Go Down

Topic: 12VDC (0.35A) Motor with 1N4001 Diode? (Read 8439 times) previous topic - next topic

Archibald

Does it make good sense to use a Schottky diode if you are driving a inductance directly from an Arduino?  Using a Schottky diode should ensure that the 'kick-back' current is taken by the diode, not by the protection diode within the chip.

raschemmel

#16
Dec 17, 2015, 07:56 am Last Edit: Dec 17, 2015, 08:41 am by raschemmel
Quote
Does it make good sense to use a Schottky diode if you are driving a inductance directly from an Arduino?  Using a Schottky diode should ensure that the 'kick-back' current is taken by the diode, not by the protection diode within the chip. 
That sounds like a good thing.
I did a diode switching test using this code:
Code: [Select]
void setup()
{
  // put your setup code here, to run once:
pinMode(9,OUTPUT);
}

void loop()
{
  // put your main code here, to run repeatedly:
digitalWrite(9,HIGH);
delayMicroseconds(6);
digitalWrite(9,LOW);
delayMicroseconds(294);


Test Pulse : 300 uS, 2% duty cycle
Vcc: 10V
Current : 0.150 A
Inductor : 100 uH rod inductor (choke)
Mosfet: FQP30N06
Filter cap from Drain to GND : 4 uF
diode UUT-1 : 1N4001
diode UUT-2 : STTH5128 (Fast Recovery diode) (body mass approx 4 times that of 1N4001
diode UUT-3 : 1.5KE350A (see datasheet link below) (body mass approx 10 times that of 1N4001

1.5KE350A

Observations: The 1N4001 had a perfectly flat response
Both of the other diodes had a rounded negative transition (rising edge was straight, first half of pulse was flat, then declined several volts during the second half of pulse .

1N4001 got hot too hot to touch and I could smell it overheating.
The other two diodes only got warm.
The 1.5KE350A ran the coolest of the three.
I can post scope shots if anyone is interested but didn't see any reason to.

NOTE:  I know the other two diodes aren't rectifier diodes like the 1N4001 but I didn't have any other rectifier diodes on hand . I'll have to order some.

larryd

@raschemmel
Please show the 1N4001 vs 1.5KE350A

1.5KE are great for SWC on inputs with floating supplies to earth ground.

No technical PMs.
If you are asked a question, please respond with an answer.
If you are asked for more information, please supply it.
If you need clarification, ask for help.

yendis

My comment that 1N4001 was not based on a detailed knowledge of this application, just a hunch that as a commonly used, general purpose device, it may be better to source a diode designed especially for this type of application.  Low voltage drop, as mentioned is probably the main issue here, which should increase efficiency.

I am using a 1N4000 series diode across a 4 amp dc motor in a pwm app.  It does get quite hot, and so I will be looking around for something more suitable.
Chris

raschemmel

#19
Dec 17, 2015, 02:50 pm Last Edit: Dec 17, 2015, 06:19 pm by raschemmel
@LarryD,
What is SWC ? (what does it stand for ?)




FYI,
This was a rush job because I only had 15 minutes before I had to leave for work so when I get home I can post some photos at a lower TIME/DIV so you can see the entire waveform. Also, I disconnected the smoothing cap (4uF) because it was masking the differences between the diodes.

The main difference is the 1N4001 is so hot you can't hold onto it for more than a second whereas the 1.5KE350A is cool enough to hold onto indefinitely.

larryd

#20
Dec 17, 2015, 06:58 pm Last Edit: Dec 17, 2015, 07:47 pm by LarryD
SWC
Surge Withstand Capability
No technical PMs.
If you are asked a question, please respond with an answer.
If you are asked for more information, please supply it.
If you need clarification, ask for help.

Wawa

@raschemmel
For a fair test, could you compare a 1Amp normal diode with a 1Amp schottky diode.
e.g 1N4004 vs. 1N5819

In my previous posts I was fighting the fact that it's a common misunderstanding that "slow" rectifier diodes turn on slow.
That slow diodes turn off slow might not a problem for a relay/solenoid (switching once), but if you do that with a high-ish PWM frequency as in your test, I expect a common diode to get hotter.
Leo..

raschemmel

#22
Dec 17, 2015, 08:55 pm Last Edit: Dec 17, 2015, 08:55 pm by raschemmel
Quote
SWC
Surge Withstand Capability
 
That's what they use them for at work. Our product runs on 240vac so the peak voltage is SQR(2)=339.411,  the 1.5KE350A is perfect for that voltage (350V-339V=10.588 V range between the peak operating voltage and the VBRNominal voltage for that diode.
datasheet

@Wawa,
I'll have to order the schottky unless I can find it at Fry's Electronics.

larryd

Quote
That's what they use them for at work.
;)
No technical PMs.
If you are asked a question, please respond with an answer.
If you are asked for more information, please supply it.
If you need clarification, ask for help.

Wawa

@Wawa,
I'll have to order the schottky unless I can find it at Fry's Electronics.
The 1N5817, 1N5818, 1N5819 are common 1A schottky diodes.
Much like the 1N4001 - 1N4007
Leo..

raschemmel

#25
Dec 18, 2015, 05:50 am Last Edit: Dec 18, 2015, 05:18 pm by raschemmel
I found this

(NTE is the Fry's brand of electronic components (made in China) . You need the NTE Cross Reference to find anything if all you have is an industry standard part number (like 1N5818) Once you have the NTE datasheet you can compare it with an industry standard datasheet
Look at the OnSemi datasheet , page -2, (Maximum Instantaneous Forward Voltage ) and then compare it to the specs on the NTE datasheet .

I order some 1n5817s from China (ebay). ($3.93 , free shipping)

raschemmel

#26
Dec 19, 2015, 03:58 am Last Edit: Dec 19, 2015, 04:22 am by raschemmel
Quote
@raschemmel
For a fair test, could you compare a 1Amp normal diode with a 1Amp schottky diode.
e.g 1N4004 vs. 1N5819
@Wawa,
I don't have either a 1N4004 or a 1N5819  (yet)

I was able to get a NTE 1n5711
1N5711
and a 1N5822


NTE 1N5822

VISHAY 1N5822

I don't know about you but I can't tell them apart.

TIME/DIV: 1uS
 Amplitude: 5V/DIV
Vcc: 10V
Inductor: 100 uH choke
Mosfet: FQP30N06

larryd

No technical PMs.
If you are asked a question, please respond with an answer.
If you are asked for more information, please supply it.
If you need clarification, ask for help.

Wawa

I expect the 1N4001 and 1N4004 to test the same.
The main difference is the reverse breakdown voltage, and you don't come even close to that.

The 1N5822 is a 3Amp schottky.  You should compare that to a normal 3A 1N5400 series diode.

https://en.wikipedia.org/wiki/1N4001_and_1N5400_series_diodes

I think diode temperature is the thing you're looking for.
Most, if not all, diodes are fast enough to take the start of the back-EMF current.
Only the diodes that turn off faster are "ready" when the mosfet switches on again.  
Leo..

raschemmel

Quote
Where are you measuring?
From GND to the drain of the mosfet.
Connections
Vcc => inductor-a
mosfet Drain => inductor-b
mosfet  source => GND
mosfet gate => arduino pin
UUT DIODE-a  => ACROSS THE INDUCTOR.

Did you want something different like from diode anode to cathode ?

Go Up