karimwassef
Active member
Have we given up on modifying the eBay controller?
Use PWM to control your Jebao DCT pump using brushless motor controller from ebay
- Thread starter d0ughb0y
- Start date
d0ughb0y
Active member
I did a bit more google search, and pretty much the standard bemf circuit is what I did in the last mod, which is removing 4 resistors, and replacing 3 capacitors with 10k resistors. So I think the problem is not in the bemf circuit. I'll try other mods when I get some new ideas.
I'm starting on a parts list for my diy controller. I'll use a plain Arduino (either pro mini or nano v3), so the additional circuit will just be the mosfet array and the bemf circuit.
I'll try using the beefy TO-220 mosfets, so I will only have one design/board that will work for all pump models. I sort of like the way the green ebay board layout of the mosfet. you can just slap on a heat sink if needed.
I'm starting on a parts list for my diy controller. I'll use a plain Arduino (either pro mini or nano v3), so the additional circuit will just be the mosfet array and the bemf circuit.
I'll try using the beefy TO-220 mosfets, so I will only have one design/board that will work for all pump models. I sort of like the way the green ebay board layout of the mosfet. you can just slap on a heat sink if needed.
karimwassef
Active member
Ok. Since you're working digitally, can you please also consider a modular approach with one master controller and multiple slave power boards? I have 5 of these pumps and my plans go up to 12. Most people will likely need at least two?
Just saying that this is the time to put "expansion-capable" into the spec.
And I'm assuming these run off 0-5V external signals? Can they run straight off Apex 0-10V? Or have voltage setpoint variables for user calibration to the external controller source.
Just saying that this is the time to put "expansion-capable" into the spec.
And I'm assuming these run off 0-5V external signals? Can they run straight off Apex 0-10V? Or have voltage setpoint variables for user calibration to the external controller source.
Last edited:
karimwassef
Active member
theatrus
100-mile-commuter
I'm going off the data sheet, but its effectively the same:
http://www.allegromicro.com/~/media/Files/Datasheets/A4960-A-Automotive-Datasheet.ashx?la=en
I may add a programmable VREF signal (DAC driven option) - the driver IC allows some ratio divider internally for VREF but for maximum flexibility for the first shot, a DAC is cheap. It can be jumpered out later if its not needed.
http://www.allegromicro.com/~/media/Files/Datasheets/A4960-A-Automotive-Datasheet.ashx?la=en
I may add a programmable VREF signal (DAC driven option) - the driver IC allows some ratio divider internally for VREF but for maximum flexibility for the first shot, a DAC is cheap. It can be jumpered out later if its not needed.
please add 0-5v or 0-10v analog input if possible so I can use this with my PLC. Outputting PWM signals is very expesnive and difficult with my PLC.
Anyway, I did some testing today here is what I found (this is with a DCS 7000 pump through 3/4" pipe, flowmeter, 3 90 degree fittings and about 3 feet of pipe).
Ebay controller maxed: 170Hz/34.4LPM/9.4GPM/561GPH
Stock controller maxed: 210Hz/43.8LPM/11.6GPM/694GPH
So it appears that the stock controller has approximately 20% more output. I think we all figure that part out but nice to know we have actually numbers now.
Anyway, I did some testing today here is what I found (this is with a DCS 7000 pump through 3/4" pipe, flowmeter, 3 90 degree fittings and about 3 feet of pipe).
Ebay controller maxed: 170Hz/34.4LPM/9.4GPM/561GPH
Stock controller maxed: 210Hz/43.8LPM/11.6GPM/694GPH
So it appears that the stock controller has approximately 20% more output. I think we all figure that part out but nice to know we have actually numbers now.
Than would be great. Thank you so much. I'm actually an Electrical engineer but I deal with power in commercial buildings not really electronics. But I understand everything just not great at designing circuits and using ICs.
yea I'm not sure how accurate my flow meter is. I wonder what the head pressure is for the meter because it seems pretty restrictive if its accurate. This pump is rated for 1900GPH at no head and almost 900 at 16 feet of head and I'm not close to that at all. Maybe Ill take a dremel to the inside of it and port it out.
yea I'm not sure how accurate my flow meter is. I wonder what the head pressure is for the meter because it seems pretty restrictive if its accurate. This pump is rated for 1900GPH at no head and almost 900 at 16 feet of head and I'm not close to that at all. Maybe Ill take a dremel to the inside of it and port it out.
karimwassef
Active member
I'll run the same for my DCT15000s with my 2" flow sensor. Just need to download the arduino counter code .
FullBoreReefer
New member
Just let me know when someone gets a decent reliable kit together that I can purchase haha. I am totally lost on all this electronics talk or I would try and help lol. But I would like this control for my dct pumps.
Keep up the good work!
Keep up the good work!
karimwassef
Active member
So... On that topic... Theatrus and doughboy, are you collaborating so we can end up with one standard solution, or trying two designs?
theatrus
100-mile-commuter
Right now it's two designs. I'm also trying to fit inside of the Jebao enclosure and reuse some parts (like the heatsink and connectors).
Messy power layout.
karimwassef
Active member
Only one image come up
theatrus
100-mile-commuter
Weird. Thanks Tapatalk 
I usually host the images out of my blog system but sometimes post from the phone.
Anyway, shuffled some things around. Realized I can have the heatsink off the edge of the board for the 150W (enough room in the case) which will allow the 50W version to cover all of the backside of the FETs. I tried some other orientations, like rotating the FET array by 90 degrees, which makes the U/V/W taps direct to the pads, but couldn't quite make it work.
Most of the motor-drive is in place, I just need to calculate the sense resistor power drop to see if I need to upsize it (the A4960 includes a handy 10x amp internally).
I usually host the images out of my blog system but sometimes post from the phone.Anyway, shuffled some things around. Realized I can have the heatsink off the edge of the board for the 150W (enough room in the case) which will allow the 50W version to cover all of the backside of the FETs. I tried some other orientations, like rotating the FET array by 90 degrees, which makes the U/V/W taps direct to the pads, but couldn't quite make it work.
Most of the motor-drive is in place, I just need to calculate the sense resistor power drop to see if I need to upsize it (the A4960 includes a handy 10x amp internally).
theatrus
100-mile-commuter
They're in the same place (and same footprint, though need to confirm height) as the Jebao buttons. Exact function TBD
d0ughb0y
Active member
so I opened up the cp-40 controller, and the bemf circuit is a simplified version of the ones on the dct pump controller (1 less resistor and capacitor). So the mod on the green ebay board makes the bemf circuit identical to the jebao controller. So I think that's as far as any mod can go on the green ebay board, since the rest of the functions are in the motor controller chip.
Another difference is the cp-40 controller uses a shift register to drive the leds. It has a lot more leds than the dc pump controller. The controller chip is the same as the dc pump controller, I would not be surprised if it is some version of AVR mcu. The cp-40 also use the same mosfet as the 50watt pump controller, except it does not come with a heatsink. I don't think 50watt needs a heatsink, the mosfets are still cool to the touch while running at max speed.
I will probably skip the LM339 and connect the bemf circuit directly to the Arduino. I can use the built in analog comparator interrupt to trigger commutation change. It only requires one additional pin use (4 instead of 3) as the reference is now used as comparator positive input.
I'll trace the motor driver circuit on cp-40 controller tonight and will probably use the same circuit. I think the reason you can only use 24v is the mosfet max Vgs is 20v so the pre driver must not use >20v. I think it uses 18v.
Apparently, the reason the ebay sellers dropped the price of crossflow is jebao is coming out with a new version. No further info on what changed.
Another difference is the cp-40 controller uses a shift register to drive the leds. It has a lot more leds than the dc pump controller. The controller chip is the same as the dc pump controller, I would not be surprised if it is some version of AVR mcu. The cp-40 also use the same mosfet as the 50watt pump controller, except it does not come with a heatsink. I don't think 50watt needs a heatsink, the mosfets are still cool to the touch while running at max speed.
I will probably skip the LM339 and connect the bemf circuit directly to the Arduino. I can use the built in analog comparator interrupt to trigger commutation change. It only requires one additional pin use (4 instead of 3) as the reference is now used as comparator positive input.
I'll trace the motor driver circuit on cp-40 controller tonight and will probably use the same circuit. I think the reason you can only use 24v is the mosfet max Vgs is 20v so the pre driver must not use >20v. I think it uses 18v.
Apparently, the reason the ebay sellers dropped the price of crossflow is jebao is coming out with a new version. No further info on what changed.
Similar threads
