My Sherman Trials, Tribulations, and Triumphs

[travsformation]

4 hours ago, Patton250 said:

I can’t figure you guys out. Someone takes a product, the first of its kind mind you, kind of a prototype, the first run and does to it what it was never designed to do and it fails and you guys are devastated? Gezzzz

 

I had to add to this. You know very well veteran and Jason will get this fixed so let me ask you this. So let’s say they get it fixed and Marty takes the thing up overheat hill and it passes. But then he goes up overheat Hill for a second time immediately after the first trip and it fries that time. Are you guys going to be devastated then as well? My point is 99.999% of you guys will never put any wheel never mind this wheel under that kind of stress condition. Rant over.

Devastated is "a bit" of an overstatement. I was just hoping it would meet my needs so I could order it 

See my other post:

6 hours ago, travsformation said:

But with Marty's testing and Jason's determination to get Veteran to make this wheel "Overheat-hill Certified", we might just get what we're after, even if it's a little further down the road...

It isn't a big deal, I can wait for a revision or another new wheel. This is Veteran's first wheel, I'm sure there's plenty more to be excited about in the future even if this one turns out not to be for me

3 hours ago, Planemo said:

Well said. However, to counteract that, I think people were under the assumption that the highest power wheel ever made would make it up overheat hill, given other far lesser spec wheels have.

Further, the fact that the Sherman is a bit heavier shouldn't make any difference (as has been suggested) IMO, given an extra 10kg or so isn't a huge amount relative to the weight differences between humans.

The Sherman has a monster spec in all departments, so I would have expected it to make it up the hill. The board should be a factor of X more robust than other wheels in direct relation to the power it can produce (and maintain), taking into account the torque requirements for the diameter of the wheel. I totally accept that other wheels have also failed overheat hill but they were never touted as the most powerful wheel on the planet either.

I am sure that Veteran will sort it, they will want to, in order to maintain the stature

+1 ! 

Edited July 14, 2020 by travsformation

[Marty Backe]

2 hours ago, ShanesPlanet said:

Oh man, fail on overheat hill? Hate to see this for sure! Hey Marty, is that axle socket similar in size to ANY  O2 sensor sockets? Cutting the ribs was a good idea, but vibration is a bitch. Even with almost no pressure, I could see how that wire would become worn again.  Too bad they didnt make a simple tab in the case to hold those down properly.  Damn damn! It really is TOO bad they dont do user firmware ability. I really do hope Jason forces them to handle this, BEFORE mine ships.. Too funny, simple ride with Marty breaks what NYC couldnt. How's that for the truth about how a typical rider can give these wheels a workout. I foresee many many 3d printed frame sliders installed. Why not put that big ole boy Jeff on the hill for a REAL test. I'd think his body type is more fitting for the average usa rider. Jeff seems awesome, I'd love to share a hearty beer with that fellow! Thank you BOTH for doing the work that these companies SHOULD be doing themselves, BEFORE release. Maybe bring a fire extinguisher in the future... lotsa dry grass up there! I think you are overdoing it with that helmet, I know you say atgatt, but...

It's a 24mm socket. Over the years, it comes down to cutting your own slot. O2 sockets aren't available in 24mm.

Assuming that I passed the test, Jeff was there to do the 230-pound version of the test.

This helmet is super light weight and cool (lots of holes). Maybe the chinbar seems overkill, but the helmet is very comfortable to wear all the time, so I do.

[Gadgetfann]

1 hour ago, Marty Backe said:

It's a 24mm socket. Over the years, it comes down to cutting your own slot. O2 sockets aren't available in 24mm.

Assuming that I passed the test, Jeff was there to do the 230-pound version of the test.

This helmet is super light weight and cool (lots of holes). Maybe the chinbar seems overkill, but the helmet is very comfortable to wear all the time, so I do.

This socket actually works great https://www.ecstuning.com/b-schwaben-parts/strut-nut-socket-24mm/003046sch01a-05/

[Marty Backe]

9 minutes ago, Gadgetfann said:

This socket actually works great https://www.ecstuning.com/b-schwaben-parts/strut-nut-socket-24mm/003046sch01a-05/

Not really.

The point of using a slotted socket like I showed in the video is that you don't have to disconnect any cables to tighten the nut. Using the socket that you are referencing would require substantially more more in disconnecting all the wires from the control board.

[Girth Brooks]

4 hours ago, ShanesPlanet said:

is that axle socket similar in size to ANY  O2 sensor sockets? 

https://www.toolsource.com/misc-c-318/24mm-nox-soot-socket-wrench-p-240459.html

 

This should work if needing a 24mm socket for axle nuts instead of making one. NOX sensors for diesel vehicles are near identical to an O2 sensor.

[AtlasP]

16 hours ago, Marty Backe said:

Roll Cage Secondary Uses

I recently discovered that the roll cage makes for an excellent attachment point for lifting with a crane. Flexibility is king 

...

Photo (and crane) courtesy of @Jrkline "Wheel Whisperer" 

Considering the weight and seeing you strain to lift it into your truck maybe ewheels should start selling cranes as an accessory. ;-)

[Marty Backe]

2 hours ago, Girth Brooks said:

https://www.toolsource.com/misc-c-318/24mm-nox-soot-socket-wrench-p-240459.html

 

This should work if needing a 24mm socket for axle nuts instead of making one. NOX sensors for diesel vehicles are near identical to an O2 sensor.

Have you ever tried tightening an axle nut on an EUC? The socket must drop down below the surface of the shell to access the nut. The product that you are referencing would only work if the nut were above the surface. EUC nuts are recessed into the shell.

[.....]

I have an 02 set, but nuthing in a 24mm. Oh well, it aint like I havent scraped a cheap as crap thin walled deep well before. Recycled paperclips make good tools for that. Cheap enough. Much cheaper than buying any specialized O2 socket anyhow. Im assuming it would need be thin walled, as 24mm is usually a pretty beefy size otherwise.

[Marty Backe]

30 minutes ago, ShanesPlanet said:

I have an 02 set, but nuthing in a 24mm. Oh well, it aint like I havent scraped a cheap as crap thin walled deep well before. Recycled paperclips make good tools for that. Cheap enough. Much cheaper than buying any specialized O2 socket anyhow. Im assuming it would need be thin walled, as 24mm is usually a pretty beefy size otherwise.

On the Sherman there is enough access for use of a beefy socket. The socket shown in the video is an Impact Socket, so it's extra thick.

[Patton250]

16 hours ago, travsformation said:

Devastated is "a bit" of an overstatement. I was just hoping it would meet my needs so I could order it 

See my other post:

It isn't a big deal, I can wait for a revision or another new wheel. This is Veteran's first wheel, I'm sure there's plenty more to be excited about in the future even if this one turns out not to be for me

+1 ! 

So you’re saying because the demo unit finally died after is had been put through its paces by every EUC testing enthusiast in the country finally to be topped off by Marty bringing it into one of the most stressful tests to be done on it you’ve decided it doesn’t meet your needs? Just out of curiosity what is veteran sends a beefed  up board to Marty is able to take it up overheat hill without a problem but then because of the battery storage Marty decides to do a second and third test in a row on the same day and on the third try the new board fries. Disappointed and not want this wheel? I guess my point of questioning is there are people that do things to these wheels that no one else is ever going to do or even close and the wheels they’re doing it to her not even the final version. You could do this with any product that comes out. Perhaps these companies should offer three different versions of each product. We all know that you can use different quality parts with any product you manufacture. Perhaps it’s time EUC companies anticipate that some users of their products is going to bring it to the extreme every day. I recently purchased a mountain bike. I’m new to it. I’m sure some of you know there are mountain bikes that cost $800 and then there are mountain bikes that cost $12,000. Obviously there’s quite a big difference in quality of the frame and of parts between these two prices. I’m thinking some of you guys want a $12,000 EUC. I’m sure someone can make one. 

[FreeRide]

8 minutes ago, Patton250 said:

So you’re saying ....

For starters Veteran has a big fuse in line with the battery, if that does now blow before the wheel burns up then there is an issue they should fix or stop wasting money on fuses

Relax, there are many important things to rant on if you must. 

[Chriull]

17 hours ago, Alj said:

The current will depend on the energy stored (i e capacitance), and film capacitors will have much smaller value meaning they will take much less current.

Agree. A 160V 20uF film capacitor has about the size of such an 160V 560uF aluminium electrolytic capacitor. They do not take any of continous burden (heat dissipation) from the electrolytic capacitors, but by their much lower ESR the deliver (and take from the el. cap.) higher current peaks while the switching process. To whatever cons and pros this would lead...

17 hours ago, Alj said:

...

Also why would it be such a big current there flowing through the caps? If you just short charged cap it will not explode like this.

That's just a short one time burden.

I tried some ltspice simulation (https://forum.electricunicycle.org/topic/18815-capacitors-on-motherboardsany-tech-gurus/?do=findComment&comment=315189).

I updated this inbetween (166uH for battery wires, 2 x 560uF caps instead of one 4700uF, 0.3Ohm as ESR) and the rms current flowing through the caps gets much lower, but still, especially in low speed (low motor generated voltage) and high burden (duty cycle) situations the current gets way to high.

There should be easily some 2 digit power dissipation numbers be reached - enough to cook the caps.

 

17 hours ago, Alj said:

Most likely the problem is overheating  due to suboptimal PCB design and poor heat dissipation.

Plus this!

KS seems to have avoided this prob by putting the caps on the fan (and of course having lower power specs).

A solution could be to split the 2 560uF caps to 10 110uF caps? If they can be placed and connected without too much parasitic inductance this should drasticly improve the power dissipation problem. And if the ESR is not growing too much for the smaller values...

...and they can find the cooler place for them and/or put them into the fan airflow.

[Patton250]

8 minutes ago, FreeRide said:

For starters Veteran has a big fuse in line with the battery, if that does now blow before the wheel burns up then there is an issue they should fix or stop wasting money on fuses

Relax, there are many important things to rant on if you must. 

Brother that’s just a tad bit condescending. Just a little. Was it intentional? :-)

[Planemo]

46 minutes ago, Chriull said:

A solution could be to split the 2 560uF caps to 10 110uF caps? If they can be placed and connected without too much parasitic inductance this should drasticly improve the power dissipation problem. And if the ESR is not growing too much for the smaller values...

A separate 'cap board' would be an excellent idea. It could then be swopped out easily as part of periodic maintenance so that we never reach the point where they pop either.

If they continue with the route of 2 big caps then they deffo need to combat heat management. Maybe start heat-sinking them at the very least. I think we are definately reaching a point where the caps are the weakest link in the chain, especially over time.

[Chriull]

39 minutes ago, Planemo said:

A separate 'cap board' would be an excellent idea. It could then be swopped out easily as part of periodic maintenance so that we never reach the point where they pop either.

Good point!

Just the this construction has to bear safely currents up to max peak motor currents! So up to 1xx A.

Any trace/wiring inductance has to be kept liw, or they render more or less useless

Any failure in this board means instant faceplant!

39 minutes ago, Planemo said:

If they continue with the route of 2 big caps then they deffo need to combat heat management. Maybe start heat-sinking them at the very least. I think we are definately reaching a point where the caps are the weakest link in the chain, especially over time.

I fear this, too. Hopefully they just get really "cooking" in this high burden situations (almost stalling up an incline).

Just found a paper (https://www.google.com/url?sa=t&source=web&rct=j&url=https://pdfs.semanticscholar.org/d722/34c8f1509d583425b23b10e292d5af6596be.pdf&ved=2ahUKEwjOysufq8_qAhWFDOwKHecJBp4QFjACegQIAhAB&usg=AOvVaw0axfRmeyYwW_96d6Crc3-p) showing already up to ~ 5A capacitor current in a system with an 100W 30V motor with cap values from 820uF to 3280uF.

If this can be transformed (1) to 2000W 100V by 5A / 100W * 2000W * 30V / 100V = 30A we are in the area where "our" caps explode...

30A would by 15A per capacitor with an ESR of roughly 0.3 Ohm leading to 67W!!

Dividing this by 10 capacitors (the smaller ones i found unfortionately have around 1 Ohm ESR) would still mean 9W per capacitor.

That seem to be quite peak numbers - with half the "burden requirements" one would come to 15A total (16W per big cap, 2W per small cap). A quarter = 7.5A would lead to 4W per big cap and 0.5W per small cap.

0.5W is here the only comforting number, maybe 2 (to 4W) for short time. Everything else needs active cooling...

That's all just wild guessing from roughly scanning one paper, but hopefully shows some magnitude.

(1) sometimes it can be just plain simple - sometimes such easy solutions don't work out at all. Maybe i'll know more after reading this paper in detail...

[erk1024]

On 7/14/2020 at 2:50 PM, Marty Backe said:

Assuming that I passed the test, Jeff was there to do the 230-pound version of the test.

Jeff actually DID the 230 pound test ... just with the MSP! He was chugging along behind you just fine.

Top marks for the MSP!  

Edited July 15, 2020 by erk1024

[Marty Backe]

7 minutes ago, erk1024 said:

Jeff actually DID the 230 pound test ... just with the MSP! He was chugging along behind you just fine.

Top marks for the MSP!  

Agreed

[Marty Backe]

Failure Postmortem

Today I disassembled the Sherman so that I could remove the control board and take a closer look at the damage.

Sadly, total disassembly is required to pull the control board. I think Veteran assumed that they would have no control board failures. Considering how easy it is to do all other work on the Sherman, the work required to replace the control board is disappointing.

All the MOSFET's are destroyed, but it looks like the massive amount of current coming in from the batteries literally vaporized that section of the control board. It could be that the MOSFET destruction was secondary to the melt-down that occurred at the battery connections.

All the wiring external to the control board looks great and I see no evidence of heat stress 

 

[bryon01]

On 7/13/2020 at 9:26 AM, Chriull said:

Yes. As stated before only cc stage time gets proportionaly shorter. By the longer cv stage duration it still gets shorter, but as for the first example (8A vs 10A) just marginal.

I wanted to give proof that while correct that CV stage gets longer as you charge at higher currents, it is completely incorrect to say this concept that this time extension somehow offsets the CC stage time of the charge.  Not correct at all.  As I have the exact cells, and the gear to record all this, I did it.  Note that unless charging the cells way outside their spec, then the area under the curve for any charge profile, will be equal.  The energy in is conserved, while voltage and amperage change.  Here is the data.  Keep in mind that my pack was 4S3P, but I listed the cell amperages in the figure.  I picked 3.15 cell voltage as the cutoff based on Marty's video.  For a Veteran in a 24S10P config, the 4 tests equate to:

1) 15A Discharge, 5A Charge: 6.2055 hours

2) 15A Discharge, 8A Charge: 4.0908 hours

3) 15A Discharge, 10A Charge: 3.2219 hours

4) 2kW Discharge, 16.5A Charge: 2.0947 hours (The rated charge current for the cell.  You wouldn't hurt the cells charging the Veteran at this rate, but could break the control board BMS, which could then hurt the cells...)

Note the longer tail for the lower current charges during the CV phase of charging is due to the cutoff criteria my charger uses, which was charge rate amps/20.  For small rates, the cutoff is also lower amps allowing it to creep in a few more mAh into the cell.  Completely not worth the wait as you are adding almost no energy in so you would instead want to stop the charger at some fixed amount of amps.  Note that 4 hours is 14,400 seconds, so if you have a TRUE 8A charger, your Veteran would charge in almost exactly 4 hours from dead empty.  If the Veteran chargers are not hitting this time, it is because they likely aren't really hitting the amperage they say they are.  Liion cells are so efficient it isn't due to energy loss in the cell and shouldn't be due to energy loss from the balancer unless you have a bad cell in the pack.

I don't know how to share the figure so here it is:

https://drive.google.com/file/d/1AFxL9yxw7B-UK66v-aj_96HnYZq-DGTw/view?usp=sharing

And the raw data if anyone wants it:

https://drive.google.com/file/d/1TqPUpp1FVeYyKEvDc_x38SGjohPJEFhx/view?usp=sharing

 

Edited July 16, 2020 by bryon01
said voltage, meant current

[Nick McCutcheon]

54 minutes ago, Marty Backe said:

Today I disassembled the Sherman so that I could remove the control board and take a closer look at the damage.

Sadly, total disassembly is required to pull the control board. I think Veteran assumed that they would have no control board failures. Considering how easy it is to do all other work on the Sherman, the work required to replace the control board is disappointing.

This is definitely disappointing to hear  Hopefully the production version from ewheels will have a slightly different shell design to allow for easier removal of the board, I can't imagine it would require any radical redesigns. I wonder if @Jason McNeil would be able to suggest this change while Veteran is in the process of revising the wheel to survive overheat hill. 

Edited July 16, 2020 by Nick McCutcheon

[Marty Backe]

1 hour ago, Nick McCutcheon said:

This is definitely disappointing to hear  Hopefully the production version from ewheels will have a slightly different shell design to allow for easier removal of the board, I can't imagine it would require any radical redesigns. I wonder if @Jason McNeil would be able to suggest this change while Veteran is in the process of revising the wheel to survive overheat hill. 

Shell designs are extremely expensive to have manufactured. I'm afraid this is it :-(

[Marty Backe]

Here are some additional pictures.

After further consideration, I think there are issues with their hardware design. The amount of board destruction that occurred is something I don't recall ever seeing on any Gotway wheels. The MOSFET's self-destructed long before the board could delaminate and melt down like this.

Am I wrong in concluding that they have not sized the components (including the circuit board traces) to support the amount of current that they are allowing to flow through the board?

I would be concerned that even a momentary (few seconds) high load could cause severe weakening of the board.

Basically, if the board fails it should be some components (MOSFETs) not the circuit board itself. Right?

What do any of you hardware guys think?

 

 

 

 

 

Edited July 16, 2020 by Marty Backe

[Mr. Thompson]

First I would like to make a comment on the temperatures mentioned during the ride. Where are they being measured and how? Almost EVERYTHING I HAVE READ ON THIS FORUM REGARDING TEMPERATURE MEASUREMENT IS B.S., with perhaps the exception of "thermistor's aren't clairvoyant", a brilliant phrase which eloquently expresses a the concept of why a thermistor only measures the temperature of itself.

If motor wires are getting hot they not of sufficient diameter to carry the current. Having to remove the motor wire's insulation due to heat is an admission of a known problem.

I have mil-spec electronic equipment with electrolytic caps that can handle the heat of a RTTY shack in Vietnam or in the Sahara Desert. The cap failed from heat?!!! And all the mosfets toasted too, imagine that? This is a CHINESEUM CLASSIC, under spec components., maximize profit, who cares who gets a life changing faceplant, might as well be a virus! The Gotway Way!? Don't worry I'm sure they'll patch It up for the production version just enough to where it will run just fine for a while.

I spent the last 12 years of my professional life educating folks in the industry on the properties of thermal transfer. The amount of ignorance and outright incompetence just astounds me. At this point I'll confine my adventures to my Mten3. Faceplants are at lower speeds. If they get the Veteran sorted out I would love to own one, it's at the top of my list, of course with a minimun bondline thermal interface material between the mosfets and the cheesy provided heatsink and some serious ventilation at the top of the chassis (duct tape option for rain:). I'll add that I have all already had a conversation with Jason about this as this I perceive this is a common problem with all of the PEVs not just the Veteran.

Please don't let my comments dicourage you from buying this wheel, as I believe it is the most exciting thing short of the little Mitten. It's just that we're dealing with an industry of manufacturers who don't give a damn about our safety, all they care is about their short term profits. On the other hand we have enthusiastic dealers who DO UNDERSTAND the whole picture. I encourage you to patronize the dealers who care about the riders and are passionate about our sport. Encourage them to pour the pressure on the manufacturers to get rid of the CHINESEUM.

 

Edited July 16, 2020 by Mr. Thompson

[Chriull]

37 minutes ago, Marty Backe said:

Here are some additional pictures.

You have some additional pictures of the dead capacitor? As it seemed from the video the cap of the capacitor was perfectly intact? Just some "plastic wrapping" on the side melted? The legs are totaly gone? Is there some leakage at the "bottom" (were the legs were attached)?

Seems most the glunk is to the right of the capacitor - on the heatsink and on the PCB?

So if the "black glunk" on the heatsink is the electrolyte of the capacitor, it must have blown through the newly burned hole of the PCB? Or was reflected from some side panel and went under the PCB?

37 minutes ago, Marty Backe said:

After further consideration, I think there are issues with their hardware design. The amount of board destruction that occurred is something I don't recall ever seeing on any Gotway wheels. The MOSFET's self-destructed long before the board could delaminate and melt down like this.

Am I wrong in concluding that they have not sized the components (including the circuit board traces) to support the amount of current that they are allowing to flow through the board?

I would be concerned that even a momentary (few seconds) high load could cause severe weakening of the board.

Was imho a bit more than a few seconds on heat hill. More like 10 seconds+ at slow speed at the high incline part before it stalled?

37 minutes ago, Marty Backe said:

Basically, if the board fails it should be some components (MOSFETs) not the circuit board itself. Right?

What do any of you hardware guys think?

Obviously the most heat developed at the solder joints of the two battery cables pairs as the PCB vaporized there?! And the capacitor was just a victim of the heat that developed at this point?

Would be interesting to know the motor current at this last part of the overheat hill drive - could easily be way over anything such a PCB construction can ever handle for the duration of some seconds...

So not really any possibility to solve this without drastic design changes of the hardware - imho they'd have to implement some very strict and "intelligent" current limiting for low speeds (avoid the huge battery current "spikes"). So such low speeds at driving up overheat hill will then lead to a "firmware shutdown" - one would need pads then to go up with some higher speeds. That should (?could?) give lower battery currents and especially less duration for the high burdens?

If one compares to an MSP (or similar) with a 21700 24s4p configuration these wheels have "internal" battery current limiting, as 4p cannot deliver as much current as the 10p cells of the sherman...

Will be interesting which changes they implement in the replacement boards. 

[Patton250]

1 hour ago, Mr. Thompson said:

First I would like to make a comment on the temperatures mentioned during the ride. Where are they being measured and how? Almost EVERYTHING I HAVE READ ON THIS FORUM REGARDING TEMPERATURE MEASUREMENT IS B.S., with perhaps the exception of "thermistor's aren't clairvoyant", a brilliant phrase which eloquently expresses a the concept of why a thermistor only measures the temperature of itself.

If motor wires are getting hot they not of sufficient diameter to carry the current. Having to remove the motor wire's insulation due to heat is an admission of a known problem.

I have mil-spec electronic equipment with electrolytic caps that can handle the heat of a RTTY shack in Vietnam or in the Sahara Desert. The cap failed from heat?!!! And all the mosfets toasted too, imagine that? This is a CHINESEUM CLASSIC, under spec components., maximize profit, who cares who gets a life changing faceplant, might as well be a virus! The Gotway Way!? Don't worry I'm sure they'll patch It up for the production version just enough to where it will run just fine for a while.

I spent the last 12 years of my professional life educating folks in the industry on the properties of thermal transfer. The amount of ignorance and outright incompetence just astounds me. At this point I'll confine my adventures to my Mten3. Faceplants are at lower speeds. If they get the Veteran sorted out I would love to own one, it's at the top of my list, of course with a minimun bondline thermal interface material between the mosfets and the cheesy provided heatsink and some serious ventilation at the top of the chassis (duct tape option for rain:). I'll add that I have all already had a conversation with Jason about this as this I perceive this is a common problem with all of the PEVs not just the Veteran.

Please don't let my comments dicourage you from buying this wheel, as I believe it is the most exciting thing short of the little Mitten. It's just that we're dealing with an industry of manufacturers who don't give a damn about our safety, all they care is about their short term profits. On the other hand we have enthusiastic dealers who DO UNDERSTAND the whole picture. I encourage you to patronize the dealers who care about the riders and are passionate about our sport. Encourage them to pour the pressure on the manufacturers to get rid of the CHINESEUM.

 

I’m beginning to think that some of you people take these things a little bit too seriously. When I read post like yours I forget I’m on an electric unicycle forum I feel as if I’ve somehow found myself on an aircraft manufacturing forum.