Sabvoton SVMC 72150 doesn't do full power

[Sparfuchs]

Oh I see, these unloaded voltage drop thats why so small drop. Well let's see loaded V drop. I expect unfortunately it to be very many more volts due too few parallel cells.

Ok, just had a awesome testride in a snowy park. As it does "burnouts" in the snow it felt like way more power than without snow
That's how I imagine the bike to be in mod or gravel if it kicks 10kw instead of 5kw
I got at full throttle a voltage drop of up to 6V or 7v for a few seconds I guess. Maybe I'll try again when its not snowing or with a bodycam to record the c.a. because it's quite hard to focus on the screen if the bike does ether kick downs or wheelies at full throttle without crashing.
What do you think of the voltage drop ?

 

[Sparfuchs]

You got mxus turbo so called 3kw hub in 24"" wheel, i assume it is 9kv torque wound it is not power hungry because it is low kv instead 11 or 13 as it could be. Whats your weight? Whats ebike weight? Maybe everything is very light and you get instantly hight rpm and running and it does not need more power, can't take more power. You mentioned voltage drop is suspiciously small and would call even neglegible. Must be way bigger voltage sag or there actually is something preventing these amps "flowing".

Yes,yes, and allmost.. it's the 5T = 7.4 Kv version as it still does enough top speed and if I wanna go "crazy" I can activate flux weakening.
My weight is about 80kg/176lbs and I guess the bike is about 40kg/88lbs.
Is this actually possible ? Wouldn't the controller still "have to" do full amps at full throttle at least at beginning and do kickdowns (or however its called when the tire spins faster than it accelerates) or wheelies at full throttle if it can't accelerate fast enough ?
I've also tried it at a very step hill and it still was "limited" to about 80amps but could have done way more speed.

I'll try again with front brake a bit pulled to see what the amps do.

Everything feels like there is a "secret 80 dc amps limit" I just don't know about

 

[amberwolf]

I got at full throttle a voltage drop of up to 6V or 7v for a few seconds I guess.

Your pack is 19s4p, so:

7v, divided by 19s, is 0.37v per cell.

To have some idea how bad or good that might be, we need to know a couple of other things:
What was the resting voltage during the test?
What was Amax on the CA during the test?
What was Vmin on the CA during the test?

For instance, if Amax was only 10A, then that's pretty bad sag.

If Amax was 100A, then it's not bad.

Maybe I'll try again when its not snowing or with a bodycam to record the c.a. because it's quite hard to focus on the screen if the bike does ether kick downs or wheelies at full throttle without crashing.

Vmin on the CA will tell you what the voltage drop was, compared to the voltage with no load. Just check it right after you try some particular test, then reset the CA's trip stats (hold rigth button for a few seconds) and try the next test.

 

[minde28383]

Just now i spoke with acquittance who rides very similar setup, his weight is about the same, ebike weight slightly heavier, also has 24'' and the same motor but in 9kv, but his battery is 18s (from car, could give way more than 100a bursts); his controller is sabvoton 100a (not 150a like yours). On sabvoton app he sees max amps ie 100a on startups; feels that controller is the limit. It does not reveal anything specifically, we need to think about it more and maybe give another test when you can.

Follow amberwolfs suggestions for sure.

Yours motor is low kv, but yes, probably you should see little bit more than 80a on most powerful starts. I think battery might be little bit limiting but follow those voltage check ups, maybe results (voltage) will be revealing.

 

[Sparfuchs]

Your pack is 19s4p, so:

7v, divided by 19s, is 0.37v per cell.

To have some idea how bad or good that might be, we need to know a couple of other things:
What was the resting voltage during the test?
What was Amax on the CA during the test?
What was Vmin on the CA during the test?

For instance, if Amax was only 10A, then that's pretty bad sag.

If Amax was 100A, then it's not bad.

Vmin on the CA will tell you what the voltage drop was, compared to the voltage with no load. Just check it right after you try some particular test, then reset the CA's trip stats (hold rigth button for a few seconds) and try the next test.

what do the 7v and the result 0.37v mean ?

Ok...just did another test ride with an trip reset before to get all those values:

resting voltage: 69,2v (resting means the "stand still" voltage of the battery when I started the test I assume ?)
Amax: 88,4
Vmin: 59.8 (
So means a voltage drop of about 10v...sounds quite a lot to me ? Is it ?

I'll check again with the battery fully charged to see how it influences the results.

 

[Sparfuchs]

Just now i spoke with acquittance who rides very similar setup, his weight is about the same, ebike weight slightly heavier, also has 24'' and the same motor but in 9kv, but his battery is 18s (from car, could give way more than 100a bursts); his controller is sabvoton 100a (not 150a like yours). On sabvoton app he sees max amps ie 100a on startups; feels that controller is the limit. It does not reveal anything specifically, we need to think about it more and maybe give another test when you can.

Follow amberwolfs suggestions for sure.

Yours motor is low kv, but yes, probably you should see little bit more than 80a on most powerful starts. I think battery might be little bit limiting but follow those voltage check ups, maybe results (voltage) will be revealing.

Another test when I can ? Sure whatever it takes, just let me know !

My next try will be some more tests with the different limiting options to see what they do and also tests with flux weakening to see if that's a way to get more than 80 amps to see if its even possible.

Thanks a lot for the Screenshots. My App looks different but in case it't about the balance and there difference I'll still share below.

Now that I know that the bms can't limit the current, but only cut according to the set values I assume all those values are important for safe use and cycle life but can't be the reason for "my problem" .

So I guess I'll have to find the issue in the controller or more specific it's settings.

 

[dominik h]

resting voltage: 69,2v (resting means the "stand still" voltage of the battery when I started the test I assume ?)
Amax: 88,4
Vmin: 59.8 (
So means a voltage drop of about 10v...sounds quite a lot to me ? Is it ?

10V drop at 88A is quite a lot. But the battery was not full and perhaps not warm enough.

For 150A you will see a voltage drop of 17V which is huge. Try again with activated flux weakening and einher you will see more amps, or your BMS will cut off, or your sabvoton has the wrong firmware which limits your current

In my scooter I see 2,2V drop @175A and 20C. At 10C I see a dop around 4V.

You can also try to heat your battery to 30-35C before testing , this will lower internal resistance a lot.

 

[minde28383]

what is wrong firmware on sabvoton? never encountered it to have wrong firmware. Had hands on several sabvotons.

Like you already planned. Fully charge up you batt and repeat the tests with fully charged without field weakening and than again with with field weakening with fully charged, but i don't know how you should you load your ebike to notice field weakening amperage. You hub is 7kv and I wonder if you would see any amp increase over 80a at max speed even if battery would be able to give more than 80a. At stand still startups you could see more amps, probably could see over 100a, but when you reach your max rpm on road, your speed will be about 50-55kmh and consumption about 20-30amps, you add field weakening about 50a (preferably up to 30a), you still in 80a range. Excessive heating will be fast witch such weakening and I never burned or demagnetized motor in such way by i see alert precautions about high field weakening.

 

[Sparfuchs]

resting voltage: 69,2v (resting means the "stand still" voltage of the battery when I started the test I assume ?)
Amax: 88,4
Vmin: 59.8 (
So means a voltage drop of about 10v...sounds quite a lot to me ? Is it ?

10V drop at 88A is quite a lot. But the battery was not full and perhaps not warm enough.

For 150A you will see a voltage drop of 17V which is huge. Try again with activated flux weakening and einher you will see more amps, or your BMS will cut off, or your sabvoton has the wrong firmware which limits your current

In my scooter I see 2,2V drop @175A and 20C. At 10C I see a dop around 4V.

You can also try to heat your battery to 30-35C before testing , this will lower internal resistance a lot.

To be honest, the voltage drop was something I didn't think of enough when I "calculated" my battery packs output. I just thought as the samsung 40t is a high draw cell rated for 45amps it can't be that bad

Sure it's a linear equation from 88a to 150a ?

I usually never discharge below 2,63v per cell, but does it also harm the cycle life a lot to go below 2,5v if it's just a few seconds caused by voltage drop at bursts ? If it won't harm a lot I could just lower the low voltage as I check the voltage on the c.a. frequently anyway.. especially if It gets closer to the end.

I've really bad experience with all the sabvoton softwares because they seem to act so unpredictable to me and do things like disable my e brake and keep it disabled even if I disable and enable it again, or showing status: flux weakening enabled even it's disabled in the software and in real, or if I enable flux weakening not letting me disamble it.A reset usually solves all those problems but I've no Idea why it does that for no reason.

That's one of the reasons why I don't really trust the sabvoton if it say's no Limit and 150a dc current, because who knows what is actually set

As there are at least 3 different softwares, is there something like a "best" software ?

well, heating the battery lowers resistance but isn't it falsify the result if I don't usually get this as real performance ? Or do you mean just to compensate the winter in the result ?
But I always store it at room temperature... but I don't know how fast it cools down when I go for a ride. As there a 3 temp sensors in the pack I'll report the temperatures next ride

 

[dominik h]

I got a Frdriver ND9850 with wrong firmware on it.
I could not set max volts in the desired range 115-120V it told me 90,5V is the max limit. After the firmware update i could set and use 118V. The current limit is also written down in the firmware.

On a flat surface you will see no real difference in amps with field weakening, but uphill it is posible to get higher max amps with it.

Can you set more than 150dc amps in the software and does the controller accept the higher value or is it corrected back to 150A?

The heated battery is only for comparison.

Voltage drop is depending on internal resistance and current.
V=IxR ( sorry don't realy know what the english shortcuts for [Voltage = Current x resistance] are)
And it is nearly linear.

 

[Sparfuchs]

what is wrong firmware on sabvoton? never encountered it to have wrong firmware. Had hands on several sabvotons.

Like you already planned. Fully charge up you batt and repeat the tests with fully charged without field weakening and than again with with field weakening with fully charged, but i don't know how you should you load your ebike to notice field weakening amperage. You hub is 7kv and I wonder if you would see any amp increase over 80a at max speed even if battery would be able to give more than 80a. At stand still startups you could see more amps, probably could see over 100a, but when you reach your max rpm on road, your speed will be about 50-55kmh and consumption about 20-30amps, you add field weakening about 50a (preferably up to 30a), you still in 80a range. Excessive heating will be fast witch such weakening and I never burned or demagnetized motor in such way by i see alert precautions about high field weakening.

I guess if you use the "right" software and set the values experienced you won't have any issues. But if you still have to try around a lot by changing the values all the times with softwares and app it might "confuse" the controller or offer more potential for "unexplainable" issues. Also the different manuals with some mistakes or different values recommended are confusing for new users. Maybe it's just but me, but it's not my first sabvoton and also many threads here show that weird behavior is possible. But as there is no "better" alternative in that price ratio it's worth to get to know the sabvoton I guess :wink:

I've already tried with flux weakening a few weeks ago and the amp draw was noticeable but I don't remember the values.
I thought field weakening is only active if it gets close to the max rpm per volt to increase the top speed ?
Is it possibility that:
The controller gets with my motor to it's limit at 80amps with the sinewave speed/pulses/calculation (or however this works) if flux weakening is disabled ?
That the controller needs f.w. enabled to be able to use more amps at acceleration ?
That f.w. does not only increase the top speed but also the torque/acceleration significant ?
That the controller intended reduces the max dc amps (without showing) if f.w. is disabled ?

Really looking forward to my next test rides and checking it

Edit: I confused firmware with software in my text above. I was only talking of software

 

[Sparfuchs]

I got a Frdriver ND9850 with wrong firmware on it.
I could not set max volts in the desired range 115-120V it told me 90,5V is the max limit. After the firmware update i could set and use 118V. The current limit is also written down in the firmware.

On a flat surface you will see no real difference in amps with field weakening, but uphill it is posible to get higher max amps with it.

Can you set more than 150dc amps in the software and does the controller accept the higher value or is it corrected back to 150A?

The heated battery is only for comparison.

Voltage drop is depending on internal resistance and current.
V=IxR ( sorry don't realy know what the english shortcuts for [Voltage = Current x resistance] are)
And it is nearly linear.

That's another mystery to me what options the firmeware has to mess up things without letting me know.
I bought it as a Unlocked 150 amp (unlocked 200) from aliexpress and with my experience of aliexpress sellers it wouldn't surprise me to much if they did instead of uploading the unlocked firmeware, upload the 100a firmeware.. or the unlocked 45amps software :lol:
How do I check ? And how do I do a update ?

BUT with the two "official?" softwares I could set the dc amps to 200 amps and it saved it... but who knows if the firmeware agrees

 

[dominik h]

If the controller does not accept higher values it just shows 150A after a few seconds in the App.

Your Motor is wound for low speed, so you have only a small gap to reach the max amerage of the controller

I have a QS69V5000w which does 103kpth without flux, around 90kpth the current starts to drop and from 0-40kp/h the current raises to it's maximum.

In the same scooter with the lower kv erider 8kw Motor, it reaches max amps at 25kp/h and starts lowering amps at around 70km/h topping out at 85kp/h. With weakening field I increase topspeed to 98kp/h and dropping of current starts at 90kp/h.

Perhaps your "sweet spot" during acceleration is to short to see the full current peak.
My scooter needs 7 seconds from 30 to 90kp/h , enough time to rise current for the sabvoton.

 

[Sparfuchs]

If the controller does not accept higher values it just shows 150A after a few seconds in the App.

Your Motor is wound for liw speed, so you have only a small gap to reach the max amerage of the controller

I have a QS69V5000w which does 103kpth without flux, around 90kpth the current starts to drop and from 0-40kp/h the current raises to it's maximum.

In the same scooter with the lower kv erider 8kw Motor, it reaches max amps at 25kp/h and starts lowering amps at around 70km/h topping out at 85kp/h. With weakening field I increase topspeed to 98kp/h and dropping of current starts at 90kp/h.

Perhaps your "sweet spot" during acceleration is to short to see the full current peak.
My scooter needs 7 seconds from 30 to 90kp/h , enough time to rise current for the sabvoton.

In the software I know for sure it did accept it and I'm pretty sure I would have noticed if the would have changed my values but I can't tell for sure, that I tried with unlocked values above 150 amps in the app. As I didn't even get 100amps I decided to make it work with the normal limits first.

The "sweet spot" is a very interesting thought.
But when I tested it very step uphill and didn't get a lot of top speed it also limited at 80 amps. There should have been the potential to draw more at low speed I guess.
It does wheelies from stillstand or below 10 kph even if I lean forwards. So I won't benefit from way more torque from still stand on asphalt if I don't wanna be dropped. But I'd really appreciate more torque from 20 to 35 kph.... and as I'm writing down what I'd like I realize that for my low kv motor and the the hint of the "to low sweet spot" it might not have been the best idea to set the mid current to 250 and hoping for awesome acceleration. :lol: With the new knowledge there might be a possibility that my, at the moment still sporty preset value of 175 might lower my already low sweet spot to a not noticeable/usable low area?!

Edit: I'm afraid full throttle is full throttle so there will not be more torque at higher speed if I lower the mid current. It would just influence the throttle behavior according to it's position, right

 

[DogDipstick]

resting voltage: 69,2v (resting means the "stand still" voltage of the battery when I started the test I assume ?)
Amax: 88,4
Vmin: 59.8 (
So means a voltage drop of about 10v...sounds quite a lot to me ? Is it ?

I'll check again with the battery fully charged to see how it influences the results.

Drops 10v for 80A? 59.8v, isnt that on the verge of low voltage cut ( LVC?) The bike I have been describing drops 7v for 9000w with its 25ah battery.. 125A, 7v drop..... say.... from 84v full, to 77v @ 9000w ~125A. Recycled EV cell.

I have not been much higher than that because it is not my bike and I only build it and am afraid of the wheelie. As is, id does not reach enough tractive effort to pull very much higher amps.

If a battery is nearing LVC on the discharge, you have a poor setup, either the load ( bogging) on the motor or the motor tries to push to much, or a pack that cannot do 150A... perhaps. I dont know.

Edit: I'm afraid full throttle is full throttle so there will not be more torque at higher speed if I lower the mid current. It would just influence the throttle behavior according to it's position, right

Good so we can rule out th4rottle flukes, if the TH is true when it reaches full throttle output voltage form the potential meter.

 

[dominik h]

If you reduce mid current you have to trist the throttle more to reach the same phase current.

 

[amberwolf]

To be honest, the voltage drop was something I didn't think of enough when I "calculated" my battery packs output. I just thought as the samsung 40t is a high draw cell rated for 45amps it can't be that bad

Sure it's a linear equation from 88a to 150a ?

Look at the cell spec sheet, or reliable testing sites like this one.
https://lygte-info.dk/review/batteries2012/Samsung%20INR21700-40T%204000mAh%20(Cyan)%20UK.html
That has a graph on it that shows you what the voltage drop is at various currents, for a single cell, for any particular state of charge.
This one shows sag at 30A per cell is about 0.35v even at completley full, so for instance a 19s pack would be 19 x 0.35v = 6.65v sag even when full at 80v resting, if each cell is loaded to 30A (for 4p that's 120A). You can use the charts and math to figure out with your loading of them to 88A at a lower per-cell voltage (SoC) to see if you may have a problem with the pack beyond just insufficient parallel cells.
https://lygte-info.dk/pic/Batteries2012/Samsung%20INR21700-40T%204000mAh%20(Cyan)/Samsung%20INR21700-40T%204000mAh%20(Cyan)-Temp-30.0.png


Then you multiply that by the number of series cells for voltage, and the number of parallel cells for current, to see what you will get for brand new well-matched cells of that model, at that current.

If your voltage sag is worse than that, then your cells are either not new, not well-matched (so some don't meet specs and sag worse than others), or not matching the spec sheet for some other reason, and you'll have to test them yourself under your specific conditions to establish their characteristics.

Or there is an interconnect problem (anywhere between cells, or battery and controller) causing high resistance, and voltage sag is happening there, instead of in the cells themselves.

 

[minde28383]

App is far from ideal. But this controller is one of the most user friendly.
Regarding field weakening false indication, yes, it happened to me too. App shows it's ON but it was not activated. Never found out why but I know that It works for others and If i ever need more speed than I'll probably will try with another Sabvoton app to make it active.
Regarding Locked or unlocked. You setup draw is no where near controller limits so unlocking or increasing it to 200a it would not have effect. If seller sold it as uncloked than hopefully it is unlocked, but I noticed that some sabvoton app changes to higher amps and some do not. Even when controller is unclocked.

i mentioned regarding for not enough battery but have few more things to notice.
Regarding you cells. I would count 20a max when building pack with such cells. 20a max discharge because your cells are good specs, otherwise 15a, 10a, and there are cells in use with constant 8a etc. So 20a is a lot for you cells. Your batt consists of 76pcs, it should be at least twice as big when you got such motor and controller. By increasing cells number you would prolong pack life (longevity) - less sag, more cycles, less heat, less disbalance, would not need to torture it every time you push to its limits.

You max voltage about 80v
Let's assume on full throttle your voltages drops to about 63v (3.3v per cell), if cold even less, moreover if your your battery is not charged up max at that moment than you get voltage drop bellow 60v, and also might be starting to approach ''sabvoton lowering power limit''. You are starting with high voltage and end up with very high a voltage drop; I don't know how controller is programmed to act. It might be controller limits your power when it sees certain voltage drop. Naturally power output decreases with lower voltage but it might be that controller itself also starts to limit the power consumption to preserve your battery. Maybe what you see on CA LCD might be not true because controller works much faster than digits gets reflected on the LCD, cells approach lower voltage and controller might be limiting power consumption

Very likely test with higher discharge (or less sag) pack would be an answer regarding 80a limit.

 

[Sparfuchs]

Update:
I did a lot of testing and got following full throttle/voltage drop results from the always reseted c.a.:
From fully charged (79v/ 4,16 per cell) to 76v resting voltage, the pack had about 8-10v drop at about 90A-100A.
Below resting 75v I got the voltage drop but also the max amps get a bit lower (about 8v at 85 to 95 amps).


Unfortunately the following settings had no (noticeable) influence at the results:

current loop kp: 300 vs. 999
Flux weakening: disabled vs. enabled at 70A (but more top speed)
Throttle mid current: 175A vs.100A
Speed Limit: Internal limit (100%) vs. no limit
C.A. Throttle Mode: by-pass vs. directly plugged into controller (what should be the same)

The voltage drop of the battery pack could (theoretical?) be almost three times more (30V) than it is without getting below my lack voltage (50v), or current limit voltage (52v). Not sure how many amps I could draw before that happens, or how the drop physically reduces the max amp draw, but if a single cell in official and independent tests does up to 70 amps busts before getting hot and 45 amps for quite a while before it gets to 80°C, I'd assume I didn't push it to the limit in my 19s4p with 100 amps 25 amps per cell (just busts for few ms!) and no rising temperatures at all. And as the 300 amp bms can't limit but only cut and doesn't the problem (that probably won't be there with a bigger battery) seem to be the controller.

So my conclusion is like minde28383 said, that the sabvoton firmeware is programmed to protect the battery with a correlation of max amps and v drop. Like limiting the max amps to a voltage drop of 10v or something like max 10 amps per dropped voltage or whatever. That would explain why it could not go above 104 amps and 10v drop ?

I know, I should build a bigger battery... But is there a way to check or manipulate the firmeware or is there a different controller for about 150 dc amps that definitely doesn't have voltage drop protection ?

 

[Sparfuchs]

Or are there any ideas how I could "trick" the controller ? Thinking of something like a capacitor with diodes if needed, that doesn't influence the real voltage drop in the pack but keeps the voltage at bursts stable to prevent the controller from limiting/protecting ?

 

[dominik h]

Try pre-heating the battery to 35C.
Than the voltage drop is much smaller and you can see how your controller reacts.

 

[Sparfuchs]

Or are there any ideas how I could "trick" the controller ? Thinking of something like a capacitor with diodes if needed, that doesn't influence the real voltage drop in the pack but keeps the voltage at bursts stable to prevent the controller from limiting/protecting ?

Maybe I overestimate "my idea", but it might be a way to solve a problem that many sabvoton users out there might have ?
As it's a way different topic that should also be found independent of my problem I made a new thread:

https://endless-sphere.com/forums/viewtopic.php?f=14&t=118970

 

[Bobby.T]

Kp is most likely a PID regulation parameter for the current control.

If so:
PID parameters are called kp (proportional), ki (integral) and kd(derivative) constants. They each have their function in the regulation and only adjusting one of them will eventually lead to overshoot of the current. This means too high kp and you will get a surge of acceleration which might be uncomfortable or dangerous. With a too low kp the acceleration(=current) and speed will grow slowly so you’ll not see high currents.

The correct value depends on what you want your motor to do - fast and agressive wheel flipping or docile grocery getter or something in between?

In motor control the correct kp parameter will be dependent on kv ( if you’ve got a low motor with lot’s of torque per amp then obviously the acceleration will be higher for the same current ramp up.)

What i’m trying say is that it isn’t true that there’s a general kp for, let’s say, 10inch motors or even for a specific QS205. There might be a value that historically has suited many riders (or at least not annoyed them to the point of contacting seller) and this is what Kathy has described.

I am running a sabvoton 96120 on a 23s70ah 150amp bms on a qs q260v4 5000w, using an apple I can set my flux weakening to 150 and get 115 km/hr on it. Its an emmo kamen. However its low speed accel is trash, barely will spin the tire on gravel. I am not sure if my controller is unlocked but when ever I try to use the PC app all I get is USB not found and the driver file doesn't show up.
So 2 questions, 1 do you know how to get the PC program to finally connect? I ordered the bluetooth so they didn't send me the cable, is it a special cable I need?
2nd, Do you know if I eventually get it hooked up to a PC and can adjust my kp, if I up the kp do I have to adjust the other 2?

 

[Bobby.T]

Or are there any ideas how I could "trick" the controller ? Thinking of something like a capacitor with diodes if needed, that doesn't influence the real voltage drop in the pack but keeps the voltage at bursts stable to prevent the controller from limiting/protecting ?

Maybe I overestimate "my idea", but it might be a way to solve a problem that many sabvoton users out there might have ?
As it's a way different topic that should also be found independent of my problem I made a new thread:

https://endless-sphere.com/forums/viewtopic.php?f=14&t=118970

I see you are hooked up to a PC, did yours come with the cable? I had ordered bluetooth so no cable. I picked up a generic USB to USB and no matter what I do I can't get it to connect.
I want to make sure my 96120 is fully unlocked and see if I can solve my weak initial acceleration.

 

[minde28383]

generic USB to USB no good. You need adapter which looks like general cable but one end is thicker because there is chip in it for data exchange between pc and controller.