How to stop a hall throttle at rest putting out 0.8v?

[jrm7997]

Hello all

I have just read an extensive post on the hall throttle by Tommycat. Like every post I have found over the last couple days, it mentions a hall throttle (HT) putting out a small amount of output even when off or at rest. But, I have not seen a way to stop the output.

My problem is that I am trying to control a BLDC hoverboard motor using the HT. I am going to use two motors and two HT. While I have been testing and figuring this out I realized that the HTs small output signal its enough to tell the control board to turn on the motor. In other words, the motor is always turning. Even with the throttle completely off. This is a bench test with the motor off the ground so it might not turn if there is weight on it but I cant stop it by hand. Its also problematic in that it will also use up battery power from the time its turned on to being turned off. Even if the motor is just sitting there.

So my question is, how do i stop the small amount of "leaking" signal to my control board. Any suggestions would be appreciated and thanks in advance.

Attached is a schematic of what I am doing (symbols are not all right but you'll get the idea).

Here are links to the board and HT that I am using .

Throttle

https://www.amazon.com/Keenso-Throttles-Control-Throttle-Electric/dp/B07YXVRQW3

Control Board

https://www.aliexpress.com/item/32998115260.html?spm=a2g0o.productlist.0.0.140e293dpZ9CIn&algo_pvid=8af7d640-0045-4bc8-9b3e-4bdabcf23d76&algo_expid=8af7d640-0045-4bc8-9b3e-4bdabcf23d76-14&btsid=0b0a556e16132453242064126e2108&ws_ab_test=searchweb0_0,searchweb201602_,searchweb201603_

[Tommycat]

Hi jrm7997,
And welcome to the forum! :-)

Controllers seem to most often come in two flavors for throttles. Those that work with hall sensored throttles, and those that work with potentiometer type throttles.
Hall sensor type are bountiful and inexpensive. Pot type... not so much. Although you can certainly make an inexpensive one (IE: not store bought) if so desired.
As you've read, hall's output ~.8-4.2vdc, while pots put out 0-5vdc. I've seen many that wanted to use their expensive pot throttles on a hall controller because of it's well liked power curve.
But your the first to want to do the opposite as your controller specs indicate throttle operation .1-5vdc.

Just wanted to mention using a potentiometer for throttle control as an option.

To your inquiry.
As the SS49E hall sensor used in your existing throttle can't directly output less than .8vdc due to it's electronic components. Either by manipulating it's input voltage, grounding, or mechanically.
A work around may be to install a diode in series with the throttle's output wire which would typically produce a .7vdc voltage drop which may be enough. (looking for a diode that would work in such a low current range...)
Or a resistor, also in series with the throttle's output, sized for the needed voltage drop may drop you in the required voltage range. (perhaps a combination of the two.)

Both these suggestions would lower the WOT (wide open Throttle) voltage by the same amount, but for your proposed application it may not be noticeable, and perhaps even desirable...

I have not tested both these theories yet but will do so this week, unless you try and verify the results first. ;-)

Looking forward to your build with great interest! Any information on your parts and pieces would be welcomed. (IE: that hoverboard must be quite powerful...?)




Regards,
T.C.

[AZguy]

For a "simple" solution an NPN transistor in an "emitter follower" configuration with a low enough Re will drop it down to 0.2V or so... Re can't be too low that it draws to much power from the supply though

For a more "exotic" but precise solution I'd go with a simple opamp circuit... If you want to go down that road I'll draw it up..

[Tommycat]

My browser translator seems to be ineffective on that link. ;-) Any chance to get a layman's explanation? (keeping in mind the OP would need to have less than point one vdc closed to 5vdc WOT.)

[AZguy]

Search "emitter follower" on wikipedia and that will take you to the page

I understand that it will drop even above 0.8V and at full twist but that doesn't mean it can't be made to work, it all depends. It's a better approach than just a diode as you can go with a lower resistor to keep the drop more consistent across the range

The best way to go and if the full range is needed then an opamp circuit is the way to go... but it's a bit more complicated... not a lot but I have no idea what skills the OP has or what he's willing to tackle... if he's interested in the opamp approach I'll knock out a good circuit in less than five minutes but won't waste my time if there's no chance he wants to go down that road...

[Tommycat]

Verified that the diode solution works, but as expected it will drop the WOT voltage.

Data points:

Using a 4.84vdc supply. 1K resistor for controller load. ***

Base line= .85 closed/ 4.06vdc WOT.
1 diode= .16vdc closed/2.9vdc WOT.
2 diodes in series= 0vdc closed/ 2.27vdc WOT.

This is just for proof of concept. This may not give you the WOT you may want, as this is about a typical 1/2 throttle position.

Had a more difficult time with resistors as the amount needed seem to choke the voltage down to much and snuffed out the operation. ***But this really depends on your controller's internal circuitry resistance to the hall sensor signal input.***

If you do testing with either, be sure to:

Keep amperage draw of your 5vdc voltage source to throttle below 10mA. ( .01 amps) Typically the hall sensor will draw around 6.5mA. Monitor frequently! Especially after changes...

Just wanted to mention using a potentiometer for throttle control as an option.

[Tommycat]

Something to watch whilst waiting on input... ;-)

[jrm7997]

Good Morning gents

I did go with the hall throttles based on price and a mistake. I'd never heard of a hall throttle until I got them thinking they were potentiometer throttles. once I had them and tested them (no resistance WTF?) I realized they were something new to me. However they work well as an input to the MC I am using except for the 0.85v. I'm pretty sure they will work for my purpose even if they will move as soon as they are turned on or the weight might stop them.

Tommycat. Will the hoverboard motors be big/strong enough to pull an RV6 around? I don't know but I'm having fun trying. This is actually the second version. The first was a mobility chair but it wasn't heavy enough and the wheels skipped. So i traded it for another mobility chair that will probable be version number 3 after this one. Unless this one does the job.

AZguy, i looked up the follower you mentioned (see attachment along with a photo of the prototype I'm working on) and would be interested in trying it out. It definitely sounds counter intuitive but with my little bit of electronics knowledge that's how I have always found electronics. confusing but it works. :)

Actually, I have a pretty good knowledge of electronics. I am an electrician turned teacher. I taught electronics for about 5 years 25 years ago. I have a bunch of 3904 transistors, two TIP31c and access to a very good electronics shop in Langley BC.

From Wikipidia

In electronics, a common collector amplifier (also known as an emitter follower) is one of three basic single-stage bipolar junction transistor (BJT) amplifier topologies, typically used as a voltage buffer.

In this circuit the base terminal of the transistor serves as the input, the emitter is the output, and the collector is common to both (for example, it may be tied to ground reference or a power supply rail), hence its name. The analogous field-effect transistor circuit is the common drain amplifier and the analogous tube circuit is the cathode follower.


I cant say I understand all of this. Vin would be my Hall in. V out would be my speed control to the MC board. But V+ could be my 5V board power or (collector is common to both (for example, it may be tied to ground reference or a power supply rail), hence its name) to ground?

If i have my 5V board power to V+, my hall input to Vin then the way I see this is the Vin would control the V out. R_E would reduce the 0.85v Vin affect on Vout? What R value would you suggest for R_E?


Thanks for helping out. I need to go turn the heat on the garage so I can work out there for a bit today. I have to wire the two throttles in and Ill be able to test it with both motors and throttles. It worked on the bench so ..... :) Once I hear from you about the emitter follower Ill wire one up and give it a shot.

Thanks

Tommycat. There is a big formation group flying Vans products in the pacific northwest. I think they are the blackjacks. I've seen them at Oshkosh, Sun and Fun and at the Arlington WA airshow. Well organized and fun to watch. Once i retire I'm thinking about getting into some formation flying. We have a small group up here called the Poppy Flight . A remembrance day group.

[jrm7997]

I was just thinking I could bench test this myself. I might have to jury rig a power source for the lower voltage but should be ok. I’ll let you know what I find

[AZguy]

Re depends on the 5V source and what it's capable of but if using standard 1/4W resistor then 100ohms would be a good place to start as long as the +5V can source 40-50mA. You want it small in order to keep the drop reasonably consistent - it won't be the full 0.85, more like ~0.65 at the low end so that's another factor. If you use a "darlington pair" instead of the normal NPN it will drop ~1.25V or so and then it would make it so you'd have to crack the throttle a bit before seeing anything go to the ESC. Since it sound like you are handy enough with soldering, etc. I'll throw a opamp circuit at you later if you like - it would look something like this:



I'm a pilot and a really big fan of those RV's =]

[jrm7997]

That looks easy enough to breadboard. I have a good bench power supply that can do all this. BTW, while i think about this, a small company called evo-in-motion makes a very slick little breadboard power supply. https://www.eimtechnology.com/product-mego When i was teaching they gave me one as a sample.

Just to clarify a couple thigs. All the resistors above are 100 ohms? vin is base, vout is emitter and 5v is the collector? Thanks i just don't like the smell of blown transistors. :)

[AZguy]

No the resistors are not all the same - this was just the basic topology... 100 was for the emitter follower

Since you are interested I'll work them out, likely have them today, this one's easy

If you want it adjustable, in particular for the offset (but also the "gain") I'll add optional potentiometers

I can make sure all the parts are available at digi-key or mouser too... it'll be cheap

[Tommycat]

Sharp looking tester. Variable output, snap on integration. But best of all, overload and short circuit protection! Found for <$50 U.S on Amazon...

AZguy, do you have a parts list so I can poke around and see what I have in my kit, or have to order...?

[Tommycat]

I like to use a small USB (5vdc) rechargeable back-up/charging battery that has a switch for bench testing. Get close and the bugs worked out using this first before using you controller regulated power which typically only produces <100mA.

Best results I've gotten so far is with a Darlington circuit using (2) PN2222 transistors with a 100K resistor between the base of the second transistor to batt negative. .07vdc closed @ 5.83mA and 3.41vdc WOT @ 5.76mA. With 5.1vdc supply input. Note: no resistance on output. But when I was previously testing for typical controller input resistance, I would use a 1K resistor on the output. It would drop my .85vdc unloaded sensor output to .8vdc, which is what my controller would pull it down too.

I fear that the magic of electronics comes at a slight voltage drop cost. So a boost in input voltage might be needed for full power output range? But perhaps good enough for testing at this time...?
Looking forward to AZguy's Opto circuit. I'm just a pure novice, but having fun learning. Sorry about the probable unnecessary warnings lol. But it took me just 2 hours to toast my new testing breadboard's power supply.

I was wondering if tug wheel traction would be an issue. Does your mission include portability, or perhaps lack of hanger space?





Thanks for the heads up on the formation groups... will look them up. Do I see a smoke mod in your future? ;-) I would speculate your a member of another great forum group, VAF!

[jrm7997]

See my comment above about the breadboard power supply but i like your usb idea. I could work up here in the warmth instead of having to heat up the garage each time i go out there.

Good advice on testing before throttling :) and I wont know how much a voltage drop will affect its usability until I load it up with some weight. It might work fine or i might be into version #3 the full mobility chair. FYI, if that one fails I'm keeping my eyes out for a snow blower, lawn tractor or ????? D8 :)

Definitely a Vans fan but I bought I didn't build so i usually go there for the "off topic" comment on the bottom of the first page. Todays took me a second but i got it. For AZguy so you know what we are talking about go to the bottom of this page https://vansairforce.net/

No smoke. I gave that up 20 years ago. :)

[Tommycat]

Found the Blackjacks, but it looks like their web site is down. Or bad link perhaps.

https://vansairforce.net/patches.htm

But most importantly, you have to check out this thread on VAF! Tons of great ideas and information from lads who have been there.( no D8s suggested, but I've seen mention of electric golf carts now and then that sounds interesting. :-)
Check out the joystick link for sure.

GO HERE NOW!


Thumb's up on the "off topics" LOL, sometimes I never get em....

[Tommycat]

Some inspiration as provided by the thread above...

The wheel "capture" mechanism is neat!

[AZguy]

This gets you very close for 0.8 to 4.2 in

Resistors need to be tweaked for different input range

The MCP6041 opamp is available in DIP (easy to solder to or breadboard) for <$0.70... nice general purpose for up to 6V with rail-to-rail in and out

If you'd like the simulation I can provide

If I were going to put potentiometers for fine tuning I'd put a 5K in between R1 and R2 with the wiper going to the opamp and R3 and I'd put another in series with R3 (connect the wiper to one end)

[pismocycler]

Thanks AZ. I think I will change the topography and insert a silicon diode to produce an offset voltage to the op amp of about 0.6-0.7V (between a feedback resistor and ground). The diode will reduce the net offset to 0.18 - 0.28 V. Gain is not critical so about 1.1 is OK. I want to keep pots out of the circuit so I can surface mount the diode, resistors, and op amp. Then splice the board into the throttle wire harness. Might also consider a jumper to bypass the circuit back to stock if needed. But thansk for the idea in the first place. Simple and a lot cheaper than buying a BAC 4000/8000 just to tune the throttle.

[AZguy]

I would avoid using the diode and just adjust the resistors

The op-amp has far more precision than the diode and especially over temperature where the diode will drift - not that it really matters since we're talking 100mV kind of range - mostly just that it's unnecessary since you can get there just adjusting resistors...

The OP's requirement was fulfilled by a simple two values of two resistors which is just sort of elegant more than anything else but fiddling values will get you there, just maybe not with that elegance...

[pismocycler]

AZ
Not too nit too much. The diode is forward biased by the current flowing in the feedback resistors, so low current draw. I imagine the offset compensation resistor divider needs about 10X more current to avoid impacting the gain. In the diode circuit offset voltage compensation is coarse at best, I just want it below 0.6 V to avoid the controller lockout. Neither of these issues are critical so to each his own.

[AZguy]

A local place carries some decent power supplies... easy for me since I can just drive the five miles to pick them up, shipping might kill the prices

I've got one of these for general purpose:

https://www.circuitspecialists.com/0-30V_0-5A_bench_power_supply_with_60_memories.html

I've got several others but that's a good one for the price, not quite as accurate as they say but both current and voltage are within a couple of digits worst case, most of the time within one digit... easy to adjust and enough voltages and current for 90+% of bench top stuff... has constant current so will charge batteries but you need another supply in series to get up to the bike battery voltages

[jrm7997]

Sorry I've not replied. Its been too cold to work in the garage and I seem to have picked up a pretty wicked cold. Gotta tell you I'm wondering how I could have got a cold with all the washing, disinfecting, masking etc. But i'm glad I didn't have to do the nose swab Covid test. Been there done that and don't want to do it again. I got the gargle tests this time and came up negative. I got the test just in case. Just a cold :)

I have been following that tug thread. Its where I got the idea to keep my eyes out for a busted snow blower. If i can get this one to pull well enough I'll definitely be looking into that wheel grab. It looks pretty slick.

I looked up that power supply. That's not a bad deal with a lot of options. But I think i have myself covered for the little i do. I took a picture of the two power supply's i have. The one on the left works well, 30vdc and I think its good for 2 amps. The one on the right is a throw away I got back into working condition. Everything works and I love two things about it. First the isolated 110vac. Just a tad safer. Second you cant break it and it puts out 10 amps. Oh and third, if you have a dead car battery you can start your car with it :) Just kidding. They are build like a brick sh** house so keep it around.

I breadboarded the op amp you sent. Unfortunately I only had one op amp (T082acp) and a mish mash of 50K and 25K trim pots to put it together with. I did have two 22K resistors though. But i think with too many jumpers and places for errors it didn't work. Could also be the op amp i have :) Because I've been sick I have not gone out but will pick up the appropriate parts to try your cct with. Thanks BTW i appreciate it.

As for the hover board I got her all wired up a few days back and it didn't work. Worked on the bench but not after wiring the two controllers together. I didn't put much time into it because it was cold out there and my brain really couldn't get into it. Its going to start raining here in a day or so and should warm up. At least enough that the little electric heater can keep it warm enough.

So ill keep you updated.

[73Eldo]

That is a really cool power supply! I wonder why there were not more like that. Lots of features in one box for a general repair or tinker.

[AZguy]

The 082 (you meant TL082?) doesn't have rail-to-rail in or out - you pretty much need that for this application. You could do it with a rail-to-rail out only but it would require more resistors - trying to keep it simple

[jrm7997]

I forgot I was going to mention shipping (comment in your power supply post). Shipping to or within Canada is an absolute fricken joke. Canada post is great but the cost of shipping is ten times or more. I once shipped a center console from a 2004 Chevy truck to a guy in the south eastern USA. Shipping from Vancouver was going to be around $300. I shipped it from Point Roberts WA for less than $100.

What we usually find is that shipping is generally free within the USA. So i have my shipments go to a mail box in Point Roberts, drive across the border and bring it back. If its under $300 they don't usually charge taxes or duties. Unfortunately with Covid and the borders being closed I can't use Point Roberts. I even have some Honda CT90 parts sitting in Point Roberts since March last tear that i cant get too. :(

So a lot of my buying has been toned down a lot. Just waiting for this too end so I can start RVing and everything else.

[jrm7997]

Well I'm about to order the parts. Having a bitch of a time finding the op amp. One of the stores was closed the other wanted $5 for an equivalent and so on. Found out last night that the closed store got bought out and has moved. Just called and they can get MCP6041 in a couple weeks. I have one more option left but I've just been getting a busy signal.

Azguy i was rereading this and noticed the comment that you also fly and are in Arizona. Anywhere near Shea and Hwy 51? My brother in law in that area. We go down pretty regularly and one of my favorite areas to visit. :) what do yo ufly and out of where?