Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Heater fan low speed protection w/ LM2907/LM2917

Status
Not open for further replies.

jelliott

Member
Is there an easy way to use a frequency-to-voltage IC to switch off a relay when the tach output of a brushless DC fan falls below a certain threshold?

Based on Figure 20 in the LM2097/LM2917 datasheet, the answer would appear to be 'yes,' but it's not even apparent to me from the datasheet how much load current the chip can pull. Has anyone on here been through this before?

The fan's tach output is two pulses per rev, and I want to pull power from the load if its speed falls below roughly 1500 RPM.

Thanks!
 
hi jell,
One option could be a retriggerable Monostable, set the mono timing for ~330uSec [ as you have two pulses/rev]
E
 
Is there an easy way to use a frequency-to-voltage IC to switch off a relay when the tach output of a brushless DC fan falls below a certain threshold?

Based on Figure 20 in the LM2097/LM2917 datasheet, the answer would appear to be 'yes,' but it's not even apparent to me from the datasheet how much load current the chip can pull. Has anyone on here been through this before?

To help understand Figure 20, go to Figure 5. Note that the "load" in Fig 20 is attached to the collector of the output transistor in "Op-Amp section" in Fig 5. The table immediately above Fig 5 shows the current that the LM2917 can sink (about 50 mA). Then go to Figure 18, which gives more information on how much current the output transistor can sink.
upload_2013-11-15_4-30-52.png


Note, the output can also source current from the emitter of the output transistor (Pin 4 in Fig 20).

You should be able to find a relay that will operate with 50 mA or less.

John
 
Hmm, I really like the two alternate suggestions. The Jaycar kit is especially appealing because I'd love to avoid all the effort of designing and fabricating a custom PCB between now and the end of the year. But it's out of stock on their website--does anyone know of an alternate source?

A retriggerable monostable IC would also (besides providing the desired safety cutoff in the event of a fan failure) modulate the heater output if I intentionally ran the fan at a lower speed. What would that circuit look like? Would it be as simple as the LM2907 Figure 20 circuit? (And are there affordable MOSFETs out there that can switch 5 A at 48 V?)

Thanks!
 
Have you considered how you will restart the fan? The proposals so far will turn it off, so the tacho pulses will stop, so the circuit will keep it off.
 
If you go the retriggerable monostable route, you may want to revisit the suggested timing. ;) 330 uS between pulses = 660 uS (0.66 mS) per revolution or 1515 revolutions per second (approx. 90,000 rpm).

John
 
Have you considered how you will restart the fan? The proposals so far will turn it off, so the tacho pulses will stop, so the circuit will keep it off.

Morning alec,
If I am reading the OP's correctly.?
The fan's tach output is two pulses per rev, and I want to pull power from the load if its speed falls below roughly 1500 RPM

He wants to switch the power to a 'load off' and keep the fan running.
Eric
 
My bad. I thought the load was the fan :oops:
 
Morning alec,
If I am reading the OP's correctly.?


He wants to switch the power to a 'load off' and keep the fan running.
Eric

That is correct. I want to pull power from an electric heater element if the fan stops spinning for any reason.
 
A retriggerable monostable IC would also (besides providing the desired safety cutoff in the event of a fan failure) modulate the heater output if I intentionally ran the fan at a lower speed.
When the fan speed drops below a threshold do you want the heater (a) to switch off completely or (b) to operate under PWM such that the average heater current progressively drops as the speed drops further.
If 'a', then under what conditions should the heater be switched back on?
The fan's tach output is two pulses per rev
Does that pulse have a 50% duty cycle or some other figure?
 
When the fan speed drops below a threshold do you want the heater (a) to switch off completely or (b) to operate under PWM such that the average heater current progressively drops as the speed drops further.
If 'a', then under what conditions should the heater be switched back on?
Option A was my original plan (in which case the heater would be switched back on if fan speed recovered above the cutoff threshold), until I saw Eric's suggestion above and got the idea for Option B, which I'd probably prefer if it'd be just as easy to implement. (My day job has been consuming > 65 hr/wk lately, so I'm afraid I won't have time to design and fabricate a complex custom PCB between now and Christmas [when I intend to install this new heater in my brother's electric car], so I'd love to find a solution that's simple enough to be implemented by soldering a few resistors and capacitors to the legs of an IC and encapsulating it all in a blob of epoxy, without the need for a PCB.)
Does that pulse have a 50% duty cycle or some other figure?
Yes, 50%.
 
If I go with Option A, a la Figure 20 from the LM2907 datasheet, does it matter what the value of the capacitor on pin 3 is? It's not labeled in Figure 20.
 
Since your load is a heater, which will get hotter if the fan slows down or stops, why don't you use temperature to control the heater. It is much
simpler than monitoring speed.
 
If I go with Option A, a la Figure 20 from the LM2907 datasheet, does it matter what the value of the capacitor on pin 3 is? It's not labeled in Figure 20.
It is labeled in Fig 5 as C2. It is part of a low pass filter (integrator). If C2 is bigger, then ripple is less, and response time is increased. Decreased ripple makes it more sensitive to smaller changes in RPM at the cost of a slower response. C2 is discussed more fully on page 8 (Eq. 3).

The datasheets example circuits often have a value of 0.1 to 1.0 µF for C2.

John
 
Since your load is a heater, which will get hotter if the fan slows down or stops, why don't you use temperature to control the heater. It is much
simpler than monitoring speed.
It's hard to imagine anything much simpler than Figure 20 from the LM2907 datasheet, but that's an interesting idea (I'll call it Option C)--I know that's how household electric space heaters are protected. But for my application (automotive cabin heater) I'm afraid it would take a bit of work to get right--normal vs. abnormal temperatures at the heating elements will be different in the dead of winter with the system set to draw outside air vs. on an autumn day with recirculation selected. I also feel like speed-monitoring options are more fail-safe. I can imagine all kinds of things going wrong with a thermistor circuit or with the installation of the thermistor, that would give you a latent failure that would go undetected until it's too late. Whereas the likely failure modes of Options A and B would result in an inoperable heater--inconvenient but not hazardous. The other thing that's making me lean towards Option A is that if the battery voltage falls off (be it from climbing a hill with with heavy passengers in the car, or from a dangerously-low battery), the fan will slow down and the heating element will shut off, shedding 5 A worth of load from the struggling batteries (and in the transient case, it'll come right back on after the car crests the hill and battery voltage recovers a bit).
 
out of interest what car is it?
 
hi jell,
One option could be a retriggerable Monostable, set the mono timing for ~330uSec [ as you have two pulses/rev]
E
Where should I start if I want to go with this option? I was just about to sit down and prototype my original LM2907 idea on the breadboard, when my brother (for whom this heater is to be a Christmas gift) called to say that he'd prefer the retriggerable monostable PWM option so that different fan speeds give different power levels from the heater. But I've never designed any kind of PWM circuit before.
 
It's a US Electricar 953A, a.k.a. "Lectric Leopard," i.e. a Renault 5 converted to an electric drivetrain by US Electricar Corporation of Athol, Massachusetts.
interesting concept! i can see why you want to eek every last drop of power from the batteries, cant be easy packing enough into a Renault 5. i dont suppose a micro is an option for you?
no idea how you would go about doing it analogue but your in good hands on here ;).
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top