MAC GMAC Hot Rod Info

[Bullfrog]

Neppie...I did not know that, thanks. I knew I liked you for some reason :lol:.

The GMAC is a great motor and Justin has made some good improvements but it is not available in a front/100mm and I asked if Grin Tech could get me a front MAC but unfortunately they declined.

Grin Tech is a little pricey but you get what you pay for IMO...plus their customer service is great and they provide the forum we are communicating on and Justin has always been super helpful to me and the entire ebike community...he has produced some really good youtube videos lately that explain the intricacies of electric motors and IMO they should be mandatory viewing for anyone interested in ebikes. This is one good example for EVERYONE: https://www.youtube.com/watch?v=c96n0Ma2rLY&t=4858s

I don't doubt Grin sells good batteries but I still buy my batteries from Paul (you did good with the name )...everything else I get from Grin Tech for the most part and why I'll probably go with the Shengyi motor. With a kick bike, I don't need the torque a MAC or an eZee can produce and I am going as light as reasonably possible with the Kick Bike.

 

[neptronix]

Grin can be pricey but it's a good place to go when you want quality stuff that's hand picked by a guy with a discerning eye and you don't want to spend days of research.

I like doing my own research and try to spot unknown killer products, but Justin is really competing with me on that these days and often has something in stock and ready to sell before i've even heard of it :thumb:

On a kick bike like yours, the center of gravity is really poor, so i wouldn't go with super high power anyway!

 

[Jil]

Thanks a lot for your answers everybody
I've tried the motor simulator, it appear that either with 52V or 72V battery, I can run a bit higher than 50 kph continuously without overtheating (flat land, no wind). But from my experience the 52V battery voltage goes quickly donw below 50 V, thus limiting a bit the max speed. And the amperage is also higher with the 52V, leading to a faster heating of the Phaserunner.
I am aware of the limitations of this setup for my use, I will try first a 72V bigger battery, to see the impact.

 

[Bullfrog]

Thanks a lot for your answers everybody
I've tried the motor simulator, it appear that either with 52V or 72V battery, I can run a bit higher than 50 kph continuously without overtheating (flat land, no wind). But from my experience the 52V battery voltage goes quickly donw below 50 V, thus limiting a bit the max speed. And the amperage is also higher with the 52V, leading to a faster heating of the Phaserunner.
I am aware of the limitations of this setup for my use, I will try first a 72V bigger battery, to see the impact.

Jil...I sent Grin Tech an email because I wasn't sure going to 72v will help your Phaserunner...below is a quote from their response.

"at the same speed, the system would need the same amount of phase current and it's independent of the battery voltages. Therefore the Phaserunner would be as hot with a 72V battery as a 52V battery."

Your battery amperage/current will be lower but the phase current will be the same. Your controller is basically a DC to DC converter and takes the input from the battery and drops the voltage and increases the amperage.

Sorry to be the bearer of bad news but hopefully I saved you some money and frustration.

 

[neptronix]

The phaserunner's maximum continuous power handling is close to a 9FET controller even though it only has 6FETs, due to it's optimal design.

Not really ideal for a hot rod bike.

 

[Bullfrog]

Jil...I know the Phaserunner is nice and small but if you want to go faster or accelerate quicker, Neptronix has an excellent point.

I like a 12 or 18 FET Infineon Controller for a MAC. I use a 12 FET with 3077 FETs which is more efficient than the 4110 FETs but the 3077 can't handle more than about 60v so if you think you ever might go to a 72v battery, I'd go with 4110 FETs. With the 8T MAC you may want to go with something that can provide a good bit of phase amperage like the 18 FET.

Only problem is finding one...since EM3ev stopped selling motors/controllers, I have not found a good place to buy them except directly from MAC and you can only do that once before they require you to buy a minimum of 8 pieces so use your one time "sample" buy carefully .

 

[pwd]

Jil...I know the Phaserunner is nice and small but if you want to go faster or accelerate quicker, Neptronix has an excellent point.

I like a 12 or 18 FET Infineon Controller for a MAC. I use a 12 FET with 3077 FETs which is more efficient than the 4110 FETs but the 3077 can't handle more than about 60v so if you think you ever might go to a 72v battery, I'd go with 4110 FETs. With the 8T MAC you may want to go with something that can provide a good bit of phase amperage like the 18 FET.

Only problem is finding one...since EM3ev stopped selling motors/controllers, I have not found a good place to buy them except directly from MAC and you can only do that once before they require you to buy a minimum of 8 pieces so use your one time "sample" buy carefully .

FYI, I purchased the 6fet "500W" one from here: https://www.aliexpress.com/store/group/BLDC-bike-motor-controller/937734_253125770.html?spm=a2g0o.store_home.pcShopHead_11014174.1_0 and it is an Infineon 4 clone (KH6XX) and work very well. By the looks of them, their larger controllers appear to be Infineon 4 clones too.

 

[Jil]

Jil...I sent Grin Tech an email because I wasn't sure going to 72v will help your Phaserunner...below is a quote from their response.

"at the same speed, the system would need the same amount of phase current and it's independent of the battery voltages. Therefore the Phaserunner would be as hot with a 72V battery as a 52V battery."

Bullfrog, many thanks ! I really appreciate your help, you saved me time and money :thumb:
It's bad news indeed, so I understand I should work more on the cooling of the Phaserunner... or use it more wisely :wink:

 

[Bullfrog]

Jil...I sent Grin Tech an email because I wasn't sure going to 72v will help your Phaserunner...below is a quote from their response.

"at the same speed, the system would need the same amount of phase current and it's independent of the battery voltages. Therefore the Phaserunner would be as hot with a 72V battery as a 52V battery."

Bullfrog, many thanks ! I really appreciate your help, you saved me time and money :thumb:
It's bad news indeed, so I understand I should work more on the cooling of the Phaserunner... or use it more wisely :wink:

You are very welcome....mounting it on a finned block of copper would be best (aluminum would be almost as good and cheaper) or maybe a small fan like used to cool the inside of a computer...or both .

I mounted my Infineon controller on the front of my head tube to ensure it got a lot of air flow.

 

[colsaunders2]

Despite a wheel rebuild, split open torque washers, and 2200 watts of peak power, the frankenMAC 8T lives with over 2000 miles on the odometer. I have the Baserunner L10 programmed to 40A battery/80A phase and 6A field weakening; that translates to 40 mph on the flats and a personal best of 44.3 mph.

Seeing as neptronix's speed record still stands, some mods are in order:

1. Full-synthetic Dexron VI ATF for cooling. I've seen posts about long-term effects on wire insulation, etc, but I've found nothing more specific than that oils can "wash out" the plasticizers. Maybe I'll toss some old wires in a jar with ATF for science. "Frank" is clutch-free and regen-enabled, so I'm not worried about stray fluid on the rear brake rotor.

2. The Baserunner came in a Hailong base plate, which has worked well as a heatsink bolted to the bottle mount behind my triangle battery back. To improve thermal conductivity, I can use Arctic MX4 between the mating surfaces, and then toss in a thermal pad between the heatsink and bicycle frame.

That should afford some headroom to increase field weakening. Recently I took the shoulder of the nearby interstate/highway/state route (it's co-numbered) to the next exit so I could actually get somewhere without having to drive or ride twice as far. Oh, I definitely kitted up for that excursion! Can we hurry up and legalize high-speed DIY builds already? Promise I'll be safer than the dude on the motorcycle who's not even wearing a helmet!

 

[markz]

Xie Cheng controller not Infineon, MCU chip is infineon used inside.
Xie Cheng sells controllers and anything else. I see their cn website is no longer.
Others have modded them and sold them (Keywin, Lyen, Perhaps even Telsanv, Luna) with beefed up traces, silastic for vibration control and better components/crtmodules.

Jil...I know the Phaserunner is nice and small but if you want to go faster or accelerate quicker, Neptronix has an excellent point.

I like a 12 or 18 FET Infineon Controller for a MAC. I use a 12 FET with 3077 FETs which is more efficient than the 4110 FETs but the 3077 can't handle more than about 60v so if you think you ever might go to a 72v battery, I'd go with 4110 FETs. With the 8T MAC you may want to go with something that can provide a good bit of phase amperage like the 18 FET.

Only problem is finding one...since EM3ev stopped selling motors/controllers, I have not found a good place to buy them except directly from MAC and you can only do that once before they require you to buy a minimum of 8 pieces so use your one time "sample" buy carefully .

 

[99t4]

Maybe I'll toss some old wires in a jar with ATF for science.

Don't forget to heat it to the temperatures that the hub interior sees. :wink:

 

[colsaunders2]

Maybe I'll toss some old wires in a jar with ATF for science.

Don't forget to heat it to the temperatures that the hub interior sees. :wink:

Well, I'm afraid the ATF won't have time to settle in and attack anything. "Frank" ejected nearly all of the fluid within 24 hours. I probably should have borrowed Bullfrog's lab coat and notes seeing as how he's the scientific one here, but this is what I did:

Rode 1.6 miles around the neighborhood, starts, stops w/ regen. Motor shell was 82 F before and after, basically ambient temperature.

I dismounted the wheel and loosened 2 of the disc rotor bolts, 1 to insert ATF and the other to vent air. Used a syringe to inject about 3 oz. Some of it seeped out the other side thru the freehub and pilot hole (axle), no biggie. Ride #2, just 1 mile. Motor shell was 92 F - OK, it's working.

Spun up the wheel a few times off the ground. Power consumption was about double (330 vs 160 watts) at full speed. Mind you, it's going 56 mph. Shell was 100 F and rising. Getting some dribble from the motor cable.

I don't have a thermistor installed, so my theory was that the improved conduction via ATF would result in a higher shell temp vs running dry. Confirmed! :thumb:

Early this morning, I rode about 6 miles. Seemed a bit "draggy". Upon completion, I noticed that a dark sludge had sprayed out from between the shell and freehub, all over the wheel. Oops.

Rode to work and left the Deathcycle outside in the sun. 7 hours later, I find a rotor coated in ATF. Looks like it sweated out thru the bolt holes. Wipe 'er up, brake still works but I got regen anyway.

And, ATF is still dripping onto the garage floor. Thermal expansion is no joke. There seriously can't be more than half an ounce left.

Conclusion: you need to seal every bloody surface if you want to run the MAC wet. Not sure that it's even feasible if you have a side-exiting L10 cable like I do :lol:

 

[colsaunders2]

Has anyone added a thermistor or temp sensor to their MAC after purchase? I ordered a pack of NTC 10K B3960s for cheap; the videos/pictures I've seen of MAC/GMAC internals show insulated wires soldered to 2 adjacent pins in the LM35 portion of the PCB, which implies they're using a thermistor - not an IC.


Mine is also lacking an internal speed sensor, but Frank is clutchless so I could theoretically bridge the green hall pin to speed. Then I would run 2 wires from the PCB and splice with the currently-snipped gray and white wires. Anyone know if the Cycle Analyst can handle 653 pulses/sec on the speed wire? :lol: That's 40 mph on a 27.3" wheel, in case you were wondering.

 

[Bullfrog]

Just a little info on reaching the practical limits of a MAC/GMAC....

I plan to try a 72v battery, a controller set to 65A battery/180A phase, 12T MAC, 686mm OD rear tire. Based on many runs using the Grin Tech Motor Simulator, IMO that is about the limit of the MAC design before magnetic saturation starts to make more amperage impractical.

The 12T MAC is the only MAC/GMAC that I'd consider running on 72v, all of the others will have too much heating due to internal losses and you may have over heating problems if you run it hard without pedaling.

The best set up for most will probably be something like a 10T MAC/GMAC powered by a 52v battery with a maximum of 40A...that should run great and have plenty torque .

 

[Bullfrog]

Has anyone added a thermistor or temp sensor to their MAC after purchase? I ordered a pack of NTC 10K B3960s for cheap; the videos/pictures I've seen of MAC/GMAC internals show insulated wires soldered to 2 adjacent pins in the LM35 portion of the PCB, which implies they're using a thermistor - not an IC.
therm.jpg

Mine is also lacking an internal speed sensor, but Frank is clutchless so I could theoretically bridge the green hall pin to speed. Then I would run 2 wires from the PCB and splice with the currently-snipped gray and white wires. Anyone know if the Cycle Analyst can handle 653 pulses/sec on the speed wire? :lol: That's 40 mph on a 27.3" wheel, in case you were wondering.
mac.jpg

I have always ordered my MAC/GMAC with a temp sensor installed...I'd recomend checking with Grin Tech.

For a speed sensor, the Cycle Analyst has a magnetic pick up to take an input from your wheel and calculate/display speed. I usually screw something up if I start soldering on the PCB :lol: .

 

[colsaunders2]

A belated update here:

DIY thermistor upgrade was successful. However, the right-side bearings fell apart when disassembling the motor. This is why it's critical that you use the jam-nut on the axle inside the freehub. Otherwise you place excessive lateral forces on the bearing, and eventually metal bits get into the core etc.

ATF is bad news. It mixes with anything already in the motor and flies out at high velocity, mainly from the freehub area. Black sludge everywhere. Even got into my spokes and ruined a tube. On the plus side, it was an opportunity to re-grease the gears.

Last month my Baserunner died. Grin was awesome and sent me a replacement. It's good that I hadn't tried to hack a speed sensor into Frank, because BR #2 doesn't have speed pass-thru (it's what was available, and my request to pay the difference for the Phaserunner was politely declined).

Another thing that's probably obvious to many here: you need lock washers, and good torque washers too (love the Grin ones). Treat your torque arm as a fail-safe, unless you're into high-torque applications that exceed all of the above. I hadn't been using them and just about trashed my sliding dropouts - thankfully I was able to replace them. Nord-lock is probably best but split washers seem to work OK. I don't have to tighten my bolts so often now.

3100 miles (5000 km!) so far. I'm with Bullfrog, Mac motor is best motor.

 

[Bullfrog]

A belated update here:

DIY thermistor upgrade was successful. However, the right-side bearings fell apart when disassembling the motor. This is why it's critical that you use the jam-nut on the axle inside the freehub. Otherwise you place excessive lateral forces on the bearing, and eventually metal bits get into the core etc.

ATF is bad news. It mixes with anything already in the motor and flies out at high velocity, mainly from the freehub area. Black sludge everywhere. Even got into my spokes and ruined a tube. On the plus side, it was an opportunity to re-grease the gears.

Last month my Baserunner died. Grin was awesome and sent me a replacement. It's good that I hadn't tried to hack a speed sensor into Frank, because BR #2 doesn't have speed pass-thru (it's what was available, and my request to pay the difference for the Phaserunner was politely declined).

Another thing that's probably obvious to many here: you need lock washers, and good torque washers too (love the Grin ones). Treat your torque arm as a fail-safe, unless you're into high-torque applications that exceed all of the above. I hadn't been using them and just about trashed my sliding dropouts - thankfully I was able to replace them. Nord-lock is probably best but split washers seem to work OK. I don't have to tighten my bolts so often now.

3100 miles (5000 km!) so far. I'm with Bullfrog, Mac motor is best motor.

colsaunders2...thank you for the update and the info.

If you are going to use any kind of cooling fluid including ATF, I'd recomend thoroughly cleaning everything on the interior as spotless as possible. I went through a lot of paper towels when cleaning mine :lol: .

If you add more than about 4.5 ounces of any fluid you will go above the axle seal and probably have a serious leak. I sealed the cover/motor interface with Permatex Ultra Gray which worked great. 4.5 ounces adequately connects the windings to the hub so heat can be rejected to the atmosphere so more than about 4.5 ounces doesn't increase the cooling capability much if any. If you remove a disc brake bolt and rotate it so it is at the lowest point possible, you can add fluid through the hole until it just barely runs out...that is about 4.5 ounces. Of course don't forget to put the brake rotor bolt back in . I used a turkey injector to squirt fluid in the brake rotor bolt hole.

Where I had a leak with ATF was it got between the copper wires and the insulation covering them and was pulled out of the motor via capillary action. I never had a leak at the axle seal or the cover interface.

I have never actually touched a GMAC motor but Justin tells me it is sealed a lot better than the MAC.

The reason I went to distilled water and Motul MoCool was because if it leaked, it did not make a mess and was not a problem if it got on my brake rotor.

After all of the cooling experiments I did, I'd recomend cleaning the interior thoroughly and covering the ring gear teeth with Mobil 1 SHC 100 grease. Just make your motor selection i.e. 12T, 10T, etc. smartly and run it in a wheel small enough in diameter where additional cooling is not required.

The Grin Tech Motor Simulator is a huge help for predicting temperatures and you can run up to about 140C without any damage to the motor. I use 2% grade in my simulations to get realistic data.

 

[Bullfrog]

I just tried my 12T MAC on a 20s/72v battery and an 18 FET Infineon Controller. The maximum battery amperage my Cycle Analyst registered was 77A and the Grin Tech Motor Simulator suggest the maximum phase amperage would be ~160A.

My one word summary of the test is "Disappointing"...even though the Grin Tech Motor Simulator suggested ~20% increase in acceleration, my seat of the pants dyno did not register any difference between the 72v set up and my 52v set up which had a maximum battery amperage of ~58A which equates to ~140A phase current.

My top speed was ~31 mph which matches the Grin Tech Motor Simulator pretty good.

At some point, all electric motors no longer have a linear relationship between amperage and torque as the amperage is increased and I believe I have passed that point for my MAC. Based on my testing, battery amperage above and beyond about 50A does not do much more than produce excess heat...of course the relationship (amperage/acceleration=constant) becomes more and more non linear the higher you go with the amperage so there is no defined point where you can say it occurs.

Based on everything I have learned to date, I have to say the best set up is a 10T or a 12T MAC (depending on your desired top speed) in a 27.5" rim (mainly for tire selection) and running a 52v battery with -25R/HG2 (or similar high amperage cells) and a controller that can handle ~50A of battery current and supply ~140A of phase current.

Something I did not address above is the potential for over heating when you drive a 12T MAC with a 72v battery. Ensure you monitor your motor temperature if you attempt anything similar to what I describe above.

Back to the drawing board :lol: .

 

[neptronix]

Try that same test on an 8T with higher amps and let me know if you're still disappointed

Yeah with a winding like that, more volts = more power way more than more amps = more power.

 

[Bullfrog]

Try that same test on an 8T with higher amps and let me know if you're still disappointed

Yeah with a winding like that, more volts = more power way more than more amps = more power.

The problem I had with my 8T MAC was it over heated...and that was steady state on 52v .

I wouldn't mind trying it on 72v but I sold it due to the over heating.

Now I only have a 10T and a 12T.

As much as I rave and preach about how great the MAC/GMAC is, you do have to use them appropriately or like any geared hub motor they will tend to over heat .

 

[neptronix]

The problem I had with my 8T MAC was it over heated...and that was steady state on 52v .

I wouldn't mind trying it on 72v but I sold it due to the over heating.

Makes sense. no matter the winding, that shell just can't diffuse enough heat to cople with the load at those speeds.

A MAC motor with half the pole count, but the same dimensions, is rated as being 2kW. that tells you how thermally handicapped the hub version is!!

 

[neptronix]

If you add more than about 4.5 ounces of any fluid you will go above the axle seal and probably have a serious leak. I sealed the cover/motor interface with Permatex Ultra Gray which worked great. 4.5 ounces adequately connects the windings to the hub so heat can be rejected to the atmosphere so more than about 4.5 ounces doesn't increase the cooling capability much if any. If you remove a disc brake bolt and rotate it so it is at the lowest point possible, you can add fluid through the hole until it just barely runs out...that is about 4.5 ounces. Of course don't forget to put the brake rotor bolt back in . I used a turkey injector to squirt fluid in the brake rotor bolt hole.

Hey Bullfrog..
I'm thinking of getting into fluid cooled geared hubs.. and you seem to know some stuff

What would be the ideal amount of whatever kind of fluid in a MAC? I'm guessing that 2oz would be just enough to form a thermal bridge and stay well below the axle.

Also.. how is your fluid cooled motor doing? How often is it dripping etc after you sealed the covers?

I'm wondering if this is a daily driver-able thing when done properly.