Ripple voltage???

Intlwaters

Help Support Intlwaters:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

HTV Boats

Well-Known Member
Joined
Nov 8, 2006
Messages
2,510
Was doing some SAW prop tests yesterday and was reviewing the data on my Castle ESC's and looking at the ripple voltage I don't know what's good or bad. On my Castle 240 ICE LV on 4S I was seeing 2-3.1v max during hard draw and on my 200 ICE HV on 10S it was 2-2.8 max. Both have cap banks. Peaking at 260amps (4S) and 190amps on 10S. All info welcome.

Mic
 
1527 1D 6S Castle Ice 240 LV plus capbank in my saw rigger ripple 2,50v and peaking at 286,3A so yours looks good to me. Gill
 
For normal running you don't want ripple spikes of more than 10% of pack voltage. For extreme SAW running you can live with more. My 2011 Q offshore record was set with a 1515 motor, and I saw max ripples currents of 3.2 volts (@ 280 amps). The actual ripple on the run out to the start of the first pass was higher than the two record passes - this was probably because the packs still warming up. The P Offshore record was set with a 1521 and it showed a ripple of 3.4 volts (@ 375 amps) - a LOT more considering the pack voltage. The big motor drew a lot more current. All records were 1P which didn't help hold the ripple down. All wires were very short.

.
 
Last edited by a moderator:
So how much uf (micro farad) are you guys adding to the controllers on you P and Q SAW set ups??
 
Mike,

I have been using 2 1000uf Rubicons (from OSE) in parallel on my Swordy's. On one 240 Castle I have the Castle pack which has 4 caps. They are much smaller, less half the physical size of the 1000's. Can't read the actual capacitance through the shrink wrap. On my 200 ICE I have a Hefei which solders easily between the wires. Can't see spec again but they are huge. Bigger than the 1000's. Etti has a 5000uf HV bank on OSE

http://www.offshoreelectrics.com/proddetail.php?prod=hef-cap-hv&cat=136

In my tunnel boxes I have space limitations and the Hefei is about as large as I can go. I have a Fighter Cat that has 5 large caps in line and 12ga. wire. Not sure how much would be too much as the factory Castle is not very large and they should know what's needed. I guess it amounts to some overkill just more protection IMO. I don't think caps reduce spikes as much as they spread out the duration. Kind of like the in line pressure tanks on top of your water heater keeps your pipes from pounding. Pressure is the same just acts like a shock absorber.

Mic
 
Last edited by a moderator:
mine is Etti 5000uf saw capbank from OSE...Gill

DSCN4733.JPG
 
Last edited by a moderator:
I blew my castle 240 when I overloaded it. My fault. It did not have caps added. The ripple was 5 to 6 when it decided to be a pop tart. I now run cap banks and don't overload the esc. So having said that I would use those numbers as BAD numbers to see.
 
Thanks for the posts and pics guys.

Mic, I picked up some of the 35V caps for P and Q here. 80 cents a piece and I think shipping was only 2 bucks. http://www.thecapking.com/35rx1000.html

Hey John, we don't know what a good number is unless we have a bad number to compare it to. Frying a $300 controller sucks though!! :( I have a Castle Ice sitting on the bottom of the Capitol pond from last year (Note to self: Screw down hatches on SAW boats). The FE community has a rich history of learning from the misfortunes of others. You are now an official member of the "I took one for the team" club. See you this weekend. Bring one of your cap banks so I can take a look at it.
 
Last edited by a moderator:
Thanks for the posts and pics guys.

Mic, I picked up some of the 35V caps for P and Q here. 80 cents a piece and I think shipping was only 2 bucks. http://www.thecapking.com/35rx1000.html

Hey John, we don't know what a good number is unless we have a bad number to compare it to. Frying a $300 controller sucks though!! :( I have a Castle Ice sitting on the bottom of the Capitol pond from last year (Note to self: Screw down hatches on SAW boats). The FE community has a rich history of learning from the misfortunes of others. You are now an official member of the "I took one for the team" club. See you this weekend. Bring one of your cap banks so I can take a look at it.
That's kind of interesting as the caps OSE sells for .75 have a different part number and the specs on ripple current read 2480 vs 1430 on the ones you listed. Anybody know if one is better than the other? On my smaller esc's I just glue and shrink wrap a pair shorten and solder 12ga wire to the leads in parallel then to pos and neg on the ESC.

Mic
 
In general as Jay mentioned you want to keep the ripple voltage low as a percentage of the total pack voltage. The sensitivity to ripple voltage increases dramatically the closer you get to the caps voltage limit. For example, if you run a 8S rated ESC at 8S and the caps are only rated for 35V it is quite easy for the bus voltage plus ripple voltage (30V +5V ripple=35V) to exceed the maximum voltage of the caps. This is often why user's or companies recommend never running an ESC at it's rated voltage. Some companies match the Cap voltage levels too close to the ceiling voltage level in my opinion. It's a compromise between ripple voltage performance and capacity. One might assume it would just be better to add 100V caps on a 50V system, but then you only have 50% bus voltage utilization from the caps.

The type, capacity and quantity of caps also make a difference. I will also point out that actual ripple voltage tends to be higher than what is logged by our hobby ESC's. I have conducted tests on professional level 3 phase controllers where measured ripple current from the datalogger was 45% of the actual peak as measured with a specific scope and current probe. So although you are logging 4V ripple's, in actuality the ripple is closer to 10V.

Adding capacity helps to some extent, but placing the extra caps away from the board does not gain as much as if the caps were directly mounted on the board. The extra lead length can in some cases causes a resonance which can actually be more detrimental than beneficial.

I'm going off on too much of a tangent here, but good rules to follow:

Use heavy gauge short wire with good connectors (this assumes good solder and soldering skills as well)

For high current applications, leave some headroom between the operating voltage and the rated voltage of the ESC.

If extra caps are deemed necessary, use the same type and voltage level. Keep the caps as close to the board as possible and avoid long length of wire to remote mount them.

View attachment 51248

Here is a shot from a SAW pass running a 6S1P, Note 400A peaks and ripple is around 4V peak. ESC ran perfectly with no extra caps, just short wires, a stout battery and good connectors.

TG
 
That's kind of interesting as the caps OSE sells for .75 have a different part number and the specs on ripple current read 2480 vs 1430 on the ones you listed. Anybody know if one is better than the other? On my smaller esc's I just glue and shrink wrap a pair shorten and solder 12ga wire to the leads in parallel then to pos and neg on the ESC.

Mic
Mic, Which OSE caps were you referring to??

Great info Tyler!!
 
Last edited by a moderator:
I forgot Steve sold loose caps because I have been looking in the Brushed ESC section where the cap banks are located.

The difference in the two caps is temperature rating. The ones OSE sell have a range up to 105 degrees C and the ones I purchased are rated up to 130 degrees C. Don't know if the higher temperature range is needed in our application. If our Caps are pushing 220 F my guess is we have bigger problems to worry about. LOL

Some of Tyler's info pretty much reinforced what I have been taught and my experience last year with my SAW boat. The rest was over my head. LOL Unfortunately I was using a ESC without data logging so I didn't have the numbers to prove it one way or another. I'm not going to lie and say caps don't make a difference in some instances. I did add caps to my first P-mono heat racer and it did help keep the ESC cooler. My gear at the time was not the best quality. So I guess it's up to us individually to decide whether Caps are a necessity or a band aid. I may do some testing with one of my boats with and without at SAW's this year.
 
Last edited by a moderator:
Thanks for the posts and pics guys.

Mic, I picked up some of the 35V caps for P and Q here. 80 cents a piece and I think shipping was only 2 bucks. http://www.thecapking.com/35rx1000.html

Hey John, we don't know what a good number is unless we have a bad number to compare it to. Frying a $300 controller sucks though!! :( I have a Castle Ice sitting on the bottom of the Capitol pond from last year (Note to self: Screw down hatches on SAW boats). The FE community has a rich history of learning from the misfortunes of others. You are now an official member of the "I took one for the team" club. See you this weekend. Bring one of your cap banks so I can take a look at it.
Don gave me the caps you sent our way. Thanks! See ya Friday at the races. Took one for the team......... I like that. Makes me feel like it's not such a loss. Just paying the dues.
 
I will add a note since many of you are running extra caps, but limit the scope to aluminum electrolytic designs.

Many caps look alike and there are many clones of good caps like the Rubycon caps so be cautious.

You want to look for low or ultra low resistance caps with high or ultra high ripple. I have attached come data sheets of the Rubycon caps popularly found. Most high performance caps like these will have a 105C rating which is good. Rubycon has one series RX30's which have a 130C limit. The temp rating is important to consider. As the ripple currents are buffered this causes heating within the caps. A higher rating allows for better thermal performance.

I would shy away from generic or suspect brands if at all possible. You can find good caps from places like OSE or though electronics resellers like Allied Electric, Mouser, etc. These big online shops will have thousands of varieties so narrow the list by voltage range, temp rating and connection style. This should shorten the list to 20 or so.

TG

e_RX30.pdf

E_YXG.pdf

e_ZL.pdf

e_ZLG.pdf

e_ZLH.pdf

e_ZLJ.pdf

e_ZLK.pdf
 

Attachments

  • e_RX30.pdf
    147.9 KB · Views: 38
  • E_YXG.pdf
    229.2 KB · Views: 21
  • e_ZL.pdf
    223.3 KB · Views: 14
  • e_ZLG.pdf
    202.2 KB · Views: 14
  • e_ZLH.pdf
    217.6 KB · Views: 18
  • e_ZLJ.pdf
    192 KB · Views: 24
  • e_ZLK.pdf
    110.7 KB · Views: 68
Sooo. As much as I hate to eat crow, I am willing to admit when I was wrong. I didn't need cap banks last year when running Q power on my SAW boats. But stepping down the voltage to P- Power, the increased amp draw caused me to learn a valuable lesson the hard way this past weekend. The cheapo Fightercat 300 served me well but couldnt take the extra amps. Not a catastrophic failure but she got hot enough to melt solder causing a short on the board. i switched over to a Castle ice 240 and proptley blew a cap on the first pass. The second one is about to let loose. The board looks to be okay. I may wire a cap in her and do a quick test, then send her to Castle for repairs. I havent downloaded the data logger yet but will do so in the next day or two. I've got a dozen 1000uf 35v 130C Rubycon's sitting on the work bench. LOL
 
Back
Top