Begode Master 134V 2400WH Suspension

[meepmeepmayer]

Here's a picture. Are these smaller than usual? Looks like TO-220?

[Chriull]

12 minutes ago, meepmeepmayer said:

Looks like TO-220?

More like TO 263-7

https://www.infineon.com/cms/en/product/packages/PG-TO263/PG-TO263-7-3/

Edited April 14, 2022 by Chriull

[fryman]

Doesn't each Mosfet have a failure rate?  If so, using more Mosfets would increase the failure rate of the board? 

[meepmeepmayer]

31 minutes ago, Chriull said:

More like TO 263-7

These are even smaller than TO-220.

Also, are the mosfets on the top of the PCB? Shouldn't they be on the heatsink side?

11 minutes ago, fryman said:

Doesn't each Mosfet have a failure rate?  If so, using more Mosfets would increase the failure rate of the board? 

In principle, yes.

-

No idea how to judge this. Maybe the board runs exceptionally cool? Or maybe they are trying to save on components?

Waiting for overheat hill!

[Chriull]

On 4/14/2022 at 10:28 PM, fryman said:

Doesn't each Mosfet have a failure rate?  If so, using more Mosfets would increase the failure rate of the board? 

There is no such (relevant) thing as statistical failure rates like for harddisks, cpu, ram, etc faults.

Mosfets degrade or annihilate depending on burdening.

As cooling needs and currents for the peak burdens are at the limit paralleling is the "only" option. Besides increasing the voltage (which has other disadvantages)

On 4/14/2022 at 10:43 PM, meepmeepmayer said:

These are even smaller than TO-220.

At a first glance a https://www.infineon.com/cms/en/product/power/mosfet/n-channel/ipb044n15n5/ has very nice parameters - but did not really look into details by now, nor looked at other such mosfets or d2pak-7 package limits.

Edit: this is just a randomly picked mosfet - no idea which mosfet is used for the master

On 4/14/2022 at 10:43 PM, meepmeepmayer said:

Also, are the mosfets on the top of the PCB? Shouldn't they be on the heatsink side?

They are SMD mosfets - so they are soldered automaticly on the surface. No more bad solder joints by manual soldering!

Cooling happens by the "backplane" of the mosfets beeing solidly soldered on the pcb and "heat transfer vias" connect to the other side were on a big solder space a heatsink can be mounted.

Or metal is in the center of the pcb helping for the heat transfer.

Anyway this mosfets are the future as they can be automaticly mounted - so there should exist enough great heat sink mounting techniques.

Edited April 19, 2022 by Chriull

[Tawpie]

Heat management will be "the" challenge for the D2Pak-7 package design. An advantage that PCB attached parts have is they make their thermal attachment to the heat sink using solder, so as long as the solder quality is monitored and enforced the thermal path to the PCB should be very consistent. More fun happens getting the heat out of the PCB and that part isn't clear (to me) from the video. Another wrinkle is by dumping the MOSFET heat into the board, they'll be warming up all the other parts but they would know that and will have taken it into consideration.

I'm sure they did their calculations, but the phase wires look mighty thin. Then again, this has the look of a factory video so we don't really know if what we're seeing here is what is now coming off the boats.

Yep, overheat hill will step 1. A long run at speed in Dubai will also be an interesting experiment.

Edited April 14, 2022 by Tawpie

[alcatraz]

There's no heatsink going to those components on top. Does that mean the actual mosfets are on the bottom?

Also is it fanless? What about those desert joy rides?

Looks good in general. The tolerances on all those suspension components is an interest to me. I bet that it works fine at 0 mileage, but if there's a bit of play there, it can potentially develop into sounding like a drawer of kitchen utensils on wheels, at a few thousand km. It would be cool if the parts are mostly serviceable, and that they tell us which parts to grease and how often. 

I'm curious how the suspension actually works. Are those long vertical cylinders pressurized? Are they prone to leaking? Where can they fail? 

[Tawpie]

31 minutes ago, alcatraz said:

Does that mean the actual mosfets are on the bottom?

Those are the MOSFETs (but we're only seeing 12 of them on top—weren't they rumored to be using 24?)... the package is called D2-Pak 7pin. The heat goes into the board first, then somewhere else.

Edited April 14, 2022 by Tawpie

[Tawpie]

29 minutes ago, alcatraz said:

Are those long vertical cylinders pressurized?

Probably not, from what I saw in the rest of the teardown and @GoGeorgeGo's EX20S

Edited April 14, 2022 by Tawpie

[alcatraz]

The contents in this video here so far doesn't show signs of high performance to me. 

Heatsinks, fans, ventilation solutions etc.

[redfoxdude]

1 hour ago, alcatraz said:

There's no heatsink going to those components on top. Does that mean the actual mosfets are on the bottom?

Also is it fanless? What about those desert joy rides?

As others have stated, this type of MOSFET has a large thermal connection underneath itself where it is soldered to the copper on the circuit board, and then you would pipe that heat through the bottom of the board to the massive heatsink. Hopefully there are good thermal pads between the board and that heatsink, but if not, that's probably going to be a really good potential upgrade. Assuming good thermal conductivity out to the heatsink and sufficient surface, this may be a very cool running wheel, but we'll have to see!


Other than that, there appears to be this cute little internal fan that would circulate air around the internal chamber:

1 hour ago, alcatraz said:

I'm curious how the suspension actually works. Are those long vertical cylinders pressurized? Are they prone to leaking? Where can they fail? 

I'm quite sure those are just the stanchions that the sliders go on. You can see how how the bicycle shock compresses at 3:44 in the video https://youtu.be/QphMEM5M9NQ?t=224
One end is in the red linkage, and the other is on the pivot on the sliders. Curious what the spring rate ends up like! I do hope the stanchions are strong enough to hold up to some abuse.

 

Edited April 15, 2022 by redfoxdude

[alcatraz]

I'm surprised noone has mentioned this yet.

This will be the first wheel with single cell cell groups.

It means that deteriorating cells can't be masked by neighboring cells. Issues will be detected earlier with this kind of design.

This wheel is going to have a lot of people failsearching later by individually disconnecting packs. Figuring out which one doesn't charge 100% and sending that one in for repair/swap.

And you could quite easily ride with 3/4 packs connected. If one is being serviced/replaced etc.

Edit: See that junction PCB that connects all those packs together. It could technically indicate to the user something like "Error: Pack no 3 is balancing for X hours. If you see this message for more than 3 hours then service!"

Edited April 15, 2022 by alcatraz

[fbhb]

1 hour ago, alcatraz said:

I'm curious how the suspension actually works. Are those long vertical cylinders pressurized? Are they prone to leaking? Where can they fail? 

The suspension works very similar to the King Song S18's linkage system (where Begode/GotWay also copied the Hero/EX20s suspension implementation from) with the obvious difference of Only using one slider tube per side (not pressurized cylinders at all).

My main concern with the suspension with there Only being One slider per side, is the fact that the slider tubes Also look to my eye to be very, very thin walled steel (plus the fact Ecodrift's teardown found them instantly rusting on the Hero). Hero, EX20S and the Master All share a similar suspension implementation and I would hope the sliders have been beefed up from those initially found on the Hero. 

It has to be remembered that e-Rides in the UK had 2 of the very first Hero's delivered to them firstly in HT format, closely followed by a HS version but unfortunately the HS version arrived with these similarly designed "single sided", thin walled sliders bent rendering the suspension stuck solid!

How is it even possible to bend this main structural element unless it is poor quality and thin walled, as I know there is No Way the sliders on the S18 being 3mm thick walled would be able to sustain bending force, especially with there being 4 slider tubes!

As mentioned in my previous post, Chance Hinz may well be one of the first in the US to get his Master and actually put it through it's paces on the extreme trail rides he is famous for, with his ability to compare it's performance and durability directly to the RS19!

He will be able to attest to outright performance, cooling, hill climbing, durability from numerous drops sometimes onto rocks etc. and most importantly for a suspension wheel, whether the implementation works out of the box or like mentioned elsewhere, the stock shock for his use case needs to be binned and replaced!

Edited April 15, 2022 by fbhb

[alcatraz]

They should ship Master's with a dummy pack. So you can mount it in place of the pack being serviced. Then you can mount your powerpads back like normal.

[supercurio]

9 minutes ago, alcatraz said:

I'm surprised noone has mentioned this yet.

A month ago 

9 minutes ago, alcatraz said:

This will be the first wheel with single cell cell groups.

It means that deteriorating cells can't be masked by neighboring cells. Issues will be detected earlier with this kind of design.

This wheel is going to have a lot of people failsearching later by individually disconnecting packs. Figuring out which one doesn't charge 100% and sending that one in for repair/swap.

And you could quite easily ride with 3/4 packs connected. If one is being serviced/replaced etc.

This configuration offers a great potential for a Smart BMS configuration in order to monitor every cell indeed, at last!

However since there's none and that each pack are connected to others in the end, possibly in parallel in practice it also increase the risk of things going wrong if a single cell is damaged - becomes a heater cell. Then it could dissipate the energy from 3x other packs as heat while its host pack might get cells charged above 4.2V.

This is where you really want to see packs isolated from each other so a good pack can't charge a bad pack (it can already be the case with this config), but then preferably a 4x Smart BMS config.

Edited April 15, 2022 by supercurio

[meepmeepmayer]

53 minutes ago, Tawpie said:

(but we're only seeing 12 of them on top—weren't they rumored to be using 24?)

3 rows seem to be hidden below the capacitors. The lower half of the board is essentially a field (graveyard?) of mosfets.

[alcatraz]

17 minutes ago, supercurio said:

A month ago 

This configuration offers a great potential for a Smart BMS configuration in order to, for once monitor every cell indeed!

However since there's none and that each pack are connected to others in the end, possibly in parallel in practice it also increase the risk of things going wrong if a single cell is damaged - becomes a heater cell. Then it could dissipate the energy from 3x other packs as heat while its host pack might get cells charged above 4.2V.

This is where you really want to see packs isolated from each other so a good pack can't charge a bad pack (it can already be the case with this config), but then preferably a 4x Smart BMS config.

Right. Sudden cell internal shorts should be less likely to result in a fire. You have 31 cell groups taking up the (up to) 4.2v excess. What's that, 4.2v/31 =0.135v? Sure it's going to be a lot of sudden current rushing. But fire? The heat will be greater but spread over the whole 32 cell module instead of only one group of 2-4 cells.

Worst situation would be a fully charged pack and one cell goes. Then all other cells will be brought to 4.2 + 0.135 = 4.335v.

Maybe... 

Edited April 15, 2022 by alcatraz

[meepmeepmayer]

I'm worried that these directly exposed battery boxes (on the Master or EX20S or S22) might be mechanically damaged in a crash. It would be a direct impact. Worst case it's going to kill the entire battery pack.

We'll need 3D-printed protectors, I suspect. Or maybe a one-piece combination of powerpad and protector for the Master would work.

[enaon]

30 minutes ago, alcatraz said:

This will be the first wheel with single cell cell groups.

 

the s18 is like that. 

[alcatraz]

Oxidizing sliders sounds bad. Negligent people are going to simply replace them every few years right? Maybe get them coated or something. 

The careful owners are going to grease them up every year etc.

[alcatraz]

8 minutes ago, enaon said:

the s18 is like that. 

Right, sorry.

[redfoxdude]

21 minutes ago, meepmeepmayer said:

I'm worried that these directly exposed battery boxes (on the Master or EX20S or S22) might be mechanically damaged in a crash. It would be a direct impact. Worst case it's going to kill the entire battery pack.

We'll need 3D-printed protectors, I suspect. Or maybe a one-piece combination of powerpad and protector for the Master would work.

I mean, the stock pads are pretty thicc up front! 
But I would likely want custom pads if I got one, and protective bumpers would be def be high on my list. Also I'd want to have something to maintain the lines of the original pad, it looks quite silly to me without the chunky pads.

Edited April 15, 2022 by redfoxdude

[enaon]

43 minutes ago, meepmeepmayer said:

I'm worried that these directly exposed battery boxes (on the Master or EX20S or S22) might be mechanically damaged in a crash. It would be a direct impact. Worst case it's going to kill the entire battery pack.

We'll need 3D-printed protectors, I suspect. Or maybe a one-piece combination of powerpad and protector for the Master would work.

being in a box is no different than being behind an ~2-3mm plastic side cover, like in most gotways, maybe even an improvement I think.

plus, all the "plastic" parts of the master are made out of foam, maybe the best idea this model brings to the game. 

Edited April 15, 2022 by enaon

[EUChristian]

Swore I would never buy a Gotway. Started as a Kingsong owner (16s), upgraded to the V11 (love it).  After reading 16 pages...after its paces of you fine people testing it out...this is my next wheel. 

[Tawpie]

2 hours ago, meepmeepmayer said:

We'll need 3D-printed protectors, I suspect.

3D parts goobered on is pretty much standard for GW! "Highly customizable" would be the marketing description. @GoGeorgeGo already has a pretty good looking set for his EX20S.

But I thought the front pad thing was going to protect the front batteries pretty well. It being there 'stock' tells you something though.