[retepsnikrep]

We need pics to understand all of that description i'm afraid, well I do anyway. I get the gist but a pic is worth 1000 words and all that.

Looking down the back of my standard cars engine the other day I was a little concered that with the 4/5 stacked IMA motors the drive shaft may not clear them near the outer CV joint on the long shaft side when suspension is at full travel :? The IMA motors are wider at that point than the engine block. What do you think.

 

[Hugh-Falls]

Bobbie Burns cautioned us

Looking down the back of my standard cars engine the other day I was a little concered that with the 4/5 stacked IMA motors the drive shaft may not clear them near the outer CV joint on the long shaft side when suspension is at full travel :? The IMA motors are wider at that point than the engine block. What do you think.

I'll check it out. Maybe we will need another addendum.

 

[Hugh-Falls]

Possible interference problem

Looking down the back of my standard cars engine the other day I was a little concered that with the 4/5 stacked IMA motors the drive shaft may not clear them near the outer CV joint on the long shaft side when suspension is at full travel :? The IMA motors are wider at that point than the engine block. What do you think.

Looking down from under the hood, so far the only possible interference readily apparent would be with the harmonic balance weight that is mounted mid-shaft. Looking up from under the car may reveal some other problems and will have to be done soon.

 

[Hugh-Falls]

Adaper plates and bearing plates

Years ago, the author of High End AC Drive Systems and Power Electronics for Electric Vehicles described how he designed and had machined an adapter plate for his EV conversion. As he points out, the critical measurements are the distances that represent a triangle between the motor/transmission shaft and each of the two dowel pins, and between the two dowel pins. Another way to look at is that the two dowel pins are each on its own concentric circle around the shaft and a certain distance from each other. If these two radiuses and the distance between pins are known it is easy enough to swing the three radiuses and lay out the triangle.

Most machine shops will have at least one milling/drilling machine and it will probably have a digital table position readout system. These machines require x y (Cartesian) coordinates. Your swung radius layout will need to be converted into x and y coordinates which can most easily be done with a simple CAD system or a little high school trigonometry.
Shaft (x=0), (y=0)
First dowel (x=0), (y=0 plus ?)
Second dowel (x=0 plus ?), (y=0 plus or minus ?)
The three question marks represent the only three important dimensions you will need in order to successfully mount to the transmission. The bolt holes you can transfer from the IMA motor/case mounted on the two dowel pins using a 13mm transfer punch or drill directly with a 13mm drill using the case as a drill jig. You will need to determine the location of the x and y axes on your plate which you can do with a cardboard template.

You can band saw, jig saw (or table/hand saw if it is aluminum) the outline of the plate or have it CNC machined or burned out depending on your budget and/or aesthetic needs.

If you will mount a motor on the plate you will have additional hole dimensions around the x y intersection. In addition you will have a hole there for the motor shaft. A bearing plate will be bored for bearing fit.

Also, you may need to consider the z axis or thickness of the plate.

What I'm trying to get at here is that we need to establish the three critical dimensions:... whether the three legs of the triangle or the Cartesian coordinates or both for the Insight so that those who would make bearing or adapter plates have a ready reliable reference.

Chris(Ma) kindly shared his dimensions with me and it was most helpful for me to know when I was on the right track. I invite all who have made these measurements to share them with us and publish them here. I will do so.

 

[terryb]

Instead of Stacking 4

instead of trying to adapt 4 Motors, what about mounting the 4 motors adapted to 4 axels to each wheel with no trans, just directly drive each wheel with feed back so all stay in sync? a simple cover on each side of the motor would alow for greater possibilities for mounting.

 

[Hugh-Falls]

Wheel motors?

instead of trying to adapt 4 Motors, what about mounting the 4 motors adapted to 4 axels to each wheel with no trans, just directly drive each wheel with feed back so all stay in sync? a simple cover on each side of the motor would alow for greater possibilities for mounting.

I have no idea of how to do what you suggest with the Insight IMA motors.

 

[Mark Elvin]

instead of trying to adapt 4 Motors, what about mounting the 4 motors adapted to 4 axels to each wheel with no trans, just directly drive each wheel with feed back so all stay in sync? a simple cover on each side of the motor would alow for greater possibilities for mounting.

e.g. Hub center motors.

 

[retepsnikrep]

Adapting IMA motors for hub drive has been discussed before, as have other hub motor options. Basically it looks too hard to pull off. Massive engineering reqd and complicated control electronics, as motors would not be in sync so would require an MCM and MDM for each motor + commutation sensors for each if you wanted to drive them with stock electronics. I'm going to press on with 5 stacked in engine bay. If you could make it work it would be a good option, feel free to have a go Terry If I win 135 million on lottery this week I might investigate it.

 

[Hugh-Falls]

Adapter Plates and Bearing Plates continued

A more direct way to machining plates will require some preparation and making up a cardboard template. A piece of thin poster board of sufficient size has a transmission shaft sized hole (.586") cut in center and is slipped into position on the transmission. With a small hammer tap gently over the ends of the dowel pins to transfer the impression. Be sure to mark which surface/side of board was against the transmission/clutch housing. Cut out dowel pin holes. Remount on transmission and trace outline of the housing. You can mount on the stripped IMA motor casings also and be sure you mount with the correct face turned toward the casing. You can cut out the traced shape of the plate and work out what size material you will need and how to situate it on you blank. You will need to draw a straight line across your pattern exactly through the centers of the two dowels and also draw this line on you blank and mark the centers of the dowels so that the Machinist will know how to orient the x axis of the blank on his machine.

If the Cartesian coordinates are set up so that the center distance between the two dowel pins is on the x axis and is the actual measured dimension on the part, the machinist who is to locate these two dowel pin holes can be asked to measure the distance on a stripped IMA motor casing that we send along with our drawing. He will be able to measure in his own way to make certain that he locates the holes so the parts will fit. This will eliminate any errors due to calibration of our respective measuring instruments and our measuring methods. We will know that at least the dowel pin holes will match up between plates and casings.

The coordinates for the shaft /bearing bore will be the intersection of the two swung radiuses (these are the two dowel pins to shaft measurements). The accompanying drawing shows the setup and the dimensions I used. Since I have not assembled my motor to date, the swung radius dimensions are unproven as are the resulting shaft hole coordinates so you should check them out for yourself before you use them.

The bolt holes will be transferred and drilled as previously described as will be the external profiling. If you find errors in these dimensions, please publish your results so that eventually we can come up with a set of refined and proven dimensions for bearing and adapter plates for our Insights.

 

[retepsnikrep]

I now have my fifth and final IMA motor so will be taking a trip to the machine shop with engine, gearbox and motors for a long planning meeting next week. I have printed out quite a few of the posts and diagrams/plans from these threads to discuss.

Thanks Hugh for that latest info. I think i mentioned I may/will have a bearing plate/spacer between 4th & 5th motors, so 5th Motor (that attaches to gearbox) appears standard and accepts clutch/gearbox etc without issue. This 5th (end) motor will also be the one with the commutation sensors as in the oem setup.

How accurately has anyone measured the available space between rotors when motors are stacked? You can't really see when they are pressed together to measure it very well? Ideas?

 

[Hugh-Falls]

Adapter plates and Bearing Plates Continued II

When I called the machinist to check on what he thought of the Cartesian coordinate dimensions, he said he did not use them because he felt he could get better accuracy on his particular equipment by a more direct method.

After he had measured the IMA case and drilled and reamed the two dowel pin holes in the bearing plates, he simply assembled the IMA case on the plates and located the shaft bearing bore directly by swinging his indicator around the inside of the stator. He said it was possible to get within .005 T.I.R. (total indicator reading). (Fortunately, because the IMA cases appear to be a little springy, I had bolted the stator into the case to stiffen it before I sent it to him).

At this point, none of the coordinate measurements I have written of here are confirmed to be accurate and will remain so until someone else comes up with a set. Also, it appears there may be no need for the coordinate dimensions at all depending on how you intend to machine the plates.

Both bearing plates were bolted and tension pinned together during the machining processes. The holes will be filled with epoxy before final assembly.

A tentative revised bearing plate drawing is attached.

 

[E27006]

measuring the gap

I think you can buy a crushable material for such engineering measurements.

Place a strip in situ, assemble/disassemble the units and then remeasure the strip.

In absence of the proper material perhaps blutack or plastercine might be adequate

 

[Mark Elvin]

It's called plastigauge, but it's pretty thin, used to measure bearing clearances on crankshafts & the like.

Have you thought of slip gauges if there is room to get in with some?

 

[Hugh-Falls]

Rotor axial offset

How accurately has anyone measured the available space between rotors when motors are stacked? You can't really see when they are pressed together to measure it very well?Ideas?

In the very first post of this thread (Aug. 26, 2010) I asked readers if anyone could furnish this offset dimension. Chris(MA) had referred to it in one of his posts but it may have become lost since then. Different writers speculated about how to get this dimension. I have measured it directly as the "end face of crankshaft to block mounting face dimension". It is what I used to determine the proper relationship between rotor and stator. Is this the dimension you require?

 

[Hugh-Falls]

Gap?

Peter, Mark and E27006: Please check out the drawing made by Chris(MA) for his "Bug Build" motor. You will see that what I understand to be the space you refer to as the "gap" is quite a bit beyond the range of Plastigage type material.

I may be wrong but check it out. You will need three dimensions to calculate the gap.
1.) Case width dimension.
2.) End face of crankshaft to block mounting face dimension. Also known as the protrusion of rotor mounting face beyond case mounting face dimension.
3.) Rotor thickness which is the dimension between the crankshaft mounting surface (face) and the flywheel mounting surface (face) of the rotor.

Add rotor protrusion dimension to case width dimension and subtract the rotor thickness dimension = offset ( item 2 plus item 1 minus item 3 equals offset).

To calculate the gap subtract the rotor thickness from the case width (item 1 minus item 3 equals gap).

 

[retepsnikrep]

I was thinking along those lines Hugh sounds very sensible I have booked the 18th july to speak with my machine shop and got a ton of papers and notes to take with me.

I don't clearly understand your assembly strategy from that long verbal description a few posts ago, do you have any mock up pics or diagrams of that.

 

[Hugh-Falls]

Terms definition

My understanding of some of these terms:

"Offset" is the distance from the case mounting surface toward inside of case to the surface of the rotor flange on which the flywheel mounts.

"Gap" is what I assume refers to the space between each of the consecutive rotors.

"Protrusion" refers to the distance the rotor crankshaft mounting surface is beyond the case mounting surface. The stator body also extends beyond (outside) of the case as well.

International standard power transmission practice for the precision mounting of pulleys, sprockets, and gears is to use tapered bushings to secure them to the shaft. There are at least three different standard bushing systems. Each system has its pros and cons. For the drill jig, the "split taper bushing" appeared to be best for synchronizing accuracy. For mounting the rotors, the "taper bore" bushing was the most compact. You may want to familiarize yourself with these bushing systems if you plan to mount the rotors on a through shaft. You may want to eliminate an adapter entirely and simply taper bore the rotor for one of the bushing systems. This would permit use of a larger diameter through shaft which would contribute rigidity.

Much of this material is available as industrial surplus. The bushings are best purchased new because the new/surplus price difference tends to be insignificant and the quality of the new part is assured. Standard roller chain "taper bore" sprockets can be modified easily to adapt the rotors to a standard keyed shaft for instance.

High quality mounted piloted flange sealed/lubricated ball bearings can simplify machining and assembly. Sealmaster Skwezloc is one of the better brands of sealed ball bearings units. It is best to avoid bearing press fits and other tight fits because in our type of ad hoc engineering frequent assembly and disassembly may become necessary.

 

[Hugh-Falls]

Assembling the stacked IMA motor

............I don't clearly understand your assembly strategy from that long verbal description a few posts ago, do you have any mock up pics or diagrams of that.

Assembling and taking apart the stacked IMA motor does present some problems. I have nothing worthwhile to share with you at this time. The solutions I've come up with so far require fixtures that I hope to eliminate if at all possible. It is just going to take some more time and effort.

What do you think of operating the stacked IMA motor at a lower voltage to begin with? This over sized motor should drive the Insight moderately well even at a pretty low voltage; well enough for local short range trips. It could use a small light battery pack during the early developmental stage.

 

[retepsnikrep]

What do you think of operating the stacked IMA motor at a lower voltage to begin with? This over sized motor should drive the Insight moderately well even at a pretty low voltage; well enough for local short range trips. It could use a small light battery pack during the early developmental stage.

Will you be using the stock electronics or some other motor controller? You can use whatever suits your controller, The motors won't care.

As i will me driving the motor with Honda electronics i will use a standard battery/swichboard assembly voltage and mcm/bcm to test drive one or more of the stacked assembly on the bench. I'll probaly also do the same with the final install driving one motor then adding others motors in one at a time to the mix and testing in between each.

Each of my motors will have a dedicated MDM (contents) i.e. IGBT, Filter caps, snubber, and voltage converter module operating off a central DC power bus. Only the master motor will have phase current sensors. I may trying running two motors in parallel off one IGBT later. That would reduce my IGBT count to 3 instead of 5.

 

[Hugh-Falls]

Peter: I asked the wrong question. Has anyone determined what the lowest voltage is at which the stock Insight MDM can be operated successfully without modification?