Flyback diode - - - NEEDED ? ? ? Or RECOMMENDED ? ? ?

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,853
I've taken some stock I have and am bodging together a blower for either a fireplace or for an out-door fire pit. The sole purpose is to force air into the firebox to get a better / hotter burn. Not looking for anything serious in the way of temperatures, just want to reduce the amount of smoke. So from stock, I've taken an old light dimmer from a 1989 Toyota Celica. It functions as a PWM. The "Sink" terminal is either ground or open. The circuit below shows a 10KΩ resistor to pull up an N channel MOSFET (also from stock); a 2SK1277 out of a pool spa controller board. It's rated at 250 V, which is enough for what I'm doing. I intend to vary the speed of the blower so that I can control the oxygen without blowing ashes out of the pit or blow the flame out.

My question concerns whether I need to include a flyback diode or not. I certainly can put one in, but why do so if one is not needed? So I'm turning to you experts in such matters. As far as the blower motor goes - it comes out of a BMW automobile, and I have no idea whether it's a permanent magnet type or series wound. I suppose I could check, but it's out in the garage and I'm sequestered in the basement tinkering with junk.

Variable Blower.png
 
Last edited:

dendad

Joined Feb 20, 2016
4,451
A flyback diode will give smoother speed control at the low end as well as clamping the high voltage spike.
Try running the motor at low speed and place the diode on while it is going to see if it makes a difference.
The energy in the collapsing field is used to power the motor instead of being wasted.
 

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,853
@dendad
Thanks DD. I'll give that a whirl.

So far I've only run small motors like CD motors, not wanting to damage anything. This evening when I have some time I'll take my test rig outside and do all the hookups just to see if there's a difference. Would you recommend 1N4007? Something better?
 

dendad

Joined Feb 20, 2016
4,451
A diode with the current raring of the motor is a good rule of thumb. So I thing a 1N400x will be a bit small as it is only 1A. Schottky diodes are often recommended for faster switching, but a "normal" diode will work too. Do you have any bigger diodes?
 

crutschow

Joined Mar 14, 2008
34,285
whether I need to include a flyback diode or not.
The idea for using PWM with a motor is to generate a DC current through the motor that is adjustable by rapidly turning the power full-on and full off, with the average current roughly equal to the maximum current times the duty-cycle.
So when the switch turns off, you want a path for the current that is flowing to keep flowing due to the motor inductance.
This is provided by the diode.
Without the diode you lose efficiency, and large voltage spikes can be generated across the switch.
 

dl324

Joined Mar 30, 2015
16,846
Don't know what the motor current is, but I wouldn't dismiss 1N4001 without more info. Those diodes will withstand 30A surges.
upload_2019-5-27_15-31-4.png

A higher reverse voltage rating than 50V would be overkill.
 

crutschow

Joined Mar 14, 2008
34,285
Don't know what the motor current is, but I wouldn't dismiss 1N4001
The motor is stated to be 6A on his schematic, and the diode will likely have to carry a good percentage of the average motor current during the off time of the switch.
So I would recommend at least a 5A diode (Schottky for better efficiency) for reasonable margin.
 

dendad

Joined Feb 20, 2016
4,451
The motor is stated to be 6A on his schematic, and the diode will likely have to carry a good percentage of the average motor current during the off time of the switch.
So I would recommend at least a 5A diode (Schottky for better efficiency) for reasonable margin.
Yes indeed!
For the price of a diode, it is not worth skimping. The current through the diode will be quite significant when the motor is under load or stalled.
In this case, bigger is better!
 

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,853
Diodes are not something I've had a lot of experience with. Sure, I've used them to rectify AC, but - - - . I'll have to look through my stock or maybe use a diode out of a wall wart, the kind that is a box connected to AC via a cord, not hanging directly on the wall. I have several and have noticed some have large diodes in them. Three legged devices with the capability of being bolted to a heat sink. If I can't find something substantial I'll have to shop for something that will do the job.

I appreciate the advice. Keep in mind the FET I'm using is quite robust. I have several of them, and they have fast recovery diodes internally. Does anyone think I could use one of those just for the diode part of it? Not spending any money and I don't know when the next time will be that I'll want to use one of these 2SK1277's.
 

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,853
The current through the diode will be quite significant when the motor is under load or stalled.
Simple air pump is all it is. Centrifugal pump type. Block the air flow and the motor "Should" spin faster as it's not going to be doing any work. I don't expect that I will ever be stalling the motor, not unless something goes wrong with it. It comes out of a car, so it should be designed to last a good ten years under normal use. Since the car it came out of is a 2017, the motor should have plenty of life in it. Add to that, I won't be using it but on rare occasions, and / or in cases of emergency, heating in the winter. I've been considering a fire-box with forced air to draw heat out into the room. Should there be a power failure during winter - very rare occurrence - I can still provide plenty of heat for the house.

But the main purpose of the pump is to give air to a starving fire, be it in a fire pit or in the fireplace. Sometimes I like to do something just because it can be done. Doesn't mean it's a great idea. Sometimes I come up with some pretty dumb ideas. Like putting rechargeable batteries in my TV remote then installing one of those wireless charger kits so I can keep the remote charged just by laying it on the charging station. Like I said, not always a "Good" idea, sometimes it's just an idea and I want to see if I can tackle the task. And, no, I haven't modified any TV remotes.

One idea I was messing with was to use some UV LED's in a flashlight configuration then powering it from a TV remote. Point it across the parking lot (motel) and mess with the TV's in other people's rooms. Never happened. Likely will never happen. But I'm always thinking. And occasionally I'm doing some "Good" thinking.
 

Kjeldgaard

Joined Apr 7, 2016
476
I appreciate the advice. Keep in mind the FET I'm using is quite robust. I have several of them, and they have fast recovery diodes internally. Does anyone think I could use one of those just for the diode part of it? Not spending any money and I don't know when the next time will be that I'll want to use one of these 2SK1277's.
I would to say yes to using a similar Power MOS as a free-wheel diode, just remember to tether the gate to the source.
 

ElectricSpidey

Joined Dec 2, 2017
2,758
I know your question was about flyback diodes, but I was surprised you chose a 10k pull-up, don’t you think the turn on might be a little slow, with a FET like that?
 

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,853
@electricspide: You're talking to the worlds foremost expert on MOSFETS. NOT! I chose 10K as a starting point to test whether the circuit was viable or not. This is only to regulate speed of the motor. For one thing, 10K just happened to be laying on my workbench from a different project. I have a pretty good stock of standard resistors, so selecting a better suited resistor is possible. Second thing, the PWM module (from the car) only provides sink, no source. So the pull-up is there to turn the FET on. At 10K with a 12 volt power source (or car battery) the current will be 1200µA; so that will not affect battery life, nor does it put the module in danger. It's just being used to switch the FET on and off. And preliminary tests with tiny motors prove it works.

While I have everyone's attention, I've found some diodes in a blown UPS. A MUR460, rated at average current of 4 amps. The Nonrepetitive Peak Surge Current (Surge applied at rated load conditions, half wave, single phase, 60 Hz) is rated at 110 amps, See here for the data sheet. Keep in mind, I'm probably going to be running this motor a lot slower than originally intended for automotive uses. Though it's a blower it won't be used as a blower, more as a gentle airflow into the firebox to keep the embers hot and burning. Too much airflow will blow ashes all over the place as well as potentially spreading hot ashes outside the firebox - a condition to be avoided. [edit] Question concerning this diode: Will it be sufficient at full power? [end edit]

If used for a different purpose such as extracting heat from the firebox I may use a higher airflow. Possibly even full power, but that's something that may not happen until I've experienced a power failure during winter. Which again is a rare occurrence.
 
Last edited:

Thread Starter

Tonyr1084

Joined Sep 24, 2015
7,853
In need of a reality check. Does this look right? Looks right to me, but again, I'm not the expert here.

The 640 came out of a failed UPS. The diode check shows 0.495 one way and open the other. So it looks like that portion is good. I'm tying the gate to ground and not back to Source because source will be at zero volts (respectively) during the PWM High but left floating during the PWM Low. Right? Wrong? Why so (or not so)?

Variable Blower.png
 

crutschow

Joined Mar 14, 2008
34,285
I'm tying the gate to ground and not back to Source because source will be at zero volts (respectively) during the PWM High but left floating during the PWM Low.
That will work okay, but it has no advantage over just connecting the gate to the source.
The MOSFET stays turned off for both conditions, since connecting the gate to ground just generates a minus Vgs voltage when the switch is off, which is (further) OFF for an N-MOSFET.

It has the disadvantage that any power supply spikes will appear across the gate junction, which usually has a 20V max rating, thus could zap the gate.
 

ebeowulf17

Joined Aug 12, 2014
3,307
@electricspide: You're talking to the worlds foremost expert on MOSFETS. NOT! I chose 10K as a starting point to test whether the circuit was viable or not. This is only to regulate speed of the motor. For one thing, 10K just happened to be laying on my workbench from a different project. I have a pretty good stock of standard resistors, so selecting a better suited resistor is possible. Second thing, the PWM module (from the car) only provides sink, no source. So the pull-up is there to turn the FET on. At 10K with a 12 volt power source (or car battery) the current will be 1200µA; so that will not affect battery life, nor does it put the module in danger. It's just being used to switch the FET on and off. And preliminary tests with tiny motors prove it works.

While I have everyone's attention, I've found some diodes in a blown UPS. A MUR460, rated at average current of 4 amps. The Nonrepetitive Peak Surge Current (Surge applied at rated load conditions, half wave, single phase, 60 Hz) is rated at 110 amps, See here for the data sheet. Keep in mind, I'm probably going to be running this motor a lot slower than originally intended for automotive uses. Though it's a blower it won't be used as a blower, more as a gentle airflow into the firebox to keep the embers hot and burning. Too much airflow will blow ashes all over the place as well as potentially spreading hot ashes outside the firebox - a condition to be avoided. [edit] Question concerning this diode: Will it be sufficient at full power? [end edit]

If used for a different purpose such as extracting heat from the firebox I may use a higher airflow. Possibly even full power, but that's something that may not happen until I've experienced a power failure during winter. Which again is a rare occurrence.
You mentioned the fan speed/power setting several times there in relation to diode longevity. I don't think it makes any difference at all to the diode whether you're running the fan at 1%, 10%, 90%, or 99%.

The determining factors for how much power the diode will dissipate are peak motor current and PWM frequency. The peak motor current will be essentially the same across most settings (you're using PWM to alter the average current, but not changing the peak current.) Most PWM controllers I'm familiar with maintain a constant frequency and only alter duty cycle. So basically, unless you reach completely on or completely off, any other setting is the same as far as the diode is concerned.
 
Top