Makita (JDM) Vacuum Remote Start

[Bakafish]

I live in Tokyo and have a Japanese domestic market (JDM) shop vacuum. It has a very useful feature where a corded tool, when plugged into the provided power receptacle on the front, will start the vacuum and then it will shut itself down after a 5 second delay. I use several cordless tools though, and although the newest tools from Makita use a ridiculously overpriced Bluetooth system to emulate this behaviour, this vacuum and the tools I use (primarily a Festool Tracksaw which supports an incompatible Bluetooth system anyway) do not have any support for this.

The Goal:
Use a cheap Chinese wireless dongle to remotely start and stop the Vacuum.

The Challenge:
Internally the circuit board that governs the entire operation including the current sensing system is completely potted, so I can't just tap into the existing trigger circuit as I had hoped. All external wiring from the board to the switches and outlet are at line voltage and carry full current. I want to avoid using some kind of switched main outlet as the vacuum can draw as much as 20A when it is supplying power to my bigger saws, and it's just inelegant.

The Proposal:
The outlet on the front (Tokyo is 100v/50Hz) that does the current sensing is live when in the "automatic" mode. It requires ~10-20 watts of current draw before the vacuum will turn on, it turns off after a 5 second delay when the load is removed (this is intentional as it completes the dust extraction.) My thinking is that if I create a dongle that:
1) Plugs into the automatic outlet which provides power
2) When triggered by the remote, puts a purely resistive load, just enough to trigger the vacuum
3) After creating the load required to start the vacuum, it goes open circuit (no load) for ~4 seconds before again drawing current to keep the vacuum on

My thinking is this switching avoids drawing a continuous load and all the heat and safety issues that would result from that.

The Resources:
A Chinese wireless key fob and 4-12vdc based board with small relay (7A @ 120v, would prefer it to be running low voltage...)
A potted 12vdc 0.45A power supply that is low enough draw to not trigger the vacuum when plugged into the automatic outlet
Multimeters, Oscilloscope, soldering station and everything required to make a circuit board.
Easy access to Akihabara for parts

The Request:
I'm a bit of a newb, I understand a lot of the fundamentals but I don't like playing with line voltage without some adult supervision (it really is 100v here, not 110 or 120 here if that matters as far as design.)
My expectation is that there would be a straight forward design that uses a transistor capable of handling 100v that gets switched on by the relay from the remote to drive a simple circuit that shorts across a large current limiting resistor, while a timer (maybe a capacitor?) modulates that connection so it is very momentarily engaged every 3 or 4 seconds. The short across the resistor would be fused so if something went wrong it would just blow 'open' rather than create a smoking ball of death...

Is there something 'off the shelf' that does this, or is there a better approach given the conditions outlined above? I'm hoping to get some help with the design, and learn some basic solid state switching of 'high voltages' along the way. Any of you able to advise?

[Bakafish]

I have a circuit designed around a 7555 timer that pulses a half second (10% duty cycle) every 5 seconds. Will wire that to a small 4A solid state relay tied to a 500 Ohm 50W power resistor. Targeting a 20 watt load every 5 seconds. Will update with a schematic and total cost after the parts come and I verify it works.

Again, any pitfall avoidance advice would be appreciated.

[Renate]

I would not switch in a fake load to activate the current sensing.
I would get a big solid state relay and wire that into the vacuum circuit too.
Even with your load idea you still have to switch AC.

If the thing with the Bluetooth is the bulk of your problem I would make this work with Bluetooth.
A Raspberry Pi Zero W can handle the Bluetooth at the cost of $15 or so.
Yes, you will have to figure out the protocol of what's coming out of the Bluetooth tool.

If you have other things and you want to use a clicker, you could work with that Bluetooth connection too.

Easy access to Akihabara for parts

I'm jealous.

[Bakafish]

I've avoided the whole Bluetooth scam. They use a proprietary protocol, a ridiculously expensive (~$50) dongle, only some tools support it, etc...

Festool tools at least trigger off of the batteries themselves, but like cordless batteries, all these manufacturers maintain walled garden profit centers instead of agreeing to a universal standard and interface. I can't blame them, but it is customer hostile.

As far as wiring into the vacuum circuit, I can hook up a big relay on the mains, but it actually doesn't help with my current workflow. It would need to be set to the "On" mode and the corded tools wouldn't trigger it (or even work until it was switched on.) Currently, the vacuum stays in "Auto" mode and my corded tools share a Festool cord with a "Plug It" bayonet connector. It makes it really easy to grab a tool, plug in the AC quick connect and the hose, then just get work done. Putting an additional switch in front of that is ugly and is just trading off one inconvenience for the other. As of now when I switch to a cordless, I need to go switch the vacuum to manual (which sucks), but if my device works as I proposed I can in-line it with the Festool quick change electrical cord and the vacuum will trigger off of a corded tool, or the remote, all without any direct intervention.

Obviously this is a small nuisance, but reducing friction in workflows really makes the job more pleasant. A new Festool vacuum that does both current sensing and a remote fob start is about $2000 here in Japan, so a few hours and $100 to mod my little Makita to do the same job is a win in my book. (To be complete, Makita sells a $200 Bluetooth based retrofit device in the US that is similar to what I'm trying to build, but they do not make a pure Remote Trigger fob, you need to buy the $50 modules for your tools, again if they support the feature. I assume they know people would just buy one fob and not spend hundreds of dollars on their stupid transmitters.)

[NiHaoMike]

Most likely it's sensing current rather than power, so a motor run capacitor (10uF gives about 30VA) will make a nondissipative "load" to trigger it. Add a 100k or so resistor across it to avoid surprises and use a zero cross SSR to switch it in.

[Renate]

No, I'm suggesting wiring in the new SSR in parallel with the potted blob switch for automatic.
You can reverse engineer the Bluetooth protocol.
It can't be rocket science, probably just GATT.
I've done that with the iRig BlueBoard foot switch which only has user software for Apple so I can do Android.

[Bakafish]

Most likely it's sensing current rather than power, so a motor run capacitor (10uF gives about 30VA) will make a nondissipative "load" to trigger it. Add a 100k or so resistor across it to avoid surprises and use a zero cross SSR to switch it in.

If I understand you correctly, the main advantage to this would be a smaller resistor as it would just be draining the cap between pulses. My concern is that in my testing, it takes a sustained (~20W) load for at least a half second (it may be a full second, I will know when I actually build the circuit and start testing it.) A single 10w bulb isn't enough, but 2 of them will trigger it. If I blip the load too fast it will not trigger, so I think a cap charging will not give a long enough draw to work.

The resistor and SSR can both handle a continuous 20 watt load, assuming I manage the heat (based on the photos I've seen of the Makita version I mentioned above, I think they are just switching in a constant load and dissipating the heat), but if I can pulse it with the timer circuit I can reduce the continuous load on the SSR/resistor, limit the heat generation and hopefully improve reliability.

[Bakafish]

No, I'm suggesting wiring in the new SSR in parallel with the potted blob switch for automatic.
You can reverse engineer the Bluetooth protocol.
It can't be rocket science, probably just GATT.
I've done that with the iRig BlueBoard foot switch which only has user software for Apple so I can do Android.

The CN remote relay unit was $9... I do not want the complexity of Bluetooth at all, none of my units have the transmitters, I do not want to buy them. I want a button on the hose that I can push to make it suck, this achieves that goal.

[NiHaoMike]

If I understand you correctly, the main advantage to this would be a smaller resistor as it would just be draining the cap between pulses. My concern is that in my testing, it takes a sustained (~20W) load for at least a half second (it may be a full second, I will know when I actually build the circuit and start testing it.) A single 10w bulb isn't enough, but 2 of them will trigger it. If I blip the load too fast it will not trigger, so I think a cap charging will not give a long enough draw to work.

A capacitor connected to AC will draw current like a resistor does, but at 90 degrees out of phase so it doesn't dissipate power.

[Bakafish]

A capacitor connected to AC will draw current like a resistor does, but at 90 degrees out of phase so it doesn't dissipate power.

Understood. But you need to drain that charge with a resistor to 'reset' it for the next cycle, right? And unless I'm using some seriously massive (and expensive) cap, I can't imagine it will draw current for the ~500ms required to trigger the sensor. So the only advantage I see is that I can spec a smaller resistor to dissipate the energy stored in the capacitor over the 90% downtime of the duty cycle, but you add an expensive capacitor. As the power resistor I'm using was about $4, and I will be able to adjust the duty cycle using the 555 to ensure it triggers, I still think my design may be preferable. I'm new to this stuff though, so if I'm missing something obvious let me know.

[NiHaoMike]

Understood. But you need to drain that charge with a resistor to 'reset' it for the next cycle, right? And unless I'm using some seriously massive (and expensive) cap, I can't imagine it will draw current for the ~500ms required to trigger the sensor.

The capacitor is constantly charging and discharging when connected to AC. The only purpose of the high value resistor across it is to discharge it when disconnected so it doesn't unexpectedly hold a charge to to zap you when you work on it. Also, motor run capacitors are cheap.

[kripton2035]

some ideas here : http://kripton2035.free.fr/master-slaverepo.html
it's for european 220v system but should be easily transposed.

[Bakafish]

The capacitor is constantly charging and discharging when connected to AC. The only purpose of the high value resistor across it is to discharge it when disconnected so it doesn't unexpectedly hold a charge to to zap you when you work on it. Also, motor run capacitors are cheap.

Oh! I didn't realize that a cap discharges when left in (an AC) circuit, makes sense I guess. I think you are saying that I can just switch a motor capacitor 'in circuit' using the SSR and it will present a load that triggers the sensor. No need for a timer circuit as it will happily stay under load without excessive heat as long as I'm likely to need. Can you give me a suggested Farad rating? I have some big caps I pirated from an old AC I just replaced, I believe they were there for the compressor and cooling fan motors. 

[Bakafish]

some ideas here : http://kripton2035.free.fr/master-slaverepo.html
it's for european 220v system but should be easily transposed.

The current sensing circuit is already built into the vacuum, unfortunately it was totally potted so I can't fiddle with it. I'm trying to create a safe way to remotely trigger it without having to replicate all that functionality. That linked circuit is a really handy design though!

[NiHaoMike]

Oh! I didn't realize that a cap discharges when left in (an AC) circuit, makes sense I guess. I think you are saying that I can just switch a motor capacitor 'in circuit' using the SSR and it will present a load that triggers the sensor. No need for a timer circuit as it will happily stay under load without excessive heat as long as I'm likely to need. Can you give me a suggested Farad rating? I have some big caps I pirated from an old AC I just replaced, I believe they were there for the compressor and cooling fan motors. 

Which values do you have? 7.5uF would be close to the minimum needed for it to work (so I would say you want at least 10uF) and the maximum would be limited by the maximum current of the circuit, far larger than common motor run capacitors. Do make sure it's a motor run capacitor, since motor start capacitors are not rated to withstand voltage for an extended time. (Post a picture if you're not sure.)

[Renate]

The capacitor idea is really clever, but I hate it.
I don't want to put big capacitors across a power line.
We used to do that with electrolytics for fun.

Heck, I don't even want to put resistors across a power line.

[Monkeh]

The capacitor idea is really clever, but I hate it.
I don't want to put big capacitors across a power line.
We used to do that with electrolytics for fun.

Heck, I don't even want to put resistors across a power line.

Putting big capacitors, resistors, and inductors across a power line is what you do every time you use any appliance. Get used to it.

[Bakafish]

Which values do you have? 7.5uF would be close to the minimum needed for it to work (so I would say you want at least 10uF) and the maximum would be limited by the maximum current of the circuit, far larger than common motor run capacitors. Do make sure it's a motor run capacitor, since motor start capacitors are not rated to withstand voltage for an extended time. (Post a picture if you're not sure.)

I have 2 Nichicon 360μF (350μF tested) 85° 250 working voltage units and a single Marcon CE62W 1000μF (950μF tested) 85° 330vdc capacitor. All appear to be electrolitic. My little Peak LCR45 tester could only check their capacitance using DC, they were all under 5% of their indicated values.

From the research you inspired I believe these are intended to be used as 'start' (and filtering) capacitors.

[Bakafish]

I believe that this (or equiv) is the kind of unit you are suggesting?

https://americas.rsdelivers.com/product/kemet/c274ac35100sa0j/kemet-10%ce%bcf-polypropylene-capacitor-pp-470v-ac-5/3778920

[NiHaoMike]

I believe that this (or equiv) is the kind of unit you are suggesting?

https://americas.rsdelivers.com/product/kemet/c274ac35100sa0j/kemet-10%ce%bcf-polypropylene-capacitor-pp-470v-ac-5/3778920

That's one kind, but it's also common to find the kind that's in a metal can, for example this one:
https://smile.amazon.com/MARS-Motors-Armatures-Capacitor-Micro-Farads/dp/B00CD75IPU/

[Bakafish]

I can confirm the motor run capacitor (Kemet 15μF，±5% C878BF35150SA0J) triggered the auto sense outlet as NiHaoMike predicted 

Based on that result, I will create a much simpler setup that utilizes the 7A relay of the chinese remote module.

Once built and tested I will upload a schematic and parts list for Googlers looking for a similar solution.

[Bakafish]

utilize a zero cross SSR to switch it in.

Will using the 7A mechanical relay provided by the remote module cause some issue if it isn't engaging/disengaging at the zero cross point? I'm trying to keep things simple, I have a zero-cross SSR, just seemed redundant if the mechanical relay is already over spec'ed for the purpose. My understanding is that a motor run capacitor should be able to deal with being switched in and out of circuit at any point of the waveform. Am I missing something important?

[NiHaoMike]

15uF is tiny compared to the filter capacitors found in modern power supplies, so the relay shouldn't have any problem with it. If you really want to play it safe, add an inrush limiting NTC in series.

[Bakafish]

The Motor Run Capacitor being driven by the built in relay of the commodity Amazon sourced wireless switch module is working well so far. The final design is rather trivial, but I provide it here for anyone who wants to emulate it (do so at your own risk obviously.)

I did try it with a small NTC I had in my scrap parts bin, but I think the current draw was too small to quickly get it up to temperature and it didn't seem to allow the cap to draw current, so I removed it from the system.



Electrical Components:
Kemet 15μF，±5% Motor Run Capacitor (C878BF35150SA0J)
1W 37KΩ Resistor
1A Fuse
Generic Wireless Relay Remote Control Key Switch Receiver + Transmitter Fob
12VDC-AC adaptor

The key variables are going to be the amount of current you need to simulate in order to trigger the auto-switch on your vacuum and the voltage/frequency of your country. The capacitor I list above works for my particular circumstances 100v/50Hz, which is not standard. An online "Capacitor Impedance Calculator" can be used to select an appropriate value for your needs. As has been emphasized multiple times in the thread, you must use a Motor Run Capacitor that is designed to handle being continuously connected to line voltage. Do not substitute a motor start, or other type of capacitor, as they will likely fill your room with nasty smoke or other bad things™.

The AC adaptor doesn't need to be 12V as these wireless switch modules seem to take a broad range, just make sure it is compatible.

The resistor can also be changed, I used a beefy looking salvaged one I had, it needs to be able to handle the current and voltage of the capacitor you use. As stated earlier in the thread, it is there to drain the cap when it is not in use, so you don't get a nasty surprise should you somehow be exposed to it.

I 3D printed a nice sturdy enclosure that integrated an inline (unswitched) outlet so my wired tools can still utilize the same automatic port without having to mess with disconnecting anything.



I appreciate all the help and ideas from the community! 

[goldserve]

Im looking to make something similar for my Festool Midi vac here in the US. I bought the capacitor to try to see if it will activate the vacuum. Now i'm looking for an outlet box that has 2 outlets and ample room inside to capacitor and other electronics. Has anyone got any good leads on an outlet box with a short cord?