeild anyting useful. I thought there was a Spanish EV club though.
User avatar
peskanov
Groupie
Posts: 153
Joined: Sat, 15 Oct 2011, 00:55

Junkyard motors parade

Post by peskanov »

Ok then, both connected to the chassis. Thanks Weber!

Johny, I am still looking at the requirements. It seems Spanish law require a project signed by an engineer, and an responsible authorized workshop. I am still not sure if there is an inspection at the end of the process.

In Spain there is no EV scene; there is some movement in the north-east area (Catalonia) where at least 2 motorcycles have been converted and legalized.
There is also an small engineering shop (ZEVNA) trying to build a conversion bussines in the north. They have legalized at least 1 car conversion.
Now you can see why I am wandering in an Australian forum! Image.

People here is quite wary of EVs. Spaniards are willing to do something for the environment, as far as they feel doing so cost nothing at all. EVs, being expensive, look suspicious to our eyes. Also, the average Spaniard thinks "electricity is very expensive" and wonder if charging an EV could more expensive than feeding an ICE (people here does not even know the difference between Kw and Kwh!).

I know this is quite puzzling, because it's true Spain is a leader in renewables. We have about 40% of renewables in the electric mix (real production, not power capacity). Mostly wind, plus dams and a bit of solar (5 or 6 GW).
Unfortunately, the 2008 crash killed the whole thing (even a wonderful development of thermoelectric plants which was developing really fast). Many people thinks the renewables boom had relation with the economic crash; that's a really popular view among conservatives.

Anyway, that's one of my motivations: contributing a bit to popularizing EVs here.
User avatar
peskanov
Groupie
Posts: 153
Joined: Sat, 15 Oct 2011, 00:55

Junkyard motors parade

Post by peskanov »

Hi there,
my donor car it's ready!
It's junkyard material, a Ford/Mercury Cougar with ~200.000km (13 years).
I paid about 700E (about 1000 ausd), not a bad price in my country.
Image.
Not the perfect car for an EV conversion, as it is quite heavy. But I hope removing the V6 engine will lighten it a bit.
I chose this car because it has an uncanny semblance to the mythical Solectria Sunrise, specially the back side.
Image

Unrelated to this, I see LiFePo4 prices are going down in China. I hope next year I will be able to sell my SLA an buy a Lifepo pack.

A junkyard car, perfect couple for my junkyard motor.   Image
User avatar
coulomb
Site Admin
Posts: 6357
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Junkyard motors parade

Post by coulomb »

peskanov wrote: Not the perfect car for an EV conversion, as it is quite heavy.

Yes, the heavier the donor the lower the acceleration for a given torque. But light cars typically don't stay light after you add an EV capable battery, so the difference isn't so great in the end.

Plus, the car will be designed to take the weight. We stuffed some 350 kg of battery (including battery boxes) into an MX-5, and it feels quite different to when lighter, and creaks and groans noticeably in the parking lot. Presumably a heavier car would shrug off the extra weight, or at least deal with it more gracefully.

So don't feel too bad about choosing a heavier donor.

Oh, and congratulations on the find! Image
MG ZS EV 2021 April 2021. Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
Patching PIP-4048/5048 inverter-chargers.
If you appreciate my work, you can buy me a coffee.
User avatar
peskanov
Groupie
Posts: 153
Joined: Sat, 15 Oct 2011, 00:55

Junkyard motors parade

Post by peskanov »

Thanks; yes, that was my reasoning. A heavy car should be better built for a heavy pack, provided I remove as much weight as possible (AC included).
I am stuffing about 450 KG there (lead acid pack, +60 KG motor, +Curtis). I have to remove a lot of weight from the car. Image
User avatar
peskanov
Groupie
Posts: 153
Joined: Sat, 15 Oct 2011, 00:55

Junkyard motors parade

Post by peskanov »

It's alive!

https://youtu.be/zN_xl9S_WcU

Image

Jesus, doing a modern car conversion (without money and/or extensive mechanics knowledge) is a real challenge.
I don't regret it, as I learned a lot about cars and mechanics in the process. However, I don't think I would have attempted this, if I knew who much effort was going to be involved. Image
User avatar
weber
Site Admin
Posts: 2936
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Junkyard motors parade

Post by weber »

Congratulations Peskanov. Well done! Image
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
User avatar
peskanov
Groupie
Posts: 153
Joined: Sat, 15 Oct 2011, 00:55

Junkyard motors parade

Post by peskanov »

Thanks a lot Weber; I will be posting some observations the next few weeks, after doing a few runs.

My feeling is the motor will need forced ventilation. The motor wires were already warm in our few tests (runing on WYE config).
User avatar
Johny
Senior Member
Posts: 3749
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Junkyard motors parade

Post by Johny »

Congratulations from me too! Agree on forced ventilation. Traffic is the killer - no airflow and high motor current each take off.
Post Reply