Community
Wiki Posts
Search
Like Tree8Likes

Capacitors

Thread Tools
 
Search this Thread
 
Old 01-24-2019, 07:10 PM
  #16  
Tech Initiate
iTrader: (2)
 
Join Date: Dec 2013
Posts: 46
Trader Rating: 2 (100%+)
Default

So I went and read your HobbyTalk post, I see you are talking about carpet oval, which I have raced in the past, and I assume you are talking about a spec class because you mention holding the throttle constant full.

I follow your calculations in that in that case of how capacitors would be essentially pointless.

I want want to bring in to discussion a different case. I for example am racing 17.5 2wd stock buggy on carpet. I mention carpet because we use huge amounts of drake brake (45-50%) and my current layout requires actually braking a lot both for corners and for scrubbing jumps.

I am using a Hobbywing V3.1 speedo that I’ve had for years. I always used the capacitors dutifully since I’ve had it simply because I thought I had to. Due to space constraints on my new buggy I did not use the capacitors at all and haven’t noticed any problems and the car really is fast.

But it I have heard is that capacitors would help with braking, and I could believe that they help absorb voltage spikes in this process as a sort of electrical shock absorber.

Thoughts?
CedarCityJohnny is offline  
Old 01-25-2019, 07:40 AM
  #17  
Tech Champion
iTrader: (168)
 
ta_man's Avatar
 
Join Date: Nov 2006
Location: New Jersey
Posts: 5,193
Trader Rating: 168 (100%+)
Default

Originally Posted by Bry195
I thought people understood that caps arent batteries. Can you talk about the math behind a capacitor that is capturing the regenerated power from a deceleration followed immediately by an acceleration? Stuff that happens at 100ms instead of seconds. I think you made a great case for getting a better battery for low frequency response (corners...) and im not saying that means you are wrong for high frequency application. Peak current use on a battery or dc bus takes away from the continuous available bus. It also generates heat. Regenerated current also creates heat.
The people that post was addressed to didn't seem to understand that a capacitor wasn't a battery.

First off an important point: The cap isn't capturing regenerated power from braking - that energy is going back into the battery. Think about the capacity relationship between capacitors and a LiPo battery:

One Farad can store one coulomb (one amp one second) at 1V. How many equivalent Farads are there in a 5000 mAHr battery? 5 amps for an hour is 5 amps for 3600 seconds or the equivalent of 18,000 coulombs. That is about 2 MILLION times the charge held by the capacitance of a Mamba Monster X (connected to an 8.4V pack). (That calculation is an approximation since the total coulombs in the caps depends on the exact charge voltage. But I hope you get the point that in terms of energy storage, that caps are irrelevant compared to the battery.)

Regarding a shorter time interval, 100ms as you asked about, you are still dealing with the fact that the Low ESR capacitors used on ESCs have such a minute capacity compared to the main power battery that they are still not going to contribute to improved acceleration strictly based on the current they can store.

I believe the principles I mentioned in the Hobbytalk post still apply: That all you can get out of the cap is a function of the difference between the peak voltage and the voltage under load. I'll do the following calculations using a 2S pack because that's what most people use. Lets say we do this for the first braking and acceleration of a race, where the pack is close to full. And for argument, lets say you are pulling a 100 amp burst the instant you get back on the throttle.

A typical IR on a battery is a few milli-ohms. We could be generous and say our hypothetical pack has an IR of 3 milli-ohms. But a more typical IR would be 5-6 Milli-ohms and give the caps more to do. So pulling 100 amps out of a 6 milli-ohm pack gives a voltage drop of 0.6V


How much are the typical caps on an ESC going to help with that during acceleration? So the regenerative braking as pumped the pack voltage up close to 8.4V. Not all the way but close enough to 8.4 that it is not unreasonable to use that number. And thus the caps are at 8.4V. Instant acceleration pulling 100 amps drops the pack voltage to 7.8V. So the caps see a voltage drop of 0.6V. From that and their capacity, we can determine how much current they can deliver over various time intervals.

Castle has stated that the Mamba Monster X has 1080 uF (microfarads) of capacitance. Lets use that in these calculations.

The available coulombs (charge) we can get from the 1080 uF caps on a voltage drop of 0.6V is 0.6 X 1080 / 1000000 which calculates out to 0.000648 coulombs. 1 coulomb is 1 amp for 1 second. Using your 100 ms time window, the cap gives you 0.0065 amps. 6.5 Milli-Amps. For 10 milliseconds it would be 65 milliamps which would be 0.07% of the total system current. With the small capacity of regular Low ESR caps, I doubt you could put enough of them in a car to make an improvement in performance that would outweigh the detriment caused by their weight. But that's a guess - I've never tried it.

Now lets look at what the super capacitors I mentioned above would do in this scenario.

With a full 1 Farad worth of capacitance in the system you would have 0.6 * 1 or 0.6 coulombs available in the caps when you punch it again. The initial implication is that you would get 6 amps out for the 100 milliseconds. But that won't really be the case for two reasons: 1) 6 amps is three times the current limit of the caps and would likely result in them being damaged in short order, and more importantly, 2) The caps have an IR of 1000 milli-ohms, a.k.a 1 ohm which would cause a voltage drop inside the caps of 6 volts - more than the voltage available. In other words, you can't get the available charge out of those super caps in the time frame you specified. With only 0.6 V differential between the caps and the battery, and 1 ohm resistance, the most you could get out is 600 milliamps, and that will decrease as the cap discharges and the voltage differential drops. But consider that if you did pull the 600 ma out of those, the output voltage would have dropped by 0.6V to be the same as the battery, thus no current flow. There is some point at which the two effects balance. So in practice, you'd get about 300 ma out initially, not the 600. Here the time interval is irrelevant because the IR of the caps is the limiting factor in how much current you can get out of them.

Add to the above discussion that even on acceleration, you are probably not pulling 100 amps for 100 ms, and thus the voltage drop of the pack and therefore the voltage differential on the cap (to allow discharge into the system) goes down rapidly meaning the current out of the caps goes down rapidly. Even with the super caps, you are talking a peak of 0.3% additional current and only that much because we used a higher than typical IR for the battery.

Another point worth mentioning is that while the above calculations appear to be independent of the actual battery voltage when you punch it and pull the 100 amps, they aren't really. The reason is if you pull 100 amps when the pack is at 8.4V, you won't get that much current out of it toward the end of the run because the peak voltage will be lower and so the peak current will be lower. With lower peak currents, comes lower voltage difference over the cap and thus the caps contribute less (than their already small amount) as the pack discharges.

That said, you do have competing factors that change the IR of the pack as the race progresses. One is that the less charge the pack has, the higher its IR. But as you run the pack it gets warmer, reducing the IR. Without measurements I could not guess which would be the overriding factor and thus whether the pack IR would go up or down as the run progressed.


An interesting point to note is that the super caps, with 926 times the capacitance, give you a peak of only 5 times as much current (using the 10 millisecond interval) as the low ESR caps on the ESC.

I will admit to using a lot of hypothetical numbers in the above calculations and thus my conclusions could be in error. But unless someone can show me some real data that disputes my assumptions, or can point out some error in my math. I'm going to stick with the premise that additional caps don't make an observable difference in racing situations.

I know there are guys that do speed runs and claim better numbers with large cap packs (of low ESR caps) than without, but I believe that is because of reductions in ripple current leading to less heating and thus better performance of the ESC. Especially since speed runs are full throttle which means the pack voltage is relatively stable compared to the on-off-on nature of racing involving corners.

If you go back to the beginning of this post and think about the 2 million to 1 ratio, the conclusions on the effectiveness of the capacitors in adding performance make a lot more sense.

Originally Posted by CedarCityJohnny
So I went and read your HobbyTalk post, I see you are talking about carpet oval, which I have raced in the past, and I assume you are talking about a spec class because you mention holding the throttle constant full.

I follow your calculations in that in that case of how capacitors would be essentially pointless.

I want want to bring in to discussion a different case. I for example am racing 17.5 2wd stock buggy on carpet. I mention carpet because we use huge amounts of drake brake (45-50%) and my current layout requires actually braking a lot both for corners and for scrubbing jumps.

I am using a Hobbywing V3.1 speedo that I’ve had for years. I always used the capacitors dutifully since I’ve had it simply because I thought I had to. Due to space constraints on my new buggy I did not use the capacitors at all and haven’t noticed any problems and the car really is fast.

But it I have heard is that capacitors would help with braking, and I could believe that they help absorb voltage spikes in this process as a sort of electrical shock absorber.

Thoughts?
Where caps do contribute (but only the low ESR caps) is reducing the ripple current that is detrimental to the ESC electronics but there we are talking about the ESC switching frequency, many thousands of HZ (cycles per second). Also this effect is mainly important at part throttle when the ESC is switching that rapidly. At full throttle, during acceleration, the motor is turning maybe 10,000 to 15,000 RPM. 15,000 RPM is 250 rev per sec and you have 12 (I think) phase switches per revolution for a four pole motor, making the switching frequency 3000 Hz.

In the case of the 17.5 motor in a truck or buggy on turf, you have the least stressful situation for the ESC as far as ripple current. For one thing, just by virtue of its winding resistance, the 17.5 can't pull the current you would see in a 1/8th scale off road car. Second, the RPMs aren't that high and being only a two pole motor, at full throttle you are likely only seeing a switching frequency of 1500 Hz. So while you can probably get away without the caps on your ESC, I personally would find a place to put them. I can't believe your body is so form fitting to the electronics that you can't find a place for them on top of something.
ta_man is offline  
Old 01-25-2019, 11:25 AM
  #18  
Tech Master
Thread Starter
 
Join Date: Jul 2018
Location: Florida
Posts: 1,011
Default

Originally Posted by ta_man
The people that post was addressed to didn't seem to understand that a capacitor wasn't a battery.

First off an important point: The cap isn't capturing regenerated power from braking - that energy is going back into the battery. Think about the capacity relationship between capacitors and a LiPo battery:

One Farad can store one coulomb (one amp one second) at 1V. How many equivalent Farads are there in a 5000 mAHr battery? 5 amps for an hour is 5 amps for 3600 seconds or the equivalent of 18,000 coulombs. That is about 2 MILLION times the charge held by the capacitance of a Mamba Monster X (connected to an 8.4V pack). (That calculation is an approximation since the total coulombs in the caps depends on the exact charge voltage. But I hope you get the point that in terms of energy storage, that caps are irrelevant compared to the battery.)

Regarding a shorter time interval, 100ms as you asked about, you are still dealing with the fact that the Low ESR capacitors used on ESCs have such a minute capacity compared to the main power battery that they are still not going to contribute to improved acceleration strictly based on the current they can store.

I believe the principles I mentioned in the Hobbytalk post still apply: That all you can get out of the cap is a function of the difference between the peak voltage and the voltage under load. I'll do the following calculations using a 2S pack because that's what most people use. Lets say we do this for the first braking and acceleration of a race, where the pack is close to full. And for argument, lets say you are pulling a 100 amp burst the instant you get back on the throttle.

A typical IR on a battery is a few milli-ohms. We could be generous and say our hypothetical pack has an IR of 3 milli-ohms. But a more typical IR would be 5-6 Milli-ohms and give the caps more to do. So pulling 100 amps out of a 6 milli-ohm pack gives a voltage drop of 0.6V


How much are the typical caps on an ESC going to help with that during acceleration? So the regenerative braking as pumped the pack voltage up close to 8.4V. Not all the way but close enough to 8.4 that it is not unreasonable to use that number. And thus the caps are at 8.4V. Instant acceleration pulling 100 amps drops the pack voltage to 7.8V. So the caps see a voltage drop of 0.6V. From that and their capacity, we can determine how much current they can deliver over various time intervals.

Castle has stated that the Mamba Monster X has 1080 uF (microfarads) of capacitance. Lets use that in these calculations.

The available coulombs (charge) we can get from the 1080 uF caps on a voltage drop of 0.6V is 0.6 X 1080 / 1000000 which calculates out to 0.000648 coulombs. 1 coulomb is 1 amp for 1 second. Using your 100 ms time window, the cap gives you 0.0065 amps. 6.5 Milli-Amps. For 10 milliseconds it would be 65 milliamps which would be 0.07% of the total system current. With the small capacity of regular Low ESR caps, I doubt you could put enough of them in a car to make an improvement in performance that would outweigh the detriment caused by their weight. But that's a guess - I've never tried it.

Now lets look at what the super capacitors I mentioned above would do in this scenario.

With a full 1 Farad worth of capacitance in the system you would have 0.6 * 1 or 0.6 coulombs available in the caps when you punch it again. The initial implication is that you would get 6 amps out for the 100 milliseconds. But that won't really be the case for two reasons: 1) 6 amps is three times the current limit of the caps and would likely result in them being damaged in short order, and more importantly, 2) The caps have an IR of 1000 milli-ohms, a.k.a 1 ohm which would cause a voltage drop inside the caps of 6 volts - more than the voltage available. In other words, you can't get the available charge out of those super caps in the time frame you specified. With only 0.6 V differential between the caps and the battery, and 1 ohm resistance, the most you could get out is 600 milliamps, and that will decrease as the cap discharges and the voltage differential drops. But consider that if you did pull the 600 ma out of those, the output voltage would have dropped by 0.6V to be the same as the battery, thus no current flow. There is some point at which the two effects balance. So in practice, you'd get about 300 ma out initially, not the 600. Here the time interval is irrelevant because the IR of the caps is the limiting factor in how much current you can get out of them.

Add to the above discussion that even on acceleration, you are probably not pulling 100 amps for 100 ms, and thus the voltage drop of the pack and therefore the voltage differential on the cap (to allow discharge into the system) goes down rapidly meaning the current out of the caps goes down rapidly. Even with the super caps, you are talking a peak of 0.3% additional current and only that much because we used a higher than typical IR for the battery.

Another point worth mentioning is that while the above calculations appear to be independent of the actual battery voltage when you punch it and pull the 100 amps, they aren't really. The reason is if you pull 100 amps when the pack is at 8.4V, you won't get that much current out of it toward the end of the run because the peak voltage will be lower and so the peak current will be lower. With lower peak currents, comes lower voltage difference over the cap and thus the caps contribute less (than their already small amount) as the pack discharges.

That said, you do have competing factors that change the IR of the pack as the race progresses. One is that the less charge the pack has, the higher its IR. But as you run the pack it gets warmer, reducing the IR. Without measurements I could not guess which would be the overriding factor and thus whether the pack IR would go up or down as the run progressed.


An interesting point to note is that the super caps, with 926 times the capacitance, give you a peak of only 5 times as much current (using the 10 millisecond interval) as the low ESR caps on the ESC.

I will admit to using a lot of hypothetical numbers in the above calculations and thus my conclusions could be in error. But unless someone can show me some real data that disputes my assumptions, or can point out some error in my math. I'm going to stick with the premise that additional caps don't make an observable difference in racing situations.

I know there are guys that do speed runs and claim better numbers with large cap packs (of low ESR caps) than without, but I believe that is because of reductions in ripple current leading to less heating and thus better performance of the ESC. Especially since speed runs are full throttle which means the pack voltage is relatively stable compared to the on-off-on nature of racing involving corners.

If you go back to the beginning of this post and think about the 2 million to 1 ratio, the conclusions on the effectiveness of the capacitors in adding performance make a lot more sense.



Where caps do contribute (but only the low ESR caps) is reducing the ripple current that is detrimental to the ESC electronics but there we are talking about the ESC switching frequency, many thousands of HZ (cycles per second). Also this effect is mainly important at part throttle when the ESC is switching that rapidly. At full throttle, during acceleration, the motor is turning maybe 10,000 to 15,000 RPM. 15,000 RPM is 250 rev per sec and you have 12 (I think) phase switches per revolution for a four pole motor, making the switching frequency 3000 Hz.

In the case of the 17.5 motor in a truck or buggy on turf, you have the least stressful situation for the ESC as far as ripple current. For one thing, just by virtue of its winding resistance, the 17.5 can't pull the current you would see in a 1/8th scale off road car. Second, the RPMs aren't that high and being only a two pole motor, at full throttle you are likely only seeing a switching frequency of 1500 Hz. So while you can probably get away without the caps on your ESC, I personally would find a place to put them. I can't believe your body is so form fitting to the electronics that you can't find a place for them on top of something.

somebody pick that mic mix back up and hand it to me please. Very understandable and I’m not going to check your math but the hypotheticals are reasonable.



What is the max voltage you would expect on a 2s 17.5 during decel? Would t it exceeded battery voltage and what would the cap do when that happens?
Bry195 is offline  
Old 01-25-2019, 02:14 PM
  #19  
Tech Champion
iTrader: (168)
 
ta_man's Avatar
 
Join Date: Nov 2006
Location: New Jersey
Posts: 5,193
Trader Rating: 168 (100%+)
Default

Originally Posted by Bry195



somebody pick that mic mix back up and hand it to me please. Very understandable and I’m not going to check your math but the hypotheticals are reasonable.



What is the max voltage you would expect on a 2s 17.5 during decel? Would t it exceeded battery voltage and what would the cap do when that happens?
The output voltage during braking can't exceed the input voltage during the powered phase.

When finally at top speed the back EMF in the motor coils matches the input voltage (if it didn't the motor would spin faster and faster until it did).

During braking you use that "back EMF" to extract energy. But the motor can't generate a higher back EMF than it was already generating and in fact the voltage will be reduced (according to Ohm's Law) because of the resistance of the motor coils (just like the internal resistance of a battery causes a voltage drop when you draw current from the battery).
ta_man is offline  
Old 01-25-2019, 08:19 PM
  #20  
Tech Master
Thread Starter
 
Join Date: Jul 2018
Location: Florida
Posts: 1,011
Default

Would you mind explaining what TA is in TA_Man? A car......body parts....electronic thug life...LOL, you were on a role and I wanted to see if you got jokes and brains to. You put allot into your explanations. more than i would or could have. Thank you.
Bry195 is offline  
Old 01-25-2019, 08:47 PM
  #21  
Tech Champion
iTrader: (168)
 
ta_man's Avatar
 
Join Date: Nov 2006
Location: New Jersey
Posts: 5,193
Trader Rating: 168 (100%+)
Default

Originally Posted by Bry195
Would you mind explaining what TA is in TA_Man? A car......body parts....electronic thug life...LOL, you were on a role and I wanted to see if you got jokes and brains to. You put allot into your explanations. more than i would or could have. Thank you.


TA is for Trans Am, the car in my avatar, a 1971 with a 455 C.I. engine. I owned that one for 30 years from 1974 to 2004. At one point I had three all at once, one each red, white, and blue. Now I just have one, a "Pewter Metallic". 2001.

Funny story: In 2001 when I bought the "new" car, I sold one of the "old" ones, a 1978, to a neighbor who lives diagonally across the corner from me, He drove it home and it has not been out of his garage since then. I can see it from my front porch when he has his garage doors open.

I have jokes, but usually just for my wife. RC cars are serious stuff.
ta_man is offline  
Old 01-26-2019, 02:26 AM
  #22  
Tech Master
 
Join Date: Jul 2009
Location: Deep South Texas
Posts: 1,416
Default

I WISH I had the techno skills to speak or type the stiff I know..

RC cars are serious stuff.... and FUN....

I miss my 65 Impala SS 409, my 1979 Corvette 383, my 88 CBR1000F... wife's hospital bills got them all... and this time the house.


.
2 years of mechanical engineering, 3 years of electronic design..
but the job's keep moving out of the country.
Texas Instruments purchased the company I last worked for.. and they closed it..
chuck_thehammer is offline  
Old 01-26-2019, 12:04 PM
  #23  
Tech Master
iTrader: (156)
 
Join Date: Sep 2007
Posts: 1,372
Trader Rating: 156 (100%+)
Default

So after all this discussion most of which I only slightly understood. Is is safe to say a 3 cap pack or one cap like tekin makes no difference for us running 10th.
Breeze is offline  
Old 01-26-2019, 12:15 PM
  #24  
Tech Lord
 
Roelof's Avatar
 
Join Date: Aug 2007
Location: Holland
Posts: 12,343
Default

Performance wise a change of capacitor will not give a noticeable thing. But good caps and not a too low value will keep the FET's of the ESC colder and so healthier
chuck_thehammer likes this.
Roelof is offline  
Old 01-26-2019, 12:49 PM
  #25  
Tech Master
 
Join Date: Jul 2009
Location: Deep South Texas
Posts: 1,416
Default

Originally Posted by Breeze
So after all this discussion most of which I only slightly understood. Is is safe to say a 3 cap pack or one cap like tekin makes no difference for us running 10th.
I do NOT think it makes a big difference between first or second... or tenth...
chuck_thehammer is offline  
Old 01-26-2019, 08:58 PM
  #26  
Tech Master
Thread Starter
 
Join Date: Jul 2018
Location: Florida
Posts: 1,011
Default

I was ran over by a TA. I bought it with a set of ridiculously wide kragers on the rear. If push starting a car isnt complicated enough try to push start a car short steppeing so you dont get sucked under the car. LOL. Not a joke. Real life. I think that was my one and only domestic other than an international scout, porsche 914, vw type 3, triumph gt6.

Industrial applications that have a dc bus that floats between 680 and 800 volts do use capacitors. They are only used for the accel and deceleration that is in a very short time frame. They can return about 50% to the accel but they are for very specific reciprocating applications. I think a part number is HLC01.1 from bosch if you are interested.

Bry195 is offline  

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are Off



Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service -

Copyright © 2024 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.