Tachometer hep needed

[miro]

I am trying to work out a tachometer for a sawmill. The circular sawblade runs at 500 RPM, or 8.3 RPS. The blade is 42 inches in diameter.
I have built a prototype using the LM2917M-8 chip . I use 12 magnets on one of the pulleys that drive the saw blade, so I get about 100 pulses per second.
For my prototype, I mounted the 12 magnets on a 6 in diameter cast iron ring, and set that up on my lathe which I know runs at 325 RPM. The DC output is about 0.25V for that speed. The DC output seems to be fairly linear with speed.

I'm trying to figure out an analog meter circuit that would have a zero reading speed of, say 350 RPM, and a full scale reading of, say 700 RPM so the 500 RPM speed is in the middle of the meter scale.

The application note for the LM2917M-8 gets me started and I've got the prototype to make an output at the 325 RPM speed, but I'm stuck now to get the rest of the idea done.
The frequency - voltage curves on the datasheet, seem to be much higher frequencies than I am using ( eg 1000 - 5000 Hz)

So, I need help to figure out what meter movement I should use ( I can easily get 50 to 100 microamp meter movements) , and how I should get the meter zero scale to be 350 RPM and the full scale to be 700 RPM.

I am new to the forum.

miro

[pebe]

It should be fairly simple to add an op-amp to do the job.

Looking at the National Semi datasheet, are you using the circuit for the 'Minimum Component Tachometer' on page8?
If so, are you using the 10K output resistor as specified there for the emitter follower, and what is your supply voltage?

[miro]

Yes, I made the circuit as described in the application note - the minimum parts one ( I'm frugal)
And yes, I am using the 10k resistor on the output.
I tried putting a variable 10K pot on the output, but that just lowered the output voltage ( no surprise there)

I'm using a 12 V lead acid gel cell as a power supply. The tach needs to be portable to get to the sawmill where there is no convenient AC power. Te measured voltage is 13.3 V. and I have a 470 ohm resistor on pin 6
But I am open to other suggestions and devices, though.

miro

[pebe]

Hi miro,

Here is a circuit that will do what you want. A CA3140 op-amp converts the output voltage at pin4 of your 2917, to current through a 100µA meter. If you want to use a 50µA meter, then change VR2 to 10K. You can change the op-amp but it must be one with FET inputs capable of running with inputs as low as 0.2V. Bipolar types such as the 741 won’t do.

Pin4 gives a voltage of 0.25V for a speed of 325rpm, so it follows that pin4 changes from zero at the rate of 0.077V per 100rpm. So you will get 0.231V at 300rpm and 0.538V at 700rpm. That will put 500rpm at mid scale. If you want a different RPM span, then change these two voltages as appropriate when setting up.

The voltage at point ‘A’ is provided from the stabilized 7.56V available from pin6 and is the reference voltage for comparison with pin4. Set it up like this:

1. Disconnect the meter. Set VR1 and VR2 to max resistance.
2. Switch on, and adjust VR1 to give 0.231V at ‘B’.
3. Now simulate the signals at pin4 like this:
4. Temporarily connect a 100K pot, P1, (wired as a variable R) in series with a 270K resistor between pin4 and pin6.
5. Set P1 to give 0.231V (= 300rpm) at ‘A’.
6. Connect the meter. If necessary, re-adjust VR1 to give zero current in the meter.
7. Set P1 to give 0.538V (= 700rpm) at ‘A’.
8. Adjust VR1 to swing the needle to full scale.
9. Set P1 to give 0.385V (= 500rpm) at ‘A’. Meter should read mid-scale.
10. Remove the temporary resistance between pin4 and pin6.

The meter will be reading backwards (less than zero) when the motor is running at less than 300rpm. If you don’t want that, insert a 1N4148 diode at ‘X’ with anode to meter after setup.

Please note that R2 is there to limit current to prevent accidental needle damage.

If there’s something I haven’t covered then please shout.
Best of luck.

(Edit 16.01.13 - 2917M IC number corrected)

[miro]

That seems to be what I was looking for .
I'll let you know how it worked out once I've got the circuit added to the tachometer circuit.
Thank you
miro

[miro]

Actually, I was also thinking that the flat belt pulley on the sawmill blade shaft is a large diameter - about 24 inches.
On my test set up I was using a 6 in diameter disk for the 12 magnets.
On the large pulley, I could use more magnets to create a higher frequency for the same RPM and thus get a higher output.
For the moment, that will be a "PLAN B" .

I know that there are one or two other heritage sawmills that could use the same thing so I'll do the experiments and "learning scar tissue" on ours and then make up a couple more units for them.

Most of the chaps around our sawmill are deathly afraid of electronic things and much prefer mechanical things they can see, touch and feel
Thanks again,
miro

[miro]

Well, I finally did get to the task of adding the expanded range circuit to the tachometer.
I can report that even though I am not an experienced circuit board guy, I did manage to add the components to the existing circuit board, and I did manage to get it operating.
Adjusting it took a bit of tweaking, but that's part of the learning process.

I set it up so that the zero on the meter is 375 RPM and the full scale is 600 RPM. That' s a bit different from what I had in mind originally, but I think it will be better for the sawmill operators.

Our sawmill is set up t run at 500 RPM which is mid-scale for the meter and the guys will be able to see if the blade is slowing or over-speeding by simply glancing at the analogue meter face. I'm thinking that I should dis-assemble the meter and put a green zone in the middle of the scale and two red zones above and below the green zone.

I'll have to wait some more before the 4- 6 ft snow drifts around the sawmill have melted before I can actually install the tachometer and try it out.

Thanks for the assistance and this little tachometer project has spurred me onto other small projects eg a precision voltage reference circuit so that I can see what the error is on the DVM.

I'll try to figure out how to post some pictures so you can see the results.

miro

[miro]

My attempt at pictures . . . miro

A view of the circuit board, meter and mechanical tachometer I used to calibrate the electronic tach

P1040397.JPG (488.78 KiB) Viewed 15468 times

This is the cast iron ring with the 12 magnets harvested from hard drives

P1040398.JPG (522.8 KiB) Viewed 15468 times

This is the pickup coil I used on the rotating magnets to get the input signal

P1040399.JPG (539.47 KiB) Viewed 15468 times

Here's my circuit board - amateur attempt to be sure - but it works. I'll make a nice case for it later.

P1040401.JPG (489.25 KiB) Viewed 15468 times

[pebe]

Hi Miro,
I'm pleased that you have got it all working. I'm thinking about you making more for other sawmills.

Rather than dismantle the meter to mark the scale, would it perhaps be better to change the circuit to have a red and a green light instead? The operator would notice it in his peripheral vision so he would not need to consciously read the meter.

[miro]

I thought of that too, but the main use of the tach is to "train" new guys on how the sawmill feels, when it is cutting right, and more importantly when it is NOT cutting right.
After they have sawn several logs, most of the guys "get it" and are good.

The mill has a mechnical feel and a sound that tells the guys who are operating the mill that it's all good.
I'd prefer to have them use the "feel" of the mill so that when things go awry, that they react very very quickly rather than look away to see what the light or the meter indicates.

Most ( heck nearly all) of the guys are more mechanically inclined and electronic stuff makes them nervous because they don't understand it , can't see how it works, and thus are afraid f it.

SO, as a first pass at the training issue, I'll use the analogue meter, see how the training sessions go, and then decide whether or not I need to do sometng more - but it's great to have a Plan B in the hip pocket.

If I do go ahead with the RED GREEN indicator, I'll certainly be back here to check what I've done.
miro