Regeneration Power from Motors back to battery - blocked by DPDT switch - need Diode rating

[marcusob]

Hi,

Ive tried to find what rating of power diode to wire in series with an emergency shut off button, I'm going to be using a DPDT as I have 2 batteries on 2 circuits - the 11 volt battery to the motors via a robo claw motor shield (2X30A) and a 7 volt battery for an Arduino and logic power on the roboclaw.

So I'll have 2 batteries via a DPDT switch to the robo claw motor input and lb input. To feed back any regen power when the motors are not being driven I need to have a battery connected to the robo claw - BUT the DPDT switch will stop a permanent connection (switch is on the +ve) so I need a power diode in series, what rating should it be ?

I've got a 11.1 Volt 3S - 24C rated Battery X 5000mah (5AH) = 125 Max Current Amp Draw

My motors have 12 volt 313 RPM Brush Motors The specs are here

Operating Specifications:
Operating Voltage Range: 6~12VDC
Rated Voltage: 12VDC
Rated Load: 4.5 kgf-cm (62.5 oz-in)
Max No-Load Current: 0.52A
No-Load Speed: 313 RPM
Min. Stall Torque: 30 kgf-cm (416.6 oz-in)
Max. Stall Current: 20A @ 12VDC
Dielectric Strength: 250 VAC
Motor Brush Type: Graphite
Output Power at Max. Efficiency: 13W

What diode do you recommend, and I just connect it in series with the switch on the +ve ?

Marcus

 

[crutschow]

Show us a diagram of where you want the diodes.
I don't see any need for a diode in series with the switch, but if the switch is connected in a bridge configuration to reverse the motor, than diodes are typically connected from each of the motor terminals back to V+ and ground (4 diodes total) to carry the inductive and regen current.
For that 1A diodes, such as the 1N400X series, should be fine.

 

[marcusob]

Hi, so this is about regeneration power - coming from the motors when they are slowing down, they act like generators/alternators and create power - this goes back to the motor shield, but because the +ve is not connected to a battery (as there is a switch in the way), the current has nowhere to go, and thus

damage to the motor shield - this is directly from the manufacturer of the shield. The diode I'm after is D1 - what type/rating etc ? Thanks.

 

[marcusob]

I have a 11.1 volt battery, these are my motor specs **broken link removed**
and here is the motor shield (its the 30A version) - **broken link removed**

Here is my battery http://traxxas.com/products/parts/batteries/idpowercellbatteries/lipo/2872X-5000mah-111v-3S-25C

And here is my web site !!

**broken link removed**

Thanks,

 

[spec]

Hy marcusob,

I have not looked at the technical aspects of your application, so I do not know if it will work or not, but to answer your question any of theses diodes will do: https://www.digikey.co.uk/product-s...t=0&page=1&quantity=0&ptm=0&fid=0&pageSize=25

 

[crutschow]

Okay, your diode is in parallel with the switch, not in series as you stated.
The 1N400X series diodes should work fine for that.
You could use a Schottky diode, as spec suggested, but it's not necessary.

 

[marcusob]

Hi crutshow, thanks for the reply, when I type 1N400X diode into Amazon i get lots of different types with different voltages and amp ratings,is there a particular volage/amp I need ?

 

[marcusob]

I presume this will do it -
1N4004 400V 1A

https://www.amazon.ca/Stkertools-1N...8&qid=1466662512&sr=8-2&keywords=1N400X+diode

Thanks.

 

[crutschow]

All the 1n400X diodes are rated at 1A, it's just their voltage rating that varies.
In your case the 50V 1n4001 will work as well as the 1n4004.

I just noticed that the diode in the schematic you posted will not do anything to protect the switch contacts (unless it is a zener diode).
The diode needs to go between the switch output and the negative pole of the battery to do that.
The idea is to provide a path for the inductive current that continues to try to flow when the switch is opened, and a diode from the switch output to the negative side of the battery will do that. Placing it across the contacts won't.

 

[spec]

Okay, your diode is in parallel with the switch, not in series as you stated.
The 1N400X series diodes should work fine for that.
You could use a Schottky diode, as spec suggested, but it's not necessary.

I agree that the 1N400x series of diodes may be OK- from the OPs specification you simply cannot tell- but the diodes that I recommend will be more efficient and more rugged. As there is only one diode involved and the cost difference is not significant in absolute terms, why mess about.

While this is completely unscientific, I simply do not like the 1N400x diodes for any critical application- they are fairly low conductance and they are made by many companies, some not so good. The early batches of 1N400x diodes were unreliable too, so if you get one of those, which are still drifting around on the market, look out. In summary the 1N400x is cheap and nasty.

The company that I worked for had to spend a fortune replacing the 1N400x diodes in all their products.

By the way, it would be unwise to buy any semiconductors from Amazon.

spec

 

[marcusob]

OK Spec, so get one off the page you recommended, a schottky one rated at 60 volt 20a

497-12297-NDDIODE SCHOTTKY 60V 20A I2PAK

That should be ok, so I'll put it between the switch (on the +ve) output, and the -ve pole of the battery. I persume I have to get it the right way around otherwise I'll short the battery. Do these things have the direction clearly printed on them ?

 

[spec]

OK Spec, so get one off the page you recommended, a schottky one rated at 60 volt 20a

497-12297-NDDIODE SCHOTTKY 60V 20A I2PAK

That should be ok, so I'll put it between the switch (on the +ve) output, and the -ve pole of the battery. I persume I have to get it the right way around otherwise I'll short the battery. Do these things have the direction clearly printed on them ?

Digikey part number 497-12297 is for manufacturer STM part number STPS20M60SR. Here is the data sheet which shows the anode and cathode connections: https://www.st.com/content/ccc/reso...df/jcr:content/translations/en.DM00034248.pdf

spec

 

[marcusob]

Actually thinking about it if I do put the diode from the switch output to the -ve battery terminal, when the switch is closed (making a normal cricuit), the current will run straight from the +ve on the battery to the -ve on the battery won't it ?I can see the diode prevents the current fom going from the -ve to the +ve (due to the direction of the diode), but will the diode also stop the current flowing from +ve to -ve on the battery because it is producing resistance and therefore the curent will flow towards the motor shield ?

 

[marcusob]

will it be like this then ?

 

[marcusob]

Wont this mean the current flows directly from +v to -ve once the switch is closed, ie the circuit is turned on ?

 

[spec]

Actually thinking about it if I do put the diode from the switch output to the -ve battery terminal, when the switch is closed (making a normal cricuit), the current will run straight from the +ve on the battery to the -ve on the battery won't it ?I can see the diode prevents the current fom going from the -ve to the +ve (due to the direction of the diode), but will the diode also stop the current flowing from +ve to -ve on the battery because it is producing resistance and therefore the curent will flow towards the motor shield ?

You are right. To achieve regeneration you do not need to use a diode, but the crunch comes when the motor is no longer spinning fast enough to produce a voltage higher than the battery voltage. At that point the battery will once again start driving the motor. Hence, you need a diode.

I only answered your question about the diode type and specifications. You did not ask if regeneration would be possible with a simple diode in your application. This will depend on the type of motors you are using. Afraid I suspect that in this case you will be out of luck. The way to confirm or demolish this theory is to spin the motors at the speed that you hope that regeneration will be achieved and measure the current flowing into the battery.

I don't know much about regeneration circuits, but I would imagine that high performance regeneration circuits, as used on automobiles, have voltage converters to increase the voltage from the motor when it is in the generator mode. That way you would feed energy back into the battery, regardless of motor (generator RPM) and, in automobiles for example, breaking would still be effective at low vehicle speeds. This would be with a typical permanent magnet motor. But other motors produce a high voltage when off load and spinning.

No doubt, some of the ETO automobile experts can amplify or correct this if what I say is not completely right.

spec

 

[spec]

Wont this mean the current flows directly from +v to -ve once the switch is closed, ie the circuit is turned on ?

That is the equivalent of a dead short across the battery- don't try that.

spec

 

[crutschow]

Wont this mean the current flows directly from +v to -ve once the switch is closed, ie the circuit is turned on ?

Yes, obviously.
Reverse the diode polarity.

 

[MaxHeadRoom78]

In a P.M. motor, the generation voltage is a function of rpm, if you need regulation through the rpm range you need a wound field and regulator, just like as an alternator functions.
BTW are those gear motors at 300 rpm?
Max.

 

[marcusob]

ok, going back to my original question it seems that Spec - you are talking about clamping / over voltage, I am not doing anything for that (its built into the motor shield already).

I want a way to get the regen energy off the motor shield. I dont want to generate this energy either, it's an unwanted bi-product that can cause a problem, so I am seeking a way to get rid of it. So MaxHeadRoom78 - I dont want an alternator with a wound field and regulator etc.

All I really wanted is the spec of a diode that would allow energy created when the electric motors of my robot are spinning and the circuit is open i.e. turned OFF.

As a result of the motors spinning there will be energy created, this has to go somewhere i.e. back to the battery, I wanted a way to allow this even when the circuit has been cut because the circuit is turned off.

The way to do this was a simple diode in parallel with the switch, not from the output of the switch to +ve. So for anyone reading this in the future - the correct circuit is the one at the top, i.e. a diode placed across the pins of the switch, thus allowing run-off in a single direction back to the battery.

The rating of the diode should be roughly the max current the robot could generate running downhill with everything turned off and the motors in free spin with no breaking (assume everything has failed i.e. all fuses are blown, shield is off etc.). In my case this is about 20A, the whole thing is running <12 volt, so a diode rating of below 12v and can handle 20 A or below should be good.

Thanks everyone. Here is the latest correspondence from the motor shield manufacturer in response to having the diode placed from switch out to -ve.

QUOTE FROM ROBOCLAW BELOW

"That would be a diode used to clamp an over voltage(If I understand your description). You would usually use a TVS diode for that(and we have them on some models of the Roboclaw just for that purpose).

What we want is a simple one way path back to the battery if the switch is turned off or the fuse blows while the motors are spinning and energy is moving into the battery positive terminal we dont want to prevent that(bad things happen if the regen energy cant go anywhere)."