Hello. I am an extreme newbie when it comes to IC design, but I have invented a new sort of motor topology and have a couple other projects that require an SPDT relay that feeds from the high side. So I sketched up this rough idea, but I really need some pointers on a couple of things. First off, does the circuit look ok as it is? Also, I am sure I must need some sort of amplifier circuit from the output of the hall sensor, but perhaps not? I have included a picture of the circuit if anyone an help me sort this out. Thanks

 

What type of relay are you talking about? It has two coils?

 

All the modern Hall-cell IC sensors have the required amplification, offset cancellation, filtering, output span range, zero gauss quiescent value and any other required signal conditioning.
Many will allow to select the output signal type: linear voltage, PWM, or serial data.
I didn’t check the device you are considering, please post the data sheet.

 

What type of relay are you talking about? It has two coils?

Those coils are the loads. Sorry for any confusion

 

All the modern Hall-cell IC sensors have the required amplification, offset cancellation, filtering, output span range, zero gauss quiescent value and any other required signal conditioning.
Many will allow to select the output signal type: linear voltage, PWM, or serial data.
I didn’t check the device you are considering, please post the data sheet.

Here is the Ti page on the sensor I would prefer:

https://www.ti.com/product/DRV5013

 

I might need the SS41 Honeywell sensor instead. I did not realize the DRV5013 was limited to 30KHz

https://sps.honeywell.com/us/en/pro...r-position-sensor-ics/ss41-l-t2-t3-s-sp-ss51t

 

From the Honeywell datasheet:
“The open-collector sinking output voltage is easily interfaced with a wide variety of electronic circuits.”

Since it is an open collector, the output signal can be pulled all the way up to the maximum rated voltage of 24 volts.

So no. You don’t need a voltage amplifier. However to drive a high current load, an external transistor is required. See the typical connections listed on the datasheet.

 

Why do you need high-side drive? It's just a 2-phase motor of some type, so why can't the phases be switched on the negative and the common terminal (which is now connected to earth) be connected to V+?

 

Why do you need high-side drive? It's just a 2-phase motor of some type, so why can't the phases be switched on the negative and the common terminal (which is now connected to earth) be connected to V+?

These are good questions. I need an SPDT switch for a couple of different projects and one of them is a rotary motor and the other involves linear actuators. I think the motor is the bigger challenge at the moment though.

So in forward mode I want it to switch high mainly because that's what seems to work best in my test model, but I probably need to make an alternate logic for reversing. I imagine I need some sort of diverted logic so I'm not running the PNP gates backward.

I want that SS41 series sensor because it switches at 100KHz and that would allow my motor a maximum of 12,500RPM(a speed it is unlikely to ever need to be run at, but with the DRV5013 I could only reach about 3700RPM). I just need that sensor to toggle between the two lines; regardless of which way the current is flowing.

Sorry my terminology and diagrams aren't great. This is the first time I've ever attempted to design a circuit. I was hoping to avoid using any microcircuitry in my design, but obviously its just the superior option; thus I must learn these things.

 

From the Honeywell datasheet:
“The open-collector sinking output voltage is easily interfaced with a wide variety of electronic circuits.”

Since it is an open collector, the output signal can be pulled all the way up to the maximum rated voltage of 24 volts.

So no. You don’t need a voltage amplifier. However to drive a high current load, an external transistor is required. See the typical connections listed on the datasheet.

Thanks. I have been staring at these diagrams for a while and am starting to make sense of them somewhat. I need to figure out what values to give things though. The diagrams don't specify what sort of diode they're using.

On my motor project: I need to figure out how to reverse it. I could have a switch that inverts the output, but if the motor is generating power I think that would still mean it runs the mosfets in reverse.

I did see one of their circuits used an SCR. Not sure if that would be a better choice. How do they handle reverse bias?

The circuit I need must handle mostly simple DC, but I would like it to be able to process pulsed DC too. Right now with my current plan it doesn't really have a way to maintain state between pulses though. So maybe that's a feature for later?

 

What type of Motors are You trying to operate, and for what purpose ?
Please provide links to their PDF Spec-Sheets.
Why have You chosen a 2-Phase-Motor ?
.
.
.

 

What type of Motors are You trying to operate, and for what purpose ?
Please provide links to their PDF Spec-Sheets.
Why have You chosen a 2-Phase-Motor ?

For the motor project it is a completely new topology for which there are no spec sheets. It fundamentally requires a two-phase type arrangement. The main purpose for this specific prototype is to power a yard vehicle. I have an engine-generator system I have designed and I want to pair it with the motor. My motor design is versatile though and can be used for nearly anything that uses a motor.

 

Three-Phase Motors will always be superior, in almost all respects, to a 2-Phase-Motor.

If You still want to tinker-around ......................

Since You are designing your own Motor,
Switching the "Low-Side" is virtually always more advantageous.
N-Channel-FETs are cheaper, smaller, they have less Gate-Capacitance to drive,
and a lower Rds-On rating, than comparably sized and priced P-Channel FETs.
Heat-Sink requirements are usually reduced as well.

Designing a Motor is not as simple as it looks,
and designing a "new" type of Motor that is superior to the standard designs
that have been around,
almost unchanged,
for over ~100-years,
is close to impossible.

Tesla-Cars just recently came up with a legitimate, and substantial,
Motor-Design improvement quite recently,
do You really think that You can do better than their really well-paid Engineering-Team ?
.
.
.

 

Three-Phase Motors will always be superior, in almost all respects, to a 2-Phase-Motor.

If You still want to tinker-around ......................

Since You are designing your own Motor,
Switching the "Low-Side" is virtually always more advantageous.
N-Channel-FETs are cheaper, smaller, they have less Gate-Capacitance to drive,
and a lower Rds-On rating, than comparably sized and priced P-Channel FETs.
Heat-Sink requirements are usually reduced as well.

Designing a Motor is not as simple as it looks,
and designing a "new" type of Motor that is superior to the standard designs
that have been around,
almost unchanged,
for over ~100-years,
is close to impossible.

Tesla-Cars just recently came up with a legitimate, and substantial,
Motor-Design improvement quite recently,
do You really think that You can do better than their really well-paid Engineering-Team ?

Its not a traditional "2 phase" motor. So those rules do not apply. I will consider trying to switch it low instead. As to competing with Tesla.... that is a really absurd standard, but my motor does a lot of things their design does not. Superior? Probably depends on what metric you're using. Try using a Tesla motor as a servo and you'll see what I mean.

 

Ok, so I redrew the circuit to make it a low side switch. Is this more suitable? Can anyone please critique my resistor values? Any suggestions on what diode to use? Any suggestions for mosfets would be fantastic too!

 

1)
What Voltage will the Motor be run with ?
Battery ? (what type and size ), Power-Supply ? (what type and specifications ).

2)
What is your estimated Motor Current Consumption, at no-Load, and under Locked-Rotor conditions ?
And, is your Power-Supply Current-Limited, and if so, what is the Current limited to ?

3)
What is the DC-Resistance of your Coils ?, ( an Inductance-measurement would be ideal ).

4)
What is the Wire-Gauge used for the Coils ?, and approximate length of Wire in each Coil ?

5)
What is the approximate average Diameter of your Coils ?

6)
Approximately how many Turns of wire are your Coils made-up of ?

7)
Are Your Coils surrounded by an Iron or Steel Core, or are they partially, or fully, Air-Core ?

8)
Is there any provision for insuring which direction the Motor will start in ?

9)
Is there any provision for reversing the direction of rotation ?

10)
Is Motor-Braking desirable ?

11)
How will Flyback from the Motor-Windings be handled ?

12)
Will the Motor run acceptably with Freewheeling-Diodes across the windings ?
.
.
.

 

1) What Voltage will the Motor be run with ?
Battery ? (what type and size ), Power-Supply ? (what type and specifications ).

So far I have tested my prototype on direct current from batteries ranging from 6v up to 36V. This particular design seems to work best at 12v, but it is variable speed so it can run off a lot of different options.

2) What is your estimated Motor Current Consumption, at no-Load, and under Locked-Rotor conditions ?
And, is your Power-Supply Current-Limited, and if so, what is the Current limited to ?

If I am figuring things correctly the current test rig pulls about 2.88 KW at 24V at saturation without any limiting. My next model will have double the impedance though. A locked rotor isn't really much different other than being stuck waiting to reach the sensor to change phase. This design can purposely lock in positions.

3) What is the DC-Resistance of your Coils ?, ( an Inductance-measurement would be ideal ).

The DC resistance on my test rig is 0.2 ohms per phase. The next model should be at least 0.4 ohms.

The current rig SHOULD be around 2.50e-1 Henry per phase. I will try to measure it soon. The next model will be considerably better.

4) What is the Wire-Gauge used for the Coils ?, and approximate length of Wire in each Coil ?

The test motor is 12 gauge jacketed wire. About 5.2 meters per coil and the spools average out to about 12x12mm(the test model is very poorly rigged though. Its only the 3rd attempt at a motor design. The next motor under works right now will have 10 gauge jacketed wire. With an orthocyclic winding technique I can get a fill rate of between 1 and 1.2. Not ideal, but not bad

5) What is the approximate average Diameter of your Coils ?

Edit: Put the wrong answer here. The diameter of the test coils is approximately 3cm. So not particularly large. The next version will be larger around and have larger cores.

6) Approximately how many Turns of wire are your Coils made-up of ?

150 per phase. The next version will have in the range of 300-400 per phase. I will probably have to add more turns to compensate for the larger gauge wire.

7) Are Your Coils surrounded by an Iron or Steel Core, or are they partially, or fully, Air-Core ?

The test model has a fully iron core made from transformer iron. Probably just going to use the lowest carbon steel I can find for the next one.

8) Is there any provision for insuring which direction the Motor will start in ?

Yes, it will always push it to the next forward position depending on polarity.

9) Is there any provision for reversing the direction of rotation ?

I was just working on that tonight. See the attached image of my new circuit. I think mechanical DPDT latching relays would be the easiest for me to design for, but perhaps there are better options?

10) Is Motor-Braking desirable ?

The design can definitely brake.

11) How will Flyback from the Motor-Windings be handled ?

I think its all fly-forward in this design.

12) Will the Motor run acceptably with Freewheeling-Diodes across the windings ?

I originally considered fly back circuits, but the design I ended up with does not seem to need them.

I only have my test model to measure from currently, but the model I am working on now will have at least twice the density of coils, far bigger magnets, a larger stator, a larger rotor, etc. I will also be going from 12 gauge insulated wire to 10 gauge(its what I have). The winding fill factor ends up being in the range of 1 to 1.2 with good orthocyclic winding technique, but I would like to eventually get some proper squared motor wire and perhaps get them wound by a company that specializes in that.

 

Please allow me to preface my response with a few disclaimers ........
I am, by no stretch of the imagination, a Motor-Design-Engineer,
I am simply aware of some of the pitfalls of Motor-Design,
and how to avoid blowing things-up, or avoiding creating fires.

You appear to be somewhat serious about making your Motor work,
( which is much more than I can say about 99% of the people coming here with "special" Motor designs ),
and without saying, straight-out, that You can't succeed at this endeavour,
I must warn You that there are at least ~100 ways that this project can easily fail miserably,
despite your best efforts.
Regardless of their simple appearance, Motors are extremely complex devices.

The first thing that You must have in place is a Power-Supply that has built-in, adjustable, Current-Limiting.
The Current must be limited to a level that will not allow exceeding the continuous
Temperature-Rating of the Insulation of the Magnet-Wire that You intend to use.
It is best that this Current-level be limited to roughly ~30 to ~50% of the Wire's Ampacity-rating for
preliminary experimentation.
Much later, ( last ), in the development process,
and, after it is determined that everything is working as expected,
the maximum-Current-level may be gradually increased,
while regularly checking the actual operating Temperatures of the Windings with
a hand-held Infa-Red Temperature-Gun, and/or, a Temperature-Sensor embedded in the Windings.

A cooling-Fan can increase the maximum allowable Current.

The Commutating-Circuitry may also have "active" Current-Limiting for certain Modes or Conditions,
and can also be set up to reduce the Current when overheating is detected.

At low Motor-RPMs, there is reduced, or even almost zero, Back-EMF generated by the Windings.
This virtually turns the Windings into a straight piece of Wire,
which means that the Current in the Windings will go "straight-to-the-Moon" without Current-Limiting.
At higher RPMs, Back-EMF, and Inductance, come into play,
dramatically increasing the Impedance of the Windings,
which tends to continuously increase with additional RPMs.
Therefore, higher-RPM operation tends to require
greater Voltage-levels to maintain adequate Current/Power to generate the required Output-Torque.

The various Magnet-Wire manufacturers usually provide Charts outlining the
maximum recommended Current of their various Insulation-Types
under various conditions and configurations, and, of course,
for each of the various Wire-Gauges that they manufacture.

Zero RPM ...........
which can be considered to be the same as "Brake" or "Locked-Rotor",
will be a condition with the lowest-Voltage, and highest potential-Current/Power-Dissipation.
This is also a condition where there is potentially the least amount of cooling from moving Air,
unless using an externally powered Cooling-Fan.

Reversing ............
by using "Full-Bridge" driving Circuitry,
Reversing is a "no-brainer", no relays required.

This will have to do for a start.

Please correct me if I have misunderstood ........
You appear to have come up with your own special design for a 2-Phase "Stepper-Motor"
which You also deem to be a good "General-Purpose" Motor design.

It is my estimation that You have gained enough of an understanding of Motors
that You may have "run-away-with" this very basic understanding and think that
You have invented a design that has never been seen before.
You might save yourself a lot of time by coming to the realization that
your design is, most emphatically, not new.
This fact is virtually guaranteed.
It's been done more ways than You can possibly imagine,
and these older designs are rarely seen, and have fallen by the wayside,
because there are far superior designs and materials readily available today,
( of course the specific design is dependent upon the requirements of the application ).

If You just want to see if You can make it work, then "more power to ya", and "go for it".

Is it worth it ?, only You can say.
.
.
.

 

Please allow me to preface my response with a few disclaimers ........
I am, by no stretch of the imagination, a Motor-Design-Engineer,
I am simply aware of some of the pitfalls of Motor-Design,
and how to avoid blowing things-up, or avoiding creating fires.

You appear to be somewhat serious about making your Motor work,
( which is much more than I can say about 99% of the people coming here with "special" Motor designs ),
and without saying, straight-out, that You can't succeed at this endeavour,
I must warn You that there are at least ~100 ways that this project can easily fail miserably,
despite your best efforts.
Regardless of their simple appearance, Motors are extremely complex devices.

The first thing that You must have in place is a Power-Supply that has built-in, adjustable, Current-Limiting.
The Current must be limited to a level that will not allow exceeding the continuous
Temperature-Rating of the Insulation of the Magnet-Wire that You intend to use.
It is best that this Current-level be limited to roughly ~30 to ~50% of the Wire's Ampacity-rating for
preliminary experimentation.
Much later, ( last ), in the development process,
and, after it is determined that everything is working as expected,
the maximum-Current-level may be gradually increased,
while regularly checking the actual operating Temperatures of the Windings with
a hand-held Infa-Red Temperature-Gun, and/or, a Temperature-Sensor embedded in the Windings.

A cooling-Fan can increase the maximum allowable Current.

The Commutating-Circuitry may also have "active" Current-Limiting for certain Modes or Conditions,
and can also be set up to reduce the Current when overheating is detected.

At low Motor-RPMs, there is reduced, or even almost zero, Back-EMF generated by the Windings.
This virtually turns the Windings into a straight piece of Wire,
which means that the Current in the Windings will go "straight-to-the-Moon" without Current-Limiting.
At higher RPMs, Back-EMF, and Inductance, come into play,
dramatically increasing the Impedance of the Windings,
which tends to continuously increase with additional RPMs.
Therefore, higher-RPM operation tends to require
greater Voltage-levels to maintain adequate Current/Power to generate the required Output-Torque.

The various Magnet-Wire manufacturers usually provide Charts outlining the
maximum recommended Current of their various Insulation-Types
under various conditions and configurations, and, of course,
for each of the various Wire-Gauges that they manufacture.

Zero RPM ...........
which can be considered to be the same as "Brake" or "Locked-Rotor",
will be a condition with the lowest-Voltage, and highest potential-Current/Power-Dissipation.
This is also a condition where there is potentially the least amount of cooling from moving Air,
unless using an externally powered Cooling-Fan.

Reversing ............
by using "Full-Bridge" driving Circuitry,
Reversing is a "no-brainer", no relays required.

This will have to do for a start.

Please correct me if I have misunderstood ........
You appear to have come up with your own special design for a 2-Phase "Stepper-Motor"
which You also deem to be a good "General-Purpose" Motor design.

It is my estimation that You have gained enough of an understanding of Motors
that You may have "run-away-with" this very basic understanding and think that
You have invented a design that has never been seen before.
You might save yourself a lot of time by coming to the realization that
your design is, most emphatically, not new.
This fact is virtually guaranteed.
It's been done more ways than You can possibly imagine,
and these older designs are rarely seen, and have fallen by the wayside,
because there are far superior designs and materials readily available today,
( of course the specific design is dependent upon the requirements of the application ).

If You just want to see if You can make it work, then "more power to ya", and "go for it".

Is it worth it ?, only You can say.

Autistic fixations aren't really a choice lol. I'm building this thing regardless of it winning any awards. I stumbled into this design to some degree because another design I produced did not work well and I reconfigured it on a whim in a way that works beautifully. I've poured over thousands of patents and can't find anything like it though. That definitely doesn't mean its competitive, but I am certain it will fulfill my needs based on the prototype. I only need it to power a yard vehicle or maybe a small passenger car used as such. The actual function of the motor would be very much governed by the power input type. Pulsed DC can be used but the positional sensors are still absolutely necessary unless you're running a very predictable load. Back EMF isn't really an issue for most uses of this design, but you bring up a very good point about inductance increasing at higher RPMs. I will be using fairly powerful N52 magnets and it will be spinning past a lot of copper.

I'm working on a 3D model of the next prototype and I intend to have cooling built in to it. The more I think about it, the more I think I need to increase the turns in the windings. This design is basically operating at DC resistance so even bumping up to 0.4ohms per phase is not enough. I also think I should at some point make sure the power input can start pulsing when the rotor is stuck in a certain phase since pushing constant full current through it would just oversaturate it anyway. Then again maybe saturation is my friend. There is very little iron in the middle of the current prototype's windings, but I have noticed after connecting to power there is an initial surge as they magnetize but then instead of cooking as expected given the ampacity of the source: they just level out and reduce their draw. I want more iron in the next model, but perhaps I shouldn't add too much?

I like the idea of using the IR gun. I have one or two of them around here somewhere. I also need a better DC voltage source or a good circuit to connect to my batteries(I've been using 4ah lead-acid batteries from medical equipment 3x12V in series but I cut the wires and use wagos so I can connect to 3 different voltages. Even at 6V the rotor turns very forcefully and its extremely makeshift. If it had the circuitry to spin it enough to develop some torque that would help. More turns, denser coils and larger magnets should all increase the output significantly. All I have right now is a lot of 20mm wide X 10mm deep button type N35 magnets to work with and for a 2.8Kw coil at 24V/10A is serious overkill for these magnets so I am eager to see its performance with larger ones. I don't think I can afford them all at one time though so it may take me a few weeks to get them magnets I need. I need to find a reputable dealer that will sell me ACTUAL N52 magnets unlike ebay scammers. I think I already have enough wire, but I need to find steel for the armature. I think solid pieces for the cores will work just fine in this configuration but I need to find something suitable. I think other than that the main thing I need is to find good axle and bearings for it. I should have all the other parts I need.

Thanks for the pointers. Hopefully within the next month here I should have a prototype presentable enough to demonstrate.

 

I need. I need to find a reputable dealer that will sell me ACTUAL N52 magnets unlike ebay scammers

These guys are reputable - https://www.kjmagnetics.com/categor...xUiLx_maHq8CiExPmUgKe3-UlSjlIqycaAtbtEALw_wcB