Tachometers with hall proximity sensor question

[Weldingrod1]

You should check your 34 vdc supply voltage with a multimeter to see what's happening to it when you speed.up!

Sent from my SM-G892A using Tapatalk

 

[Weldingrod1]

Sorry, 24 vdc.
Also, if you measure the current you might want to do that.
A little wall wart power supply is plenty for this...

Sent from my SM-G892A using Tapatalk

 

[B2]

Thanks again for all this information, I will make some changes next week and see what happens. I just want to know one thing. I was provided with one 10mm×3mm neodymium magnet with the sensor,my question is,do yourealy need 10 or is one sufficient?

One is all you need. I put in 10 because I could and I wanted to run my lathe really slow, and the digital display had 6 digits which I did not know what to do with! So the first digit in my display is measuring 0.x RPMs. or the display shows RPMx10. When the lathe is turning at 6 RPM the display shows 000,060. But as I stated the display was smaller than any other RPM display I could find and so would fit my front panel. By the way most Neo magnets will work if physically large enough. You can get more coin magnets via the web. If they are not as thick as your original one you can stack them to create more magnetic field at the sensor. Just don't bury them in some magnetic material like steel.

PS. The difference between a "hard" magnet like Neo and a soft magetic material like steel is that the "soft" magnetic material can be demagnetized easily and so changes its magnetic oriention when put in a field. Hence, when you bury the hard magnet in the soft magnetic material the fields from the hard magnet ends (N and S poles) just travel around in the soft material to short out the field lines. I.e. the field lines from the N end must return to the S end. The steel just shortens the path preventing any of this field from reaching far away from the magnet.

 

[mksj]

I would not change your ring or magnet, I embed the magnet in steel spindles all the time and they work just fine. The tachometer you have will read down to 10 RPM, so there is no need for additional magnets. If you have the tach on and manually turn the spindle, there is a red led in the tach. sender that should light up when the magnet passes. Instead of redoing everything, just take a step wise approach
1. Use an alternate power source, as I said, a small 9V battery is ideal because it eliminates any possible electrical noise issues.
2. Check the sender unit is operating correctly, the LED lights up when the magnet is oriented correctly and it is wired correctly
3. Before I install these tachometers I check that they work, connect a 12VDC power source or 9V battery, connect the hall sensor and manually run the magnet across the sensor to see that it triggers and you see numbers on the display.
4. Electrical noise from the speed control such as a VFD usually show up as erratic RPM, not 0's or a fixed number.
5. Distance between the magnet and sensor should be around 0.1-0.2"
6. The RPM unit could be defective, I just bought one and it would show LLLL when powered, but the sender worked and it would indicate RPM's sort of. I returned it and bought a different brand. I have installed dozens of these, MacTachs, Taculators, etc., each one has there issues, but you only need one magnet for the one you have.

 

[B2]

Thanks @mksj all seems right.

However, embedding the magnet in a soft magnetic material does reduce the external field. Putting it behind the a piece of soft magnetic material virtually eliminates the field (called a magnetic shield). But many of the Hall sensors are pretty sensitive. Any you have to get closer with the sensor when you bury the magnet in soft magnetic materials. Also, not all steel is all that "soft" so this can help a little. Some stainless steels are not even magnetic. Having worked in magnetism and magnetic materials most of my life I tend to try to always optimize these sorts of magnetic structures. Hence, my use of the nylon, where it can be buried and you do not have to worry about it flying off of a glued surface.

By the way, where did you get your material for your spider. I have been looking for a thick pipe without a lot of success. I have the PM1440GT lathe which has a 2" spindle ID and the OD is about 2.5". What is the OD and the ID of your spider material?

 

[mksj]

Hey B2, agree that different materials can effect magnetic flux, but when using the neodymium magnet have quite a strong field effect. Although non-magnetic materials may do better, I did have an interesting experience with the sensor was embedded in an outer aluminum ring and the spinning magnet induced a field in the aluminum that decreased the effected of the sensor. So I often recommend using an L bracket as opposed to a contiguous ring to hold the sensor. They are pretty simple devices, but the MachTach (no longer available) and Taculator do require additional magnets due to their shorter sampling times.

I use a number of different materials for the spiders. Previously I had bought a 12" section of 2.5" OD 1.5" ID DOM steel tubing and made several spiders for other people. More recently I use 1144, 4140 or 8620 in 2.5-3.0" solid stock, if you have a spider under 4" I use a two inch long x 1.5" annular cutter and through bore from both sides, and then use a boring bar for the finishing ID. Longer spiders I use MT3 drills and then center bore with a carbide boring bar. The solid stock is readily available on eBay at more reasonable prices for rems vs. DOM tubing, but the latter requires less material removal. Do not like working with 1018.

I would probably start with 2.75 or 3" stock for the 1440GT, I like to have a 1/2" wall thickness for the spider bolts.

2.5" OD x 0.5" Wall x 1.5" ID Mild Steel Round Tube A513-Type 5 DOM | Online Metals

Order 2.5" OD x 0.5" Wall x 1.5" ID Mild Steel Round Tube A513-Type 5 DOM stock today on our website. Buy online with fast shipping direct to your door.

www.onlinemetals.com

2 1/2" Diameter 1144 Stressproof Steel Round Bar Stock - 2.5" x 10.25” Length | eBay

Find many great new & used options and get the best deals for 2 1/2" Diameter 1144 Stressproof Steel Round Bar Stock - 2.5" x 10.25” Length at the best online prices at eBay! Free shipping for many products!

www.ebay.com

2 1/2" Diameter 8620 Steel Round Bar Stock - 2.5” x 9” Length | eBay

Find many great new & used options and get the best deals for 2 1/2" Diameter 8620 Steel Round Bar Stock - 2.5” x 9” Length at the best online prices at eBay! Free shipping for many products!

www.ebay.com

2 1/2” Diameter 4140 Heat Treated Steel Round Bar Stock - 2.5” x 5.875” Length | eBay

Find many great new & used options and get the best deals for 2 1/2” Diameter 4140 Heat Treated Steel Round Bar Stock - 2.5” x 5.875” Length at the best online prices at eBay! Free shipping for many products!

www.ebay.com

Spider material HR 8620, deep boring with 1" carbide boring bar.

 

[B2]

Thanks @mksj . Thanks for the spider material info. That is a lot of boring from solid material! I was hoping for a shorter trip. Looks like you have your spider mounted on a smaller diameter spindle than the 1440. I had found the thick walled tubes at online metals, but not in the larger 2.5" OD spindle of the 1440.

Yes, the spinning field can cause eddy currents in conductors to cause opposing fields. However, if the Al housing and the magnet are not moving relative to each other then there would be no eddy currents in the Al. There must be more to it or I did not understand your configuration of parts. Now, the spinning magnet probably causes some eddy currents in the metallic sensor housing and I assume this is why the put the sensor right out at the end and embed it in plastic. I suppose, these currents in the sensor housing could also be imaged back into the Al magnet holder causing other fields, but I would prefer to assume that these are much smaller. By the way, if you do not believe in eddy currents try the following. Drop a Neo magnet down a copper (non-magnetic) pipe and observe how long it takes to fall through. As a comparison just drop the magnet without a pipe. (Catch it on something soft, rug, so that it will not chip or break.) Use a magnet that is similar in size to the pipe diameter. A cylindrical magnet with height similar to the diam is best. Or if you have number of coin magnets stack them...but I have never tried this. The opposing generated eddy currents in the pipe wall tries to support the magnet so it slows down the fall. Copper is the best material for the pipe as Copper is highly conducting. Al will work, but it is not as conductive and so is not as effective.

It is more trouble, but very impressive, is a device that I built to explain a circuit in one of my classes. I took a flash charger out of an old Polaroid camera so as to get an energy storage capacitor. I made a special ~1" diameter coil from copper tape and embedded it in epoxy so that it would not fly apart. using 4 AA batteries and the step up transformer circuit from the camera I would charge the capacitor up to a couple of hundred volts and then discharge it through the coil. On the flat coil I placed an Al washer (~1" OD with a ~ 3/8 ID hole). The rapidlly changing magnetic field would induce eddy currents into the washer and these currents generated a field to opposed the coil field . The washer would shoot up about 30 feet into the air. This is a form of what is called a rail gun. At least I had the student's attention for a while!

 

[pontiac428]

Two things, you might have your sensor input wires (gnd and signal) backwards. The difference is whether you're triggering the tach on the rising or falling side of the wave. Also, it may work better for you to use a 1K pullup resistor between your VFD supply voltage to the sensor and the signal output wire to the tach. That way, you get a full 0-5v (or whatever) square wave signal to the tach. An oscilloscope is a useful tool here, but these two things can be tried as quick tests without any real analytical effort.

 

[Suzuki4evr]

Ok I am going to leave the ring as is for now,do the 9V battery, move the sensor closer to the magnet and see what happens. If that fails,hopefully not,then do the plastic ring.

Two things, you might have your sensor input wires (gnd and signal) backwards.

Currently I have it on 24V and the negative on CM of the vfd,blue(sensor) to blue of display,black(sensor) to black of display and sensor (+) to 24V on vfd. I am also going to try the 10V supply of the vfd.

 

[B2]

Hi @Suzuki4evr

I think your objective for now is to simply determine if either the sensor or the digital display is burned out.

I do not think you have enough current out of the VFD 10 volt supply to use it. Also, it is not a good idea as you want it to be stable for the VFD speed input. I looked up the manual and is says 20mA and this is to power your VFD speed pot. Follow Mark's suggestion of using a 9V alkaline battery to power the sensor. Clip leads are nice to have, but if you do not have them or a battery holder you can sometimes just solder a wire to the terminals. It will have plenty of current to last for a while. Duracell 9volt batteries can put out as much as 1 Amp for a little while, but will get hot and the terminal voltage will drop down to about 7 Volts. I tested this on camera flash chargers and they would do this for 20-30 seconds before the flash was completely charged. I could charge the thing 30-40 times before the battery was low. So a fresh 9V battery will be plenty for your tests.

The VFD manual says there is an extra 200mA from the 24V supply for external use. Maybe? But be careful here. My Hitachi VFD manual implies that if you short out its 24 volt supply it is all over. I say implies, because there is only one place in the >600 page manual that it says anything about the current limit so I contacted the service center and did not get a straight answer... "Ohhhh don't do that" was the response.

Also, the 1K pull up resistor could be hooked up to the 10V supply but that might even be too much. Be careful hooking up to 24 volts even with the 1K resistor inserted. It is near the limit of where some of these electronics want to work and the digital display input may not like it. Don't know what the display wants to see, but it is probably set up for 5 volt logic sorts of levels. If you really want to try the pull up resistor, then make it larger so as to limit the current. Even a 10K or a 20K should be providing sufficent current.

These things are cheap and some times just die. Other times you may have zapped them and did not know you did so. I had one display that worked fine for about 3 days. Then just quit. Not all that uncommon. Cheap, mass produced stuff is not necessarily bad but it may not have been tested well. QC is an issue. Even many of the ones sold by big named US companies are commonly the same ones being sold in China, or at least were made in China. Where is your iphone made? Not in the USA. Is extensive QC testing required by the US company?