PDA

View Full Version : Question about diode application.



DICKEYBIRD
05-25-2013, 11:01 AM
Still messin' around with the DC motor/speed control tach generator and have another question; this time about diodes. Original thread: http://bbs.homeshopmachinist.net/threads/59249-Tach-generator-KB-control-PCB-question?highlight=tach+generator

I now have a simple DPDT relay wired in your basic polarity reverse config. in the output leads of the DC tach generator. The relay coil needs to be energized to reverse the output polarity only when the the spindle motor is running in reverse. I could control the relay with yet another relay controlled by Mach or I was thinking of energizing the relay coil from the output side of the KB (reversing) speed control. It outputs max 90 vdc in reverse polarity (obviously) when the motor is running backward. The KB manages dynamic braking/reverse extremely well and "knows" how to stop smoothly from full speed then reverse direction safely with no danger of electrons getting confused & blowing up something due to the operator's input errors. That's why I'd like to slave the tach-gen reverse polarity relay directly to the output of the KB. That way there'd be no danger of the tach's output being reversed whilst the spindle is spinning down.

Would it work if I was to put a "clever" diode or 2 on the motor output leads of the KB to energize the relay coil? By clever I mean one that could pass a max of 12v at a 150 mA or so and be set up to power the coil in one polarity direction only. The relay coil is rated 12v/75 mA.

I know not of these electro-magical devices but am sure ye grizzled sorcerers of said magic will know if this is feasible or if I just need a good dope slappin'.

MaxHeadRoom
05-25-2013, 11:43 AM
Here's the dilemma, you want to reverse the tach to ensure that the +ve output is always connected to the +tach terminal when the motor is in the appropriate direction, this excludes paralleling the reversing relay as the motor, may or may not be in reversal at that point?
IOW as the motor traverses through reversal, the tach output will be very low or near zero, one method to detect the point of reversal is a op amp/comparator circuit that is sensitive enough to detect the small change in tach voltage reversal and immediately pick up a relay that reverses the board connections when motoring in that direction.
Max.

Fasttrack
05-25-2013, 12:25 PM
Would it work if I was to put a "clever" diode or 2 on the motor output leads of the KB to energize the relay coil? By clever I mean one that could pass a max of 12v at a 150 mA or so and be set up to power the coil in one polarity direction only. The relay coil is rated 12v/75 mA.



Diodes don't exactly work like that. They do have a breakdown voltage, above which they begin to conduct. If all you want to do is drop a signal from 90V to 12V for a relay, there are a variety of ways to do that. If you are looking for a simple solution and don't care about isolation, then you could accomplish that using a zener diode and a resistor. The zener diode is in parallel with the relay coil and the resistor is in series. If the polarity of the signal will change, you will want another diode.

But I don't know jack about these speed controls. Based on what MaxHeadRoom says, it sounds like there may be more to the story.

MaxHeadRoom
05-25-2013, 12:47 PM
To pursue the op amp idea, do a search for LM311 zero cross over, there should be whole raft of circuits, they will drive a small relay directly.
Max.

mikem
05-25-2013, 12:56 PM
Put a diode in series with the relay coil to only conduct when in reverse mode and use a series resistor to run the relay from 90 volts. From your data, the relay coil resistance must about 160 ohms. At 90 volts the resistance that would draw 75ma is 1200 ohms. Subtract the 160 and that leaves 1040 ohms in series to run the relay at 90 volts.

IF your input voltage varies with speed, you could use a voltage regulator IC like a 7812 or 7912 to get your voltage constant at 12 volts--but your input must be a few volts higher than 12 to get it to regulate.

DICKEYBIRD
05-25-2013, 01:25 PM
Hmmm, great info thanks all! More head scratchin' to do. Probably safer to go back to plan A; triggering the tach signal reverse relay from the B.O.B.'s spindle reverse relay output & make sure to put an M5 spindle stop & a short pause in the code before switching directions.

As usual, my so called clever ideas require more cleverness than I have.:rolleyes:

MaxHeadRoom
05-25-2013, 02:11 PM
Incidentally how are you handling the M5 stop? If using braking, is the inhibit on only at motor rest? if so when is the inhibit applied?
Max.

DICKEYBIRD
05-25-2013, 02:44 PM
Good question Max, I'll have to get back to you later. I'm at *!^$!*(!) work today & getting hammered.

If I remember correctly, I'm switching 1 KBCC terminal on for forward, another terminal on for reverse and both terminals off is off/brake. I'll ck. later when I get back to the shop to be sure.

MaxHeadRoom
05-25-2013, 02:55 PM
I have never used that particular model, so looking at the manual it would appear that maybe the inhibit is activated in the braking mode, which may make sense as all control would be removed from the motor and the brake resistor connected across the armature?
Max.

DICKEYBIRD
05-25-2013, 03:28 PM
That sounds right. I don't think I'm using the enable or the inhibit terminals yet. May not need to? There's a schematic for the A.P.R.M. module and the KBMM speed control in the manual but I wouldn't know what to look for.

I put the big (6" 4-jaw) chuck on it a while back just to test it. I spun it up to 1400 rpm and entered M5 with my fingers on the brake resistor. It came to a quick stop and I could feel the resistor get warmish.

MaxHeadRoom
05-25-2013, 04:03 PM
If you wanted another relay for the tach when the reverse occurs, you could tap across RY1 coil, it is a 110vdc relay.
It may be OK in this case, if the inhibit occurs during/before reversal.
It is OK to reverse the tach if the motor is inhibited.
Max.

Paul Alciatore
05-25-2013, 05:54 PM
I re-read the other thread and was surprised to find that the tach leads had to be reversed to run in the other direction. But, whatever. Normally a tach provides pulses and any polarity of the leads would only be for power to the tach. But, I don't know everything and what works, works.

I have to think that the designers of this speed control must have had some scheme in mind for reversing the motor direction. Have you tried to contact them and ask? Perhaps there is a jumper or input or something for this situation. I say this because reversing the tach leads when the motor is turning in the opposite direction seems like a bad way to do things. I mean, what if you flip the reverse switch and the motor is still going forward? Do you reverse the tach leads immediately or do you have to wait until the motor actually stops and reverses? Then, how would you detect that reversal? This just does not sound right. Or is the operator expected to turn the motor off and wait until it stops before applying reverse power? That also does not sound like a good design.

I think you need to know more about how the circuit is designed to work before you make any further changes. Do contact the manufacturer.

MaxHeadRoom
05-25-2013, 06:18 PM
The issue is that this is not a version such as the 4 quadrant version where the direction is changed without physically reversing the armature leads, as the manual states, if using the armature reverse option in conjunction with a Tach, the +v tach polarity output has to be maintained at the + tach input, IOW the drive does not automatically recognize a tach reversal, as it would be done in the 4 quadrant version or your typical spindle/servo controller, where either a dual bridge or H bridge motor control exists.
Quote:
Note: If control is used on a reversing application, the tach wires must also be reversed so that positive (+) is always connected to B or T.
To set control for tachometer feedback.
Max.

DICKEYBIRD
05-25-2013, 08:47 PM
I have to think that the designers of this speed control must have had some scheme in mind for reversing the motor direction. Have you tried to contact them and ask? Perhaps there is a jumper or input or something for this situation. I say this because reversing the tach leads when the motor is turning in the opposite direction seems like a bad way to do things. I mean, what if you flip the reverse switch and the motor is still going forward? Do you reverse the tach leads immediately or do you have to wait until the motor actually stops and reverses? Then, how would you detect that reversal? This just does not sound right. Or is the operator expected to turn the motor off and wait until it stops before applying reverse power? That also does not sound like a good designThanks Paul. I'm guessing this control design was approaching the end of it's life cycle when it came out and the 4 quadrant designs Max mentioned were right around the corner. The DC tach generator feature (again I'm guessing) is old technology and the modifications to the design to solve the polarity issue weren't practical.

Coincidentally, I bumped into a fellow on another forum that's working with exactly the same setup I have and he's already requested some tech info from KB. I hope he gets some good info from them.

MaxHeadRoom
05-25-2013, 08:59 PM
I'm guessing this control design was approaching the end of it's life cycle when it came out

I don't think its so much at the end of the life cycle, as KB has always made the 'simpler' types of DC motor controllers, but I believe the add on board for reversing and braking was to cater to those that needed a little extra over and above the basic KB-125 etc.
For those that want reversible drive with also tach reversing, there is the KBRG-225D etc.
And then of course there is the PWM drives also.
Max.

EVguru
05-26-2013, 06:48 AM
Can you not just use a bridge rectifier on the tach output so it's always the correct polarity.

If the diode drop is an issue, there are precision rectifier circuits on the web.

DICKEYBIRD
05-26-2013, 09:43 AM
Interesting idea Paul, thanks. KB is specific about the 7V/1000 rpm for the tach input and that's the spec for the Servo-Tek tach I'm using. Dunno how much voltage drop I'd get using a bridge as you suggest or what effect the lower voltage would have. I suspect it would prevent adjusting the speed properly.

I looked up precision rectifier circuits and that's a no-go for me. Way over my head.:o

Paul Alciatore
05-26-2013, 11:33 AM
Now I understand. We are not talking about a real tach with a pulse train output. It is a simple speed sensor that gives you a DC signal that is proportional to the motor speed. Paul Compton's suggestion gave me an idea. But I need to see an actual schematic diagram, including how it is powered, to proceed with it.

MaxHeadRoom
05-26-2013, 11:53 AM
But I need to see an actual schematic diagram, including how it is powered, to proceed with it.

There is no power to it per se, it is a simple DC brushed P.M. generator.
It is a 'real tach'! Just analogue.
Max.

EVguru
05-26-2013, 12:04 PM
A bit of voltage drop from a bridge rectifier will cause a speed offset between the requested and actual speed. That shouldn't be an issue except at very low speeds.

Using Schottky diodes in the bridge circuit would keep any voltage drop to a minimum, but they should be rated for the peak output voltage of the tachogenerator.

MaxHeadRoom
05-26-2013, 12:19 PM
The output of the tach in question is I believe 7v/1krpm, so that = 142rpm/v, the drop on a bridge is around 1.2v - 1.3v, I would imagine that at low rpm that may cause some hunting due to the dead spot?
Max.

DICKEYBIRD
05-26-2013, 01:19 PM
Not sure at what RPM the lower voltage would affect real world performance but I suspect it would be a big problem dropping over a volt across a bridge as Max calculated. My lathe is direct drive and the low end is where I hope to gain stability using the tach feedback function. No reason to mess with the tach if it doesn't work well at low speed.

I did a little poking around this morning and my B.O.B has a pin 16 output terminal paralled to the onboard reverse relay's activation circuit. I can use that output to trigger a reed relay to switch 12V to the tach polarity reverse relay coil when the reverse command is sent. The B.O.B will supply 28 mA @ 5V so I should be safe with the reed relay's 20mA/5V coil. Is that bad electro-magical Ju-Ju switching a relay with a relay?;)

DICKEYBIRD
05-26-2013, 01:38 PM
If you wanted another relay for the tach when the reverse occurs, you could tap across RY1 coil, it is a 110vdc relay.By the way, I meant to thank you earlier for that tip Max as it's a good'un. If it weren't for the KB being all snug and wired into its new encosure I'd remove the A.P.R.M. and see about tapping into the coil circuit as you suggest. That'd probably be the best solution.

Paul Alciatore
05-26-2013, 03:28 PM
There is no power to it per se, it is a simple DC brushed P.M. generator.
It is a 'real tach'! Just analogue.
Max.

Ah ha! Last piece of the puzzle. Great.

Paul Compton's bridge rectifier will work with just a bit of a modification to make up the Voltage lost in the bridge. Here is my suggested circuit:

http://img.photobucket.com/albums/v55/EPAIII/DCSpeedSensorMod_zpsffe64910.jpg (http://smg.photobucket.com/user/EPAIII/media/DCSpeedSensorMod_zpsffe64910.jpg.html)

The simple addition of a 1.5 Volt cell in series with the sensor line will make up the constant, 1.5 Volt loss in the bridge. So you get a rectified sensor signal that is back to the proper level. A single AA cell will probably last a long time, depending on the current draw of the motor control circuit. Change it every year or two to be sure it does not corrode and leak.

One caution: The sensor input will never drop below 1.5 Volts so if that is a problem, you may not be able to use this. But I suspect that that represents a very slow speed and you will not be operating that slow.

Oh, sorry, some may label this Voltage generator a "tach" but I just can not bring myself to do that.

DICKEYBIRD
05-26-2013, 05:19 PM
That's purty clever Paul, thanks!

I didn't hook up the generator (do you like that description better?:D) to the control but did check how many spindle rpm equals 1.5V output and it's 192 rpm. I don't know if I'll be going that slow in reverse. Won't know that until i get into threading steel.

One nice thing I learned that I hadn't checked before was that the direction doesn't affect the voltage. 192 rpm in reverse gave me dead on 1.5v too. Apparently Servo-Tek did their homework on brush timing.

I am sure that I don't want to leave a constant 1.5V feed into the control 24/7 though. Dunno if it'd hurt but it just doesn't feel right.

I dug up some proto board and am sticking together a little reed relay circuit to switch the polarity reverse relay. If that doesn't work I may give your circuit a try.

MaxHeadRoom
05-26-2013, 05:41 PM
Apparently Servo-Tek did their homework on brush timing.


They have been making DC tach's for over 50yrs, so they have probably got it right by now!;)
Max.

DICKEYBIRD
05-26-2013, 05:57 PM
They have been making DC tach's for over 50yrs, so they have probably got it right by now!;)
Max.Yep, I could tell immediately it's a quality unit. Got real lucky & darn near stole it on ebay.

macona
05-26-2013, 06:45 PM
You might be able to use an opamp as a full wave precision rectifier, that should eliminate the voltage drop of the diodes.

http://sound.westhost.com/appnotes/an001.htm

EVguru
05-28-2013, 04:57 AM
You might be able to use an opamp as a full wave precision rectifier, that should eliminate the voltage drop of the diodes.

http://sound.westhost.com/appnotes/an001.htm

That was one of my suggestions, but Dickeybird isn't comfortable with building electronics and you may well need specialist high voltage op-amps. We've not been told just how high the motor is being revved.

I'd just try numging a bridge in to see what effect it has. If low speed control is a problem, I'd have a search and see how low a voltage drop Schottky diode I could find (with a high enough breakdown voltage). Some Schottky diodes are as little as 0.15v conduction drop.

DICKEYBIRD
05-28-2013, 08:38 AM
...but Dickeybird isn't comfortable with building electronics...Ahh, the British gift of understatement.:D