# Stepping down the AC frequency to 30 Hz

Discussion in 'General Electronics Discussion' started by Glee, Mar 15, 2011.

1. ### Glee

9
0
Mar 15, 2011
Hi all,

New to the forums. I'm trying to step the frequency of wall power down to 30 Hz to run an electromagnet. What is the easiest way to do this? Is there a simple way to switch AC at 30 Hz, while syncing with the sine waves to prevent instantaneous voltage changes? Or am I completely off the mark?

2. ### davennModerator

13,865
1,956
Sep 5, 2009
I 'm wondering why you think you need to 1/2 the mains freq. and why you think it would help an electro magnet work better ? have you got a reference where this was suggested ?
why dont you just use DC ?

Dave

3. ### Glee

9
0
Mar 15, 2011
The electromagnet is vibrating, and I would like to have it vibrate at 30Hz rather than 60Hz. Right now it is working fine with a half wave rectifier, only it's at 60Hz instead of 30Hz. Thanks!

Last edited: Mar 15, 2011
4. ### (*steve*)¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥdModerator

25,497
2,839
Jan 21, 2010
There is no easy way. Only hard ways.

You would have to convert the power to DC then make an oscillator which switches the voltage 30 times per second.

Or maybe you could deliver every alternate half cycle at 60Hz to the device via a triac and something that detects zero crossings and fires on each 4'th crossing.

5. ### Glee

9
0
Mar 15, 2011
Thanks for the reply. I was afraid of that. I don't have any experience with transformers or triacs. How would I go about making an oscillator that can handle significant power? Do you make a power oscillator or can you amplify a smaller signal?

Alternatively, what kind of circuit would detect every 4th zero crossing?

Greg

6. ### (*steve*)¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥdModerator

25,497
2,839
Jan 21, 2010
Why do you want something vibrating at 30Hz?

What power does the electromagnet require?

Will it still perform it's job if it gets less power?

7. ### Glee

9
0
Mar 15, 2011
The 30 Hz is a design requirement and I'm trying to deliver somewhere between 20-100 Watts. I'm not sure where the sweet spot is yet, but if I had to guess, it would be on the lower end of that range, maybe 30 Watts. Anything less than 20 Watts wouldn't be very useful. At what wattage does this become significantly harder?

Thanks

8. ### Resqueline

2,848
2
Jul 31, 2009
Is the design requirement approximately or exactly 30Hz, and/(or not neccessarily) in sync with the mains? Does it need to be a sine or can it be a square wave?
Apart from (the) other possibilities I tend to think about a signal generator and a PA amplifier for that kind of job. What resistance/impedance/inductance is the solenoid?

9. ### Glee

9
0
Mar 15, 2011
It can definitely be approximately 30Hz, at least +-2. I think it needs to be a sine wave, so that the electromagnet doesn't see instantaneous voltage changes, but I'm open to suggestions on that if you think I can get away with driving it with a square wave and some diodes.

Your PA amplifier idea sounds intriguing, what are the considerations with how that will interact with the electromagnet? I might go buy one and play around with it.

Thanks!

10. ### Resqueline

2,848
2
Jul 31, 2009
An ordinary electromagnet pulls in regardless of polarity, so driven with 60Hz AC it will buzz with 120Hz. With a diode in series it'll buzz with 60Hz, as you've seen.
Driving it with a simple single-ended power supply, pulse generator, & transistor is an easy way in that frequency in = frequency out, but you may get harmonics.
Driving it with an amplifier would require the input waveform to be modified (into a heavy cross-over distortion like shape) to reproduce a sine shaped ouput force.
If the electromagnet had a magnetic plunger however it would work like a speaker, being pushed out at "positive" voltages and being pulled in at "negative" voltages.

But all of this requires a certain (frequency dependent) voltage. Do you have a multimeter? Do you have the solenoid like you want it? Can you do some measurements?

11. ### Glee

9
0
Mar 15, 2011
I have the solenoid and a multimeter, but I don't have a signal generator. I can definitely do some measurements. What should I do? The multimeter is saying that the resistance of the coil is 1.2M Ohms, which seems way too high. Thanks,

Greg

12. ### Resqueline

2,848
2
Jul 31, 2009
Yeah, if it really was 1.2M then you wouldn't get any action out of it at all. There might be something wrong with the multimeter (fuse?). Try it out on other things.
I'm looking to know the DC resistance, and the current it draws at mains frequency. The size & weight of the thing would also help estimate the power it can take.

13. ### Glee

9
0
Mar 15, 2011
I think the diode may have been throwing off the resistance reading. I hooked it up to 5V, and got a resistance value of ~250 Ohms. The thing is about 50 grams, and it is made of litz wire wrapped around a rectangle that is 1"X.5." It seems to be drawing about 110mA.

14. ### Resqueline

2,848
2
Jul 31, 2009
Unless the solenoid was picking up hum/RF and rectifying it I don't see how the diode could throw off the measurement.
I'm also surprised about the use of Litz wire in a solenoid, it's expensive and usually only used where low loss is paramount - in high-frequency coils.
There's usually three measurements for three-dimensional objects, what's the third?
And, to be absolutely clear: at 5V DC you measured a current of 5/250=20mA? And at what mains voltage (117V & 60Hz?) did you measure 110mA AC?

15. ### Glee

9
0
Mar 15, 2011
Sorry, it's .75" tall, so it's 1"X.5"X.75", and weighs 51 grams.

Yes I measured 20 mA at 5V to arrived at 250 Ohms. I measured 110mA at US mains, 120V, 60 Hz. I didn't check to make sure it was at the right voltage.

I believe it is made of Litz wire, but I've never used it before so I could be mistaken. It's very thin and red, and you have to burn off the insulation to strip it to the copper core.

16. ### davennModerator

13,865
1,956
Sep 5, 2009
litz wire if multi-stranded. If what you have is just single stranded then its normal enamelled wire

you mentioned in you 2nd post a diode, where's the diode in relation to the coil connections ?
is it across the coil ? then for accurate resistance measurement of the coil just lift one leg of the diode

Dave

17. ### davennModerator

13,865
1,956
Sep 5, 2009
if the diode is in parallel with the coil then you are going to get a summed parallel resistance of the coil and diode, particularly when meter probes are connected one particular way around

Dave

18. ### Glee

9
0
Mar 15, 2011
I believe that the diode is in series with the magnet. If it was in parallel, wouldn't it short out the bottom half of the AC waveform, as opposed to just block it if it's in series? Also, for the multimeter resistance reading, it reads 1.2 M Ohms one way and no reading if the leads are reversed.

The coil is made of just enamelled wire then, the wire is not multi-stranded.

19. ### Resqueline

2,848
2
Jul 31, 2009
Ah, ok, so the diode is pemanently incorporated, so you're unable to measure on only the coil itself, and you have to include the diode in the measurements?