Page 9 of 11

PostPosted: Mon Apr 01, 2013 3:47 pm
by gary
aussie_bloke wrote:
Answering Gary's question yep that motor is from a RC car, I thought with a heavier spindle a faster motor would be better to drive it. It has a resistance of about 4.5 ohms so it would draw about 2.67A when run from 12V going by the I=V/R formula.

Yes well that pretty much explains everything.

In truth there is no reason you can't use the motor as long as you design for it's characteristics - high starting current and high speed. As Steve mentioned these are normally driven from PWM circuits (normally referred to as ESCs "Electronic Speed Control"). You can buy these quite cheaply now but I don't have enough knowledge to advise you on it as yet. I actually took advantage of the high ozzie dollar and free shipping to buy a brushless motor and a corresponding ESC with the though that they might be adaptable to NBTV as the brushless motors are effectively synchronous motors which are ideal for NBTV use - alas I think I am about a year away before I can get back into NBTV seriously.

At the very least the high speed of this motor does imply that it needs a reduction system.

If you have a flat belt (like you often find in a cassette or cartridge player) you could try it just running on the motor spindle itself. A little bit of rubber tubing on the spindle would be even better to prevent slippage. In addition you can use the pulleys you have that don't fit by either shimming them out (if too large) or drilling them out (if too small) - the same applies to the other pulley.

Having said all that MY recommendation would be to use a cassette motor (with the internal speed control disabled) as a direct or belted drive - they have been used on numerous occasions just for this purpose (using a record as the disk). Belted is best for additional torque but that can lead to slippage too if you are not careful.

Good luck.

PostPosted: Mon Apr 01, 2013 5:31 pm
by gary
Steve Anderson wrote:Measuring a DC brushed motors resistance has little bearing on its running current consumption as that is mostly defined by the inductance and the switching frequency which is determined by the commutation speed and load.

Steve A.

Sorry Steve, whilst you are correct that induction and speed (in other words back emf) has an effect on current consumption you must take into consideration that a PM motor is linear over it's torque and speed range and is given by:

I = (Vsupply - back emf) / Rarmature

where back emf is linearly proportional to speed (a constant x speed)

Thus Rarmature gives us our maximum current at start up (enough to make transistors etc pop perhaps) and will also give us an idea if the current will be high(ish) under load (windage). Indeed as the back emf is always less than Vsupply even at no load the current is at least inversely proportional to Rarmature.

PostPosted: Mon Apr 01, 2013 8:03 pm
by Steve Anderson
Gary, absolutely correct, I was perhaps oversimplifying things. Also in a brushed DC motor you'll notice the measured resistance will vary as you rotate the armature as the brushes bridge one or more commutator sections.

Sure the DC resistance can give you a 'worst case' current but normally (once started) this drops to a value which is not attributable to the DC resistance alone. (unless the motor is stalled, which is in effect what we have here). Once the magnetic core reaches saturation it cannot generate any back EMF.

With the measurements made in this thread how was a 10A fuse blown with the 4.5 Ohms measured at 12V? Measuring low resistances with your average multimeter is tricky. A 10A fuse would take quite a time to blow even at 15A, review their datasheets.

The sample .pdf shows at rated current they take four hours to blow. One hour at 35% over-current and five seconds at double rated current...and these are 'fast acting'.

Steve A.

PostPosted: Mon Apr 01, 2013 9:56 pm
by gary
Grrrr, I have been guilty here of not following the thread entirely, i.e. skip reading. I missed the fact that it had blown a 10 amp fuse, but on top of that I thought the exercise was about trying to determine the characteristics of an unknown motor and was congratulating myself on determining it was an RC motor and had a high(ish) current rating.

In fact Troy has told us that it was out of an RC car but it was so long ago I had forgotten, one of the problems of getting old...

Still, I see no reason why the technique of measuring the armature resistance should not work - I have done it hundreds of times (as a collector of DC motors) without a problem. I admit I have never tried it on an RC car motor but I have done it on scooter motors!

The current drawn by a PM DC motor should never exceed it's stalled value and that is Vsupply/Rarmature.

Troy if you haven't moved off this altogether you might want to re-measure the resistance making sure you turn the spindle a little to make sure the brush is totally across the winding - note the lowest value you read.

It does seem, given the blowing of the fuse, and also looking at the thickness of the windings that the resistance should be lower than that by about a factor of 10.

EDIT: BTW how do you know the motor was out of a Kyosho car? It's just that some of them go for big bucks as replacement units. Also the type of car might gives us the motor specs too.

PostPosted: Thu Apr 11, 2013 7:02 pm
by aussie_bloke
Ok I have gone to Jaycar and bought myself a plastic motor pulley set, got some nice small size pulleys for the motor along with the smaller spindle pieces to make it fit on the motor, perfect! :D

Repositioning the motor so the belt fits nice and snug around its pulley and the spindle pulley (sewing machine rotary knob) I powered it up, the motor runs so much better now drawing less current! Just with the 3 ohm 20W resistor bank and motor connected to the 13.8V DC power supply, I am now getting about 3.4A driving the motor while connected to the spindle via the belt instead of 4.1A which is a good improvement.

And then connecting the coarse speed controller and adjusting the motor speed to a slower speed that looks around maybe 750-1000RPM, the current drawn goes down to about 1.9A which is so much better and at least to my opinion acceptable! :D The voltage drop across the motor is about 0.9V.

Now I've been curious about the low voltage drop across the motor as Steve stated it not being right with the motor specs being 12V and it still running. I am now thinking maybe this motor isn't exactly a 12V motor as there were no specs stamped on the motor itself and I by force of bad habit "assumed" it to run off 12V. I understand a lot of RC car battery packs supply 7.2V as well, so this motor would of probably typically ran off 7.2V in a car at a RPM speed high in I guess the 5000 to 10000RPM range. Giving this factor and the fact the motor is voltage speed controlled, the motor at a very low voltage but with the same amount of current flowing through it and it running at a proportionally slower speed in accordance to the voltage level, surely there wouldn't be anything wrong with it??? Anyhow the motor seems to be running A-okay at least to my opinion.

Also with the motor's resistance, I have rotated the motor manually and watched the multimeter and notice I get a variable resistance between around 1.5 ohms to about 30 ohms.

Anyways here are some pics of the new motor and pulley arrangement along with the meter readouts and a video of it in operation:

PostPosted: Thu Apr 11, 2013 10:18 pm
by Harry Dalek
Hi troy more than likely you have the right speed but if you had the pulley wheel sizes motor and such other way around it would go faster and the motor would run slower speed.

Good construction work and i see you have a hot glue gun too no NBTVer should be with out one :wink:

PostPosted: Thu Apr 11, 2013 10:24 pm
by AncientBrit

An easy way to check speed is to make up a simple strobe disc and paste to the final pulley.

When you illuminate with a fluoro or bulb driven from ac mains you will see a near stationary pattern at the correct rpm.

Something like attached pix.
BTW ignore "300 rpm" that was before drive pulley (not shown) was modified. Speed in fact is 750rpm.

I think this site in the past had a simple pdf for printout.
If not ask and someone will post.



PostPosted: Thu Apr 11, 2013 11:44 pm
by Viewmaster
AncientBrit wrote:I think this site in the past had a simple pdf for printout.
If not ask and someone will post.



Just found this strobe on the web which covers 750RPM at 50Hz mains freq
(5th pattern out from centre) It covers 316-1500 RPM

If anyone wants lower freq then this one covers 50-300 RPM

For those living in the funny lands ( :lol: ) where the mains is 60 Hz just change the 50H or 50L to 60H or 60L in the URL's.

Don't get giddy!

PostPosted: Fri Apr 12, 2013 12:09 am
by Steve Anderson
Ah! If it were run off 7.2V battery packs which is a very common voltage for R/C models the motor is probably a 6V version allowing (even in PWM drives) for a little loss in the switching semiconductor, connectors and wiring. It only requires a few milli-Ohms at these currents to lose measurable voltage in proportion to the supply.

Well, it does seem an improvement, but I would still try a higher gearing ratio if possible, current and heat should reduce even more...

Steve A.

From the photo above it would appear that you have an approximate 4:1 reduction ratio which means nominally the motor would be running at 3000RPM which is still a bit too slow for these sort of motors. Gary did make the suggestion of running the belt direct on the motor's worth a try.

PostPosted: Fri Apr 12, 2013 5:23 pm
by AncientBrit

Nice selection of speeds on that composite strobe disc.
Well researched.
That's gone into my library.

Here's my more modest single speed version.



PostPosted: Fri Apr 12, 2013 9:29 pm
by gary
Just a couple of thoughts after watching your youtube videos (which help greatly in puting things into context).

Whilst there is nothing wrong with using the strobe there are a couple of things to keep in mind. Firstly, they need a light source that is actually flickering with the mains - so CFLs won't work - oh for the good old days of incandescent lamps. Secondly, my own experiments with synchronous motors (mains driven) have shown me that the mains frequency actually varies, well, more than I expected it would.

But note you already have a strobe of sorts on the disk - the sync encoder - you can use the pulses from this to measure the speed of the disk by measuring the time between adjacent pulses (or probably better still the distance between as many pulses as you can measure to average out any discrepancies). You may want to temporarily blank out the missing pulse with a piece of black paper (make sure it's the same ink as the sync encoder - we don't want any of Harry's problems LOL). This will make the pulse train a steady 32 pulses so that triggering is constant (or nearly so).

Alternatively, if you take the pulse train into your PCs soundcard (making sure your levels are reasonable) I have a small application that can give a very accurate speed measurement (working like a true stroboscopic RPM meter). There is, of course some delay but you can work around that.

Yet another way is if your meter has a frequency measurement function - this works very well indeed - Dick Smith used to have a digital multimeter with that function for $45 - well worth it.

Any way just some observations and options that I have successfully used in the past.

PostPosted: Tue Apr 16, 2013 2:31 am
by Klaas Robers
I wonder what happens if you run the motor without the belt to the larger pulley. 0.9 volts is definitely too low.

- What does the motor when it runs with no load, so without any belt, on e.g. 6 volts accross the mottor?
- Observe the motor speed, yes these motors may run fast.
- What is the current drawn in this situation? It should be low, say 200 mA.

I guess that the motor runs 6 to 10 times as fast.

You should get close to this situation. That will include a smaller pulley on the motor, or no pulley at all, and an even larger pulley on the disc spindle.

A smaller pulley on the motor and a larger pulley on the disc spindle incudes a better "matching" of the speed - torque characteristic of the motor to the system that it has to drive. It will keep the speed much better constant.

PostPosted: Wed Apr 17, 2013 9:54 pm
by aussie_bloke
Thanks everyone for your valuable input.

Well everyone I have some excellent news. My NBTV camera is now complete and working great :D , well at least as far as I can tell anyhow. Am yet to test it with Gary's NBTV program but at least it's making a great picture on the LED monitor and the motor appears to be syncing to speed :).

Now talking about the final construction stages I have put the final components in the circuits and wired the circuits all together and mounting them in their appropriate spots. I hot glued most of the circuit boards onto a chip board for convenience and neatness and some of the circuits are situated on the frame work.

Anyways on first testing, I was getting a negative picture so had to switch the dome sensor leads around and that fixed that problem. On tweaking the controls on the head amp and video/sync mixer circuit I managed to pull up a nice greyscale picture with what I stick in front of the camera :D

Testing the signal output from the encoder circuit it was a bit off 2.5ms (400Hz) according to the scope so I had to adjust the pot a little to get the timing right. had to stick a bit of wood in front of the encoder disc as the phototransistor was picking up light from the 150W lamp. There also seems to be a slight bit of frequency drift so I have to intermittently adjust the encoder circuit pot, plus when I do get it around 2.5ms there also seems to be some noise in the pulse. I will have to do some further tests with that, apart from that disc is syncing at least as far as I can tell.

Lastly talking about the motor, I tweaked the coarse/fine speed control and encoder circuit pot until I got a satisfactory speed and it synchronizes. Checking the voltage drop this time, it is now around nearly 1.5VDC.

Anyways here below are the pics of my final construction work along with off screen shots and a little video demo of it in operation, enjoy :) :


PostPosted: Fri Apr 19, 2013 6:02 am
by holtzman
Well done, really inspiring! Haven't been on this forum for a while, your project is a great surprise.

Re: congratulations

PostPosted: Sat Apr 20, 2013 5:11 pm
by aussie_bloke
holtzman wrote:Well done, really inspiring! Haven't been on this forum for a while, your project is a great surprise.

Thanks for the great words, the project was fun to make and about 6 months in the making and I have learned a lot in building it! :D

Anyways latest update on my camera.

First thing, a bit of a horror story, I foolishly connected the main DC power leads the wrong way around and blew up some components in some of the circuits which really upsetted me greatly!!! :cry: But fortunately after I had my drama queen session I got back into proactive mode and progressively replaced all the regulators, the blown electrolytic cap on the sync mixer board, the chips on the main sync board and the camera works again. Only thing though is I'm not sure if I've damaged other components and am a bit ho hum on how correctly the sync circuitry is working. But it seems to be working again the same as before anyhow.

Next I decided to try out the camera with Gary's NBTV The Big Picture and fired the camera up, connected it to my sound card and stuck a test subject in front of it and it came up nicely on the monitor but can't tune The Big Picture to display the image, I adjusted the encoder board and motor speed but can't get an intelligible picture to display, also even did recordings in WAV and played it back on The Big Picture and adjusted the controls but couldn't get anything intelligible. So am not sure if the sync circuits are working properly at this stage.

I have in the meantime did a telerecording of the monitor screen displaying what I put in front of the lens, this time I used a late 1960s vidicon tube B&W camera to telerecord the picture as I get a nicer less flickery/broken picture due to the vidicon tube's after image persistance/lag characteristics. I am yet to upload the telerecording but have uploaded a video log made from my digicam of doing the telerecording which can be seen here . I have made 74 video logs of my NBTV camera project surprisingly, they are all on my channel if you's want to watch them through .

Anyways will keep you's updated with further developments with my camera.