I am thinking about doing an electric supercharger project. The idea is to use a turbo compressor driven by a brushless DC hobby motor (I have found one that will do 12 kW (16 hp) at 85,000 rpm briefly). The target would be 250 - 300 hp at 7 psi boost (1.5 PR). The problem (?) is most turbo compressors are designed for best efficiency at 2 - 2.5 PR.

Compressors that come close are Garrett 58/76 or the similar BW 7675 (EFR 7670). I have plotted the engine in question on this compressor using Matchbot here https://www.borgwarner.com/go/TXCQKH. As you can see the match plots well below the efficiency island although 60% - 70% is not too bad. The 76mm tip diameter is about as small as you can go and still make reasonable boost at 80k rpm - bigger would be better.

Are there any compressors out there that are a better match for this application? Don't really want to go to Vortech with their massive volute.

 

I tweaked things a bit to resemble a normal engine, if your engine is really how you had matchbot setup let me know.
https://www.borgwarner.com/go/7HE2VA

But I would go 7163, slightly higher efficiency which means it'll take less power. However the hp posted on matchbot is just for the comp, you still need to figure for drivetrain loss, unless you are fixing the comp right to the motor.

What voltage do you have figured? The high hp motors are usually on 36+ volts.

 

Someone on Corral is doing this but I think he may be using a blower impeller, not sure. If you're not aware do a search over there.

ks

 

The engine is a Toyota 2ZZ-GE
190 hp @ 7,500 rpm NA.

The Brushless DC motor I mentioned is here https://www.aliexpress.com/item/1005003765803916.html?spm=a2g0o.productlist.0.0.42f57900lCeGLK&algo_pvid=1f0665f7-6e15-48e5-b2dc-d6cf1095fbc8&algo_exp_id=1f0665f7-6e15-48e5-b2dc-d6cf1095fbc8-5&pdp_ext_f={"sku_id":"12000027092537691"}&pdp_npi=2@dis!AUD!!505.1!!!31.46!!@210318c916537996086192399ed12c!12000027092537691!sea
Max 12 kW (16 hp, Your point #6 @ 16.7 hp is a touch high)
Max 85,000 rpm
Voltage required is 8S - 16S (34 - 68 volts) depending on motor model. I would tend toward the 68 volt version to reduce the current and wiring requirements. Something like this will need a bank of LiPo cells and a charging system that will recharge from the car battery when the supercharger is not loaded. Since you can only draw about 1/2 kW from the car alternator, 10 seconds at 12 kW will need about 4 minutes off boost for the LiPo's to recharge. Fortunately 10 AH is common for LiPo cells - which would deliver the required current (169 A) for over 3 minutes.

 

Here's his newest vid:
https://www.corral.net/threads/electric-turbo-on-a-363-dyno-tested.2514550/#post-18730469

ks

 

Yeah, I've been watching his progress from the start. He is driving the electric motor at up to 34 kW in that video. Of course with an engine that (NA) pumps more than twice my 2zz he needs a much bigger compressor. (A Vortech in his case)

 

Yeah - direct drive 85,000 rpm max.

 

I wouldn't count on the burst watts, I would size for continuous, maybe a slight bump higher.
Me personally, I would use 2 motors on a splitter gearbox. Then you could use angular contact bearings that will handle the thrust load much better than the deep groove in the bearings.
I'd use vesc's, I believe they have a 345A option, cool thing with these is they have alot more options than the standard esc's

 

I can do without the complexity and space demands. The burst ratings are 10 seconds which is way more than I need. Even the system in the video took a lot of time and effort to develop - uses 1 motor, direct drive, burst rating . .

 

Not really that complex, the right angle gearbox in my turbine rc heli uses angle grinder gears to go from ~90k rpms to 20k.

At 85k rpms the bearing inner races will already be expanding from the g forces, plus the balls g forces themselves from spinning at ~28k, then your going to add quite a bit of axial load also.

Going with a splitter box will only have a radial load on the motors, while all the thrust will be on the angular contact bearings. Plus, each motor would only be doing half the work, so much better life out of them and the electronics.
Or find a motor that has more power, I just burned up a motor by pushing it to spin a tiny hydraulic pump.
It's cool what your wanting to do, I just wouldn't rely on a Chinese motor being pushed to its max

 

I watched some of his previous vids, and yea, he isn't even running his motor at max, or the esc, yet alone at burst rating. He's also at 31k rpms because he's using a compressor designed for s/c use, very very low trim, almost no backsweep on the blades, big exducer in order to get pressure ratio up at lower rpms. The early solar turbos would be a good place to start looking, as they are designed almost the exact same way, or order an s/c wheel. Atleast with a solar compressor you could use the backplate and compressor cover

 

I am not familiar with Solar turbos.

 

A solar 3LDA-168 has a really low trim (36) wheel, that's more inline with what a s/c would use.
It's got a 40.7mm inducer, id have to some calcs to see what compressor speed would be necessary to move the required ~30-33lbs/min.

 

Inducer would need to be closer to 50mm to flow 250 - 300 hp at 1.5 PR.

 

Just spotted the Garrett 56/76 from the GT3076R. Has better efficiency at the 1.5 PR region.

 

Yep, much better efficiency, getting close to your 85k limit, but your now down to ~14.6hp/11kw. Assuming voltage drop to 3.9v/cell your at 62v, 78,000rpms ÷ 62 = 1258kv. So the 16s/1200kv motor should juuuust get you there @ ~180amps, it's totally doable. But I wouldn't expect it to like it and live there reliably, better order 3 motors. You won't need any kind of crazy batteries, the more you have tho the lesser the voltage drop.

Watched another vid of that sledgehammer, and he's using angular contact bearings like I mentioned. The motors only job is to spin, no other loads on it at all. So a simple preloaded 2 bearing shaft and tubeless is all that is needed. 85k is nothing on ceramic angular contact bearings.

Also, I calculated that with a compressor inlet flow speed of 700ft/sec (typical max flow/speed) the 40mm inducer wheel CAN deliver 33.4lbs/min of air. Problem is that the large exducer limits compressor speed to 140k rpms, so that's a no-go.
I second the 58/76 garrett wheel...

 

You mean the 56/76 wheel of course?

Meanwhile I have been looking for wheels with similar inducer diameter but larger tip diameter to reduce wheel speed at the 1.5 PR target. Found these Holset wheels but can't find compressor maps.

50/83, Tip height 6.0 (H1C) https://www.ebay.com.au/itm/281913955507
52/83 Tip height 6.05 (H1C) https://www.ebay.com.au/itm/281937058495?epid=1837350279&hash=item41a4c21abf:g:14EAAOSwgkRVULNi
54/83 Tip height 6.05 (H1C) https://www.ebay.com.au/itm/262250184294?hash=item3d0f543e66:g:OiEAAOSwv0tVWX5c
56/83 Tip height 6.35 (H1C) https://www.ebay.com.au/itm/263830800139?hash=item3d6d8a870b:g:zt0AAOSwBahVVX61
56/83 Tip height 6.3 (HX35) https://www.ebay.com.au/itm/252257887753?hash=item3abbbde609:g:5xYAAOSw7PBTmgxg

Can anyone help with maps for any of these?

 

Where are you seeing 56/76 wheel? I'm looking at garretts site and don't see anything with 56mm inducer. But if you go there, check out the 67mm gt3584, your spot is at peak efficiency and only 67k rpms.

Going larger exducer will help slow rpms for the same pr like you said, but now the inducer won't be breathing the same amount of air, pushing you closer to the inefficient choke side. Also, tip height plays a big role as well. The exducer has an outlet area, reducing tip height reduces the area, increasing velocity out of the wheel, which increases pr. I only mention it because you posted alot of wheels with different tip heights.

Blade backsweep is a very large component of rpm vs pr, the more backsweep, the more rpms needed for same pr. Early turbos were very rpm limited, and needed all the help they could get. Here's an example of one, which is also similar to most centrifugal s/c wheels.
https://www.ebay.com/itm/253232051313?hash=item3af5ce7471:g:pQkAAOSwQcJZ9ciu

 

https://www.garrettmotion.com/racing-and-performance/performance-catalog/turbo/gt3076r/

Sorry 57/76

 

Another compressor wheel worth considering. https://www.ebay.com.au/itm/262249446290?mkevt=1&mkpid=2&emsid=e112358.m43.l1123&mkcid=8&bu=44936883085&osub=-1~1&segname=16TE179852_T_PREPURCHASE_CT1&crd=20220613090000&ch=osgood
Anyone have a map?

#ad