My 84-volt Nikola Triumphs, Tribulations, and Failures

[Lucas Alexander Oliver]

On 6/13/2019 at 12:14 PM, Dzlchef said:

@Marty Backe the wheel charged fine although the battery seems to carry me about as far as my KS16S, albeit much smoother and faster.   I need to learn to start slower and keep a consistent cruise speed.  

https://euc.world/tour/575803514399088

Wow. That’s really surprising.  I have a ks16s as well and can only go about 15 miles before the battery is so low I don’t wanna ride it anymore.  But my fist ride on the nikola I started at 90% and went 22 miles. A mixture of fast stop and start and 25 mph riding and I only went down to 46 % and I’m 6’3” and 200lbs.    So I will get at least 45-50 miles out of her 

[EU_commuter]

@Marty BackeWhich wheels survived the Overheat Hill?  MCM5 certainly.  MSX?  How about any Kingsong wheels?  Thank you.

[The Fat Unicyclist]

6 hours ago, meepmeepmayer said:

FFFFFFFUUUUUUUUUUUU

Hi! Was someone wanting me?   

[Marty Backe]

36 minutes ago, EU_commuter said:

@Marty BackeWhich wheels survived the Overheat Hill?  MCM5 certainly.  MSX?  How about any Kingsong wheels?  Thank you.

Personally I've never had another wheel fail on Overheat Hill. I've had wheels overheat though. I've seen KingSong wheels blow a fuse, and a ACM2 blow a MOSFET (not my wheel). The MSX flew up the hill no problem. The KS18XL (18S too) made it up the hill, but overheated.

[Marty Backe]

 

 

 

[Kens]

38 minutes ago, Marty Backe said:

 

OMG we can put some IKEA furniture inside that shell 

Thank you Marty for your sacrifices and R.I.P Nikola. 

Btw will 100v control board works with 1600wh battery? 

[stephen]

Great video marty i had to watch the cut out couple of times ,,,Ouch! When they go they go ☹️with your weight lent forward you just hit the ground flat on ,, hope your ok and nothing hurt but the nikola pride,  at my rider weight,,200+lbs I'd have no chance

I think with your riding conditions and hill tests @Jason McNeil wood swap the board or return for exchange for the nikola plus  i think the Nikola plus be the better option , it's a shame cause you was enjoying the wheel so much just let down by the small mosfets , I'll definitely watch my beeps on hills just shows how much current going through then boards,

Hope you and Jason come to some deal , thanks for the video plus the opening of the wheel and seeing the board 👍👍👍👍👍👍👍👍👍👍👍👍👍👍👍👍👍👍 thanks again 👍👍👍 a million likes 👍👍👍👍👍

Edited June 15, 2019 by stephen

[maltocs]

55 minutes ago, Marty Backe said:

 

 

 

So sorry to hear of this failure for you and for Gotway. Lotta beeping going on there! You said you set the alarm at 90A. I'm not great with electronics, but would you set it at the same on a 100V wheel? Arrrg, so disappointed at the smaller mosfets!

[Marty Backe]

2 minutes ago, maltocs said:

So sorry to hear of this failure for you and for Gotway. Lotta beeping going on there! You said you set the alarm at 90A. I'm not great with electronics, but would you set it at the same on a 100V wheel? Arrrg, so disappointed at the smaller mosfets!

Yes, the 100-volt wheel would probably get a smaller setting for the current alarm. I'd have to experiment a bit to find the right value.

This wheel is definitely drawing a lot more higher currents than any other Gotway wheel I've ridden. That probably explains why it has such great torque. But Gotway really screwed up by going cheap on the MOSFETs. It's a mystery 

[Marty Backe]

9 minutes ago, stephen said:

Great video marty i had to watch the cut out couple of times ,,,Ouch! When they go they go ☹️with your weight lent forward you just hit the ground flat on ,, hope your ok and nothing hurt but the nikola pride,  at my rider weight,,200+lbs I'd have no chance

I think with your riding conditions and hill tests @Jason McNeil wood swap the board or return for exchange for the nikola plus  i think the Nikola plus be the better option , it's a shame cause you was enjoying the wheel so much just let down by the small mosfets , I'll definitely watch my beeps on hills just shows how much current going through then boards,

Hope you and Jason come to some deal , thanks for the video plus the opening of the wheel and seeing the board 👍👍👍👍👍👍👍👍👍👍👍👍👍👍👍👍👍👍 thanks again 👍👍👍 a million likes 👍👍👍👍👍

Thanks a lot for the kind words Stephen. I'm OK. Had all my protections on. I'd have bloody knees if not for the knee pads and I'm sure the wrist guards came into play.

Jason's taking care of me. No real question there. I have learned my lesson that I will not buy another Gotway wheel that doesn't use the large MOSFETs!

[Marty Backe]

29 minutes ago, Kens said:

OMG we can put some IKEA furniture inside that shell 

Thank you Marty for your sacrifices and R.I.P Nikola. 

Btw will 100v control board works with 1600wh battery? 

No, the 100-volt wheels use a different battery configuration.

[maltocs]

2 minutes ago, Marty Backe said:

Yes, the 100-volt wheel would probably get a smaller setting for the current alarm. I'd have to experiment a bit to find the right value.

This wheel is definitely drawing a lot more higher currents than any other Gotway wheel I've ridden. That probably explains why it has such great torque. But Gotway really screwed up by going cheap on the MOSFETs. It's a mystery 

Thanks for the response from an engineer's perspective. And thanks for the test which resulted in this huge sacrifice.

btw, I need to know what "serenity now" pills you are taking. Your response after the face plant was "well, " then you chuckled a bit, "not a good sign guys, I was hoping that wasn't going to happen" in the calmest voice possible.

Not a curse word in sight! AMAZING.

 

[Marty Backe]

2 minutes ago, maltocs said:

Thanks for the response from an engineer's perspective. And thanks for the test which resulted in this huge sacrifice.

btw, I need to know what "serenity now" pills you are taking. Your response after the face plant was "well, " then you chuckled a bit, "not a good sign guys, I was hoping that wasn't going to happen" in the calmest voice possible.

Not a curse word in sight! AMAZING.

 

I'm a practicing Stoic, literally

Edited June 15, 2019 by Marty Backe

[Unventor]

So @Marty Backe now that you have had a little time to think, what happens with your NIkola. Is it going to be your city wheel? I couldn't imagine you taking it to mountain trails after what happened.

Then again it might be a weak component that didn't live up to expected standard I guess. These thing does happens. 

I didn't know you got hurt in the process. I hope you are OK. Good thing you had protection gear on. 

Again cummunity comfort pad on the shoulder to you. 

 

[Marty Backe]

2 minutes ago, Unventor said:

So @Marty Backe now that you have had a little time to think, what happens with your NIkola. Is it going to be your city wheel? I couldn't imagine you taking it to mountain trails after what happened.

Then again it might be a weak component that didn't live up to expected standard I guess. These thing does happens. 

I didn't know you got hurt in the process. I hope you are OK. Good thing you had protection gear on. 

Again cummunity comfort pad on the shoulder to you. 

 

Thanks 

Unfortunately, I will not be able to take this wheel on any rides that contain longish steep hills. Too delicate. I'll always have to be monitoring the current alarm for sure.

Let's hope Gotway moves to a better control board for the 84-volt version.

[melodrama]

As a quite experienced power electronic designer I would say that Nikola's power MOSFET cooling design is very poor; fan is connected into other side of the heatsink where fresh air/airflow is really needed ! Fan should be located/positioned so that cool air/airflow is guided under the power MOSFET heatsink..  So it's not a big surprice that MOSFET's overheated too much.

[Seba]

7 minutes ago, melodrama said:

As a quite experienced power electronic designer I would say that Nikola's power MOSFET cooling design is very poor; fan is connected into other side of the heatsink where fresh air/airflow is really needed ! Fan should be located/positioned

You're right. It also seems that there is no temperature probe near the radiator or MOSFETs. This is a major design flaw, as PCB temperature where temp probe is apparently located in Nikola, being additionally cooled by the fan located on wrong side, doesn't reflect MOSFET tab or package temperature. This is why @Marty Backe got misleading, very low temperature readings. For example, in KS-18L/XL there is a temperature probe located close to the heatsink and MOSFET that is farthest from the cooling fan. This is why L/XL may be perceived as a "hot running wheel". Nikola wasn't "cold running wheel", it was just a "very hot running wheel with low temp readings".

From what I see, PCB design may cause MOSFET source (S) leg to overheat, as there is highest thermal resistance due to small copper pad directly under the MOSFET leg. I don't see the other side, but i assume it's similar so drain (D) legs are also connected using small pad located on an edge of larger copper plane.

Another design flaw is to use two TO-220 MOSFET instead of one TO-247. Doubled small package design is prone to many failures. MOSFET leg fracture during bending or PCB assembly will increase leg resistance. Improper soldering or improper PCB design may cause current imbalance, so it wont be divided 50/50 between both transistors. And if one MOSFET will fail, but won't short, second MOSFET failure is a matter of short time, as it takes over full current instead of part of it. As we can see in the photo, it was the case. Failed both MOSFETs on one side of a motor phase half-bridge circuit. Double MOSFET design may work well only if is properly oversided, so each single MOSFET is able to carry maximum current expected to flow in the motor phase circuit. But as failing MOSFET will likely create a short circuit, secondary MOSFET won't be beneficial.

[eddiemoy]

1 hour ago, Seba said:

You're right. It also seems that there is no temperature probe near the radiator or MOSFETs. This is a major design flaw, as PCB temperature where temp probe is apparently located in Nikola, being additionally cooled by the fan located on wrong side, doesn't reflect MOSFET tab or package temperature. This is why @Marty Backe got misleading, very low temperature readings. For example, in KS-18L/XL there is a temperature probe located close to the heatsink and MOSFET that is farthest from the cooling fan. This is why L/XL may be perceived as a "hot running wheel". Nikola wasn't "cold running wheel", it was just a "very hot running wheel with low temp readings".

From what I see, PCB design may cause MOSFET source (S) leg to overheat, as there is highest thermal resistance due to small copper pad directly under the MOSFET leg. I don't see the other side, but i assume it's similar so drain (D) legs are also connected using small pad located on an edge of larger copper plane.

Another design flaw is to use two TO-220 MOSFET instead of one TO-247. Doubled small package design is prone to many failures. MOSFET leg fracture during bending or PCB assembly will increase leg resistance. Improper soldering or improper PCB design may cause current imbalance, so it wont be divided 50/50 between both transistors. And if one MOSFET will fail, but won't short, second MOSFET failure is a matter of short time, as it takes over full current instead of part of it. As we can see in the photo, it was the case. Failed both MOSFETs on one side of a motor phase half-bridge circuit. Double MOSFET design may work well only if is properly oversided, so each single MOSFET is able to carry maximum current expected to flow in the motor phase circuit. But as failing MOSFET will likely create a short circuit, secondary MOSFET won't be beneficial.

I think the bigger question is if GW uses the 247’s, are they just moving the problem further down, like to the motor or motor wires.  As you’ve seen from EUC Guy pushing his wheel to the limits, 247’s held up well, but the motorwire insulation melted resulting in a short.  When motor gets too hot, it also affects the magnets, they demagnetize or the glue becomes undone,,.

[stephen]

7 minutes ago, eddiemoy said:

I think the bigger question is if GW uses the 247’s, are they just moving the problem further down, like to the motor or motor wires.  As you’ve seen from EUC Guy pushing his wheel to the limits, 247’s held up well, but the motorwire insulation melted resulting in a short.  When motor gets too hot, it also affects the magnets, they demagnetize or the glue becomes undone,,.

I need to inspect my motor wires now and again because i live in hilly area and sometimes do big hills where i get 90amp beeps/ vibrations often , I've had it nearly a year now 

[Unventor]

7 minutes ago, eddiemoy said:

I think the bigger question is if GW uses the 247’s, are they just moving the problem further down, like to the motor or motor wires.  As you’ve seen from EUC Guy pushing his wheel to the limits, 247’s held up well, but the motorwire insulation melted resulting in a short.  When motor gets too hot, it also affects the magnets, they demagnetize or the glue becomes undone,,.

No matter what you do there is bound to be something to give it is pushed over its limit. I don't think you see how to push a huge car with you wheel in the instruction manual...

 

[eddiemoy]

8 minutes ago, Unventor said:

No matter what you do there is bound to be something to give it is pushed over its limit. I don't think you see how to push a huge car with you wheel in the instruction manual...

 

That is why there is a fuse in the KS wheels.   Even if the firmware screws up, the hardwired fuse would blow and then everything is saved.  Just need to replace the fuse, not the board or motor. Fuse blowing in the KS wheels are rare, but it does happen.  It still results in a face plant. No way around that, but just replace the fuse and if you are properly protected, can ride again.  No need to wait for replacement components to ship. BTW, with everywheel, you get spark fuses.  

[eddiemoy]

14 minutes ago, stephen said:

I need to inspect my motor wires now and again because i live in hilly area and sometimes do big hills where i get 90amp beeps/ vibrations often , I've had it nearly a year now 

When I tested the MSX and put temp probes on the motor wires, temperature wasn’t bad.  But I wasn’t stressing it for a long periods of time.  I just did that so I would feel comfortable with my riding style.  Since the mosfets won’t blow, the other components will.  The MSX seem to be designed to take a lot of stress and no everyday kind of riding will result in melting motor wires.  But doesn’t hurt to check for a piece of mind.

[Rehab1]

13 hours ago, Marty Backe said:

I just returned from my mountain stress testing of the Nikola and it utterly failed (nothing like the smell of burnt MOSFETs in the morning).

So I owe all you guys an apology for defending Gotway's decision to revert back to the smaller MOSFETs. What can I say?

 

10 hours ago, Sketch said:

 I’m kind of surprised they went with the smaller MOSFETs on the Nikola.  Seems like a great wheel otherwise and an easy modification for Gotway to change.  Good to know the plus version has the larger MOSFETs. 

For those who might not understand how the heck a MOSFET works.

 

[diegoxxx]

https://www.facebook.com/groups/1098582130337563/permalink/1111606179035158/  Look at the information posted recently by  Authorized GotWay Service EUCservice https://www.facebook.com/groups/1098582130337563/permalink/1111606179035158/

[EUC Custom Power-Pads]

4 hours ago, Marty Backe said:

But Gotway really screwed up by going cheap on the MOSFETs. It's a mystery 

 

4 hours ago, Marty Backe said:

I have learned my lesson that I will not buy another Gotway wheel that doesn't use the large MOSFETs!

I'm really sorry for you, but thanks for the test.
So the Nikola 84V is out of the 16X or Nikola 84V purchase decision for me!

Now the question is back again....16X or Nikola Plus 100V with bigger 247's mosfets, even if this means maybe less torque for higher speed.

But what does it look like in the Plus version? I can't imagine that it looks any different there, except for the other mosfets?
I suppose we're expecting the same design flaw there as @Seba wrote and I see it the same way, don't you?

3 hours ago, Seba said:

It also seems that there is no temperature probe near the radiator or MOSFETs. This is a major design flaw, as PCB temperature where temp probe is apparently located in Nikola, being additionally cooled by the fan located on wrong side, doesn't reflect MOSFET tab or package temperature. This is why @Marty Backe got misleading, very low temperature readings. For example, in KS-18L/XL there is a temperature probe located close to the heatsink and MOSFET that is farthest from the cooling fan. This is why L/XL may be perceived as a "hot running wheel". Nikola wasn't "cold running wheel", it was just a "very hot running wheel with low temp readings".

 

I can only agree with that.

1 hour ago, eddiemoy said:

That is why there is a fuse in the KS wheels.   Even if the firmware screws up, the hardwired fuse would blow and then everything is saved.  Just need to replace the fuse, not the board or motor. Fuse blowing in the KS wheels are rare, but it does happen.  It still results in a face plant. No way around that, but just replace the fuse and if you are properly protected, can ride again.  No need to wait for replacement components to ship. BTW, with everywheel, you get spark fuses.

Unbelievable, I would have assumed that a fuse would be standard.

KS 
Gotway