Your thoughts please - Soft-start cct for 240V universal motor

Thread Starter

cognas

Joined Feb 24, 2017
58
cgallery - this thread seems to have died the death, but I'd like to ask if you ever resolved your problem, or what you eventually did. I have exactly the same issue - also primarily with a table saw. My previous (low cost) table saw had gear teeth cut into the motor shaft (1/2" diameter), which meshed with a 2" diameter gear wheel on the saw arbour, and after a couple of years of very little use, it stripped all the teeth off the 2" gear. I have bought another saw table, again low cost, though adequate for my needs - but it suffers exactly the same problem - the saw blade comes up to speed in what seems like about a mirosecond. I'm sure a soft-start of some kind would greatly extend its life. I'd really prefer to build something rather than buy an off-the-shelf solution, and I have a couple ideas (one triac-based, one IGBT-based), but I foresee problems with them both. How did you get around it?



Moderator's Note:
Please don't hijack other member's thread, now you have your own.
This thread was split from --
Modifying router speed control for soft start
https://forum.allaboutcircuits.com/...ed-control-for-soft-start.98637/#post-1100017
 

tcmtech

Joined Nov 4, 2013
2,867
On those type of universal series wound motors a simple high wattage light dimmer will easily work as a speed control or soft start device.

Now as far as those motor gears go they need grease and that's something that very cheap saws don't use much of.

Simply put, you are getting what you paid for. A very cheap poorly made tool and there is no real fix for that but to spend the money and buy a better quality one which had you spent the money you did on two crappy saws on one good one you wouldn't be having this problem. :rolleyes:

Growing up we heated the house with wood and most of it was old lumber so we cut it with circular saws. Dad was cheap. Really cheap so he would by the cheapest crappiest circular saws he could find ($20 - $30 1980's money) and we went through 2 - 4 of them a winter (striped gears or fried motors every time) opposed to having ever bought a good commercial grade saw for $100 - $150 that would have lasted for years. :mad:

That's what I did for ehating when I got my own place set up about 17 years ago. I bought a $180 Craftsman commercial circular saw around 2001 and an AEG 12" industrial chop saw a few years later. Cant break or wear out either for trying. :cool:
 

MaxHeadRoom

Joined Jul 18, 2013
28,619
I think I would investigate making one using a small 8 pin Picmicro.
For material ideas there is the Fairchild AN-3006 (fig 3) and the Picmicro AN958.
The AN958, which includes code, is primarily for a Triac burst firing method to replace a adjustable mechanical thermostat for a heater, which is not suitable for a motor, but there is a Dimmer program version for experimental use based on the same circuit.
Using a marriage of these two sources one should be able to come up with a viable solution.
Max.
 

Thread Starter

cognas

Joined Feb 24, 2017
58
Folks,

I'd like to get the thoughts of somebody who knows more about this stuff than I do.
I had a table saw that stripped the teeth off a gearwheel because it started up too quickly. It had a 240V universal motor that comes up to speed almost instantly. I couldn't get spares, so now I have another table saw, but it has exactly the same start-up characteristic, so to extend its life, I'd like to make a soft-start circuit. I could buy one - but for about twice what I paid for the saw.

I have had two ideas, one using an IGBT and one using a triac, but I foresee issues with both of them.

IGBT : These are just preliminary thoughts - I haven't started to build anything.
ss1.png
I have approximated that the smoothing cap, in order to have any effect, would discharge from 340V (peak rectified mains) to 210V in 15mS. With a motor resistance of 8-ohm (at rest), that gives a capacitor value of around 4700uF. Big, high-voltage caps are pretty pricey. Perhaps I could use a zero-crossing detector and synchronise the mark-space circuit to the mains, and eliminate the smoothing cap.
The mark-space circuit could be made with a uProcessor, or with 555's, op-amps, etc. It would generate a 0:100 mark space, increasing to 100:0 over a second or so. Its frequency would be about 1kHz. The IGBT needs to be driven by a gate driver, but the power supply of the gate driver seems to be quite complex. All the circuits I have looked at have a -ve rail below ground. Is that necessary?
In the ON position, the switch feeds power to the motor, and in the OFF position, it short circuits the motor with the intention of slowing it down quickly. I haven't tried to see if that works yet, but I think it will.

Triac : I bought one of these through Ebay. Switching it on with the pot anticlockwise, then turning it up over a 1 sec period brings the motor up to speed and everything appears to work. I didn't try sawing anything, so I don't know yet how it will stand up to having the motor under load, but as that would only ever happen when the pot is fully clockwise, there should be minimal dissipation in the triac. It occurred to me that I could replace the pot with a digital one, and ramp it up from 0%-100% over a second using some sort of oscillator - a 555 etc. The snag there is that the pot in my little circuit is a 560k lin, where digipots seem to be 200k max. Also, the datasheets I've seen show that they power on at their mid-point value, so would have to be driven down to the zero-value before the mains came on, and I don't think there's a reset pin to do that. Can anybody think of some simpler (eg R-C) circuit to replace the pot, and ramp the triac up from zero to max in a set period (1 second seems about ideal)?
 

ian field

Joined Oct 27, 2012
6,536
Folks,

I'd like to get the thoughts of somebody who knows more about this stuff than I do.
I had a table saw that stripped the teeth off a gearwheel because it started up too quickly. It had a 240V universal motor that comes up to speed almost instantly. I couldn't get spares, so now I have another table saw, but it has exactly the same start-up characteristic, so to extend its life, I'd like to make a soft-start circuit. I could buy one - but for about twice what I paid for the saw.

I have had two ideas, one using an IGBT and one using a triac, but I foresee issues with both of them.

IGBT : These are just preliminary thoughts - I haven't started to build anything.
View attachment 121408
I have approximated that the smoothing cap, in order to have any effect, would discharge from 340V (peak rectified mains) to 210V in 15mS. With a motor resistance of 8-ohm (at rest), that gives a capacitor value of around 4700uF. Big, high-voltage caps are pretty pricey. Perhaps I could use a zero-crossing detector and synchronise the mark-space circuit to the mains, and eliminate the smoothing cap.
The mark-space circuit could be made with a uProcessor, or with 555's, op-amps, etc. It would generate a 0:100 mark space, increasing to 100:0 over a second or so. Its frequency would be about 1kHz. The IGBT needs to be driven by a gate driver, but the power supply of the gate driver seems to be quite complex. All the circuits I have looked at have a -ve rail below ground. Is that necessary?
In the ON position, the switch feeds power to the motor, and in the OFF position, it short circuits the motor with the intention of slowing it down quickly. I haven't tried to see if that works yet, but I think it will.

Triac : I bought one of these through Ebay. Switching it on with the pot anticlockwise, then turning it up over a 1 sec period brings the motor up to speed and everything appears to work. I didn't try sawing anything, so I don't know yet how it will stand up to having the motor under load, but as that would only ever happen when the pot is fully clockwise, there should be minimal dissipation in the triac. It occurred to me that I could replace the pot with a digital one, and ramp it up from 0%-100% over a second using some sort of oscillator - a 555 etc. The snag there is that the pot in my little circuit is a 560k lin, where digipots seem to be 200k max. Also, the datasheets I've seen show that they power on at their mid-point value, so would have to be driven down to the zero-value before the mains came on, and I don't think there's a reset pin to do that. Can anybody think of some simpler (eg R-C) circuit to replace the pot, and ramp the triac up from zero to max in a set period (1 second seems about ideal)?
If you told us more about you mark/space circuit, we could probably figure out where to put the soft start capacitor.

Or you could just stick an inrush limiting NTC thermistor in series with the AC feed.
 

JohnInTX

Joined Jun 26, 2012
4,787
Merged hijacked and duplicate thread here.
Please start only one thread per question and don't post your questions in someone else's thread. If the questions are related, post a link to the related thread.
Thanks and welcome to AAC!
 

Thread Starter

cognas

Joined Feb 24, 2017
58
If you told us more about you mark/space circuit, we could probably figure out where to put the soft start capacitor.
As I said at the beginning, none of this is built yet - it's all just thoughts on paper, and as I also said, "The mark-space circuit could be made with a uProcessor, or with 555's, op-amps, etc. It would generate a 0:100 mark space, increasing to 100:0 over a second or so." I can't tell you its circuit diagram because I haven't decided yet how to make it. I'm not asking about the nitty-gritty at this point - just some advice on what method would be best.

Or you could just stick an inrush limiting NTC thermistor in series with the AC feed.
That's something I hadn't considered. Would it fall to very near zero resistance? The motor is about 700W.
 

Thread Starter

cognas

Joined Feb 24, 2017
58
Merged hijacked and duplicate thread here.
Please start only one thread per question and don't post your questions in someone else's thread. If the questions are related, post a link to the related thread.
Thanks and welcome to AAC!
Pardon?
What other thread are you referring to? Somebody else asked the same question, but the thread was old and had died, so I put a question on it, asking its initiator simply if he had resolved his problem. I thought it probable that I wouldn't get an answer, because the most recent update was in Nov 2013. Then on this present thread, I discussed my own problem, and the ideas I had had. I didn't think they were duplicates at all.

PS. I have just checked that thread, and the question I asked has not been posted (awaiting moderation), so I really don't know what you're talking about.
 

tcmtech

Joined Nov 4, 2013
2,867
As I said at the beginning, none of this is built yet - it's all just thoughts on paper, and as I also said, "The mark-space circuit could be made with a uProcessor, or with 555's, op-amps, etc. It would generate a 0:100 mark space, increasing to 100:0 over a second or so." I can't tell you its circuit diagram because I haven't decided yet how to make it. I'm not asking about the nitty-gritty at this point - just some advice on what method would be best.
You're making it way overly complicated for what a simple analog circuit with a few basic components can easily do.

Brush type universal electric motors like what lightweight power tools use are simple and robust devices that do not need any degree of high precision control to make them work in any number of ways. For them basic phase angle fired voltage/current regulation is more than sufficient.

Find a speed control trigger assy for a power drill or such variable speed power tool of similar wattage rating and replace the trigger potentiometer with a rotary type and y you have a heavy duty soft start system.

Or just use a 1000 watt lamp dimmer and skip the power tool speed controller hacking altogether.
 

tcmtech

Joined Nov 4, 2013
2,867
....Or use a light Dimmer rated at 10amp.

If you use a rotary pot with switch buit in, set the off to minimum speed.
Pretty sure we are going to get ignored for suggesting using the practical sense approach to this idea here.

A heavy duty phase angle fired power control (incandescent light dimmer) is too simple easy and cheap therefore it can't possibly work. :rolleyes:
 

JohnInTX

Joined Jun 26, 2012
4,787
@cognas It is preferable not to add to a dead thread. While the new thread was more specific to your current question we didn't want members to be replying to two separate threads on similar subjects - that is why it got flagged.
No worries.
 

Thread Starter

cognas

Joined Feb 24, 2017
58
Y

Find a speed control trigger assy for a power drill or such variable speed power tool of similar wattage rating and replace the trigger potentiometer with a rotary type and y you have a heavy duty soft start system.

Or just use a 1000 watt lamp dimmer and skip the power tool speed controller hacking altogether.
But I already have one, and gave a link to it in my opening post. Did you not see it? Then I discussed replacing the pot with a digital pot (that won't work. I have found they don't like more than 5V on their terminals), and I asked what sort of circuit I could use to replace the pot with.
 

Thread Starter

cognas

Joined Feb 24, 2017
58
Pretty sure we are going to get ignored for suggesting using the practical sense approach to this idea here.
Ignoring people who try, even misguidedly, to help, is not my style. Your suggestion is not at all misguided.

A heavy duty phase angle fired power control (incandescent light dimmer) is too simple easy and cheap therefore it can't possibly work. :rolleyes:
It DOES work. I already said that. Now, if you have a suggestion of some simple circuit to replace the 560k pot, that would be most helpful.
Actually, that isn't a fair question unless you know the circuit diagram of the device, so I'll take it apart this evening and work out its schematic, then I'll post it.
 

Thread Starter

cognas

Joined Feb 24, 2017
58
@cognas It is preferable not to add to a dead thread. While the new thread was more specific to your current question we didn't want members to be replying to two separate threads on similar subjects - that is why it got flagged.
No worries.
But my addition to that (old) thread related directly to that thread. It asked the initiator if he had resolved his problem. There was no solution to it in his thread, and he had allowed it to go cold, but in the interrim, he may have solved it in some way after some time, and omitted to update his thread. He may not still be around, who knows? However, if you are the moderator on whose auspices my question is waiting, I think it would be worth posting it. Before attempting to post it, I tried to contact him directly through his name, as that is possible in some forums, but I couldn't work out how to do it here. If my question is posted, he should get an e-mail notification, and if he answers with a solution, then his solution may well be mine, as the issue is identical. If he gave me a solution, I would update my own thread with the finding. I wouldn't simply abandon it.
 

MaxHeadRoom

Joined Jul 18, 2013
28,619
But I already have one, and gave a link to it in my opening post. Did you not see it? Then I discussed replacing the pot with a digital pot (that won't work. I have found they don't like more than 5V on their terminals), and I asked what sort of circuit I could use to replace the pot with.
I have been mulling over trying one or both the circuits outlined in my post #3, but it should be easy to adapt one to allow the Triac circuit to ramp up when ever the power is first applied, a pot would just be nuisance to operate over a period of time IMO.
I am doing this as time allows so I cannot promise a definite time or outcome.
Max.
 

ian field

Joined Oct 27, 2012
6,536
As I said at the beginning, none of this is built yet - it's all just thoughts on paper, and as I also said, "The mark-space circuit could be made with a uProcessor, or with 555's, op-amps, etc. It would generate a 0:100 mark space, increasing to 100:0 over a second or so." I can't tell you its circuit diagram because I haven't decided yet how to make it. I'm not asking about the nitty-gritty at this point - just some advice on what method would be best.


.
There are online archives of old TV schematics - around the 80s, many TVs had discrete component SMPSUs, a lot of them had soft start.

It could be as simple as studying a few schematics and adapting the concept.
 

Thread Starter

cognas

Joined Feb 24, 2017
58
I have been mulling over trying one or both the circuits outlined in my post #3, but it should be easy to adapt one to allow the Triac circuit to ramp up when ever the power is first applied, a pot would just be nuisance to operate over a period of time IMO.
I am doing this as time allows so I cannot promise a definite time or outcome.
Max.
Thanks for your attention to this, Max, however, I have made some discoveries. Please read my post below.
 

Thread Starter

cognas

Joined Feb 24, 2017
58
Folks, some observations :

1. The switch arrangement to stop the motor quickly by shorting it doesn't work. It seems to take about the same amount of time to stop when it is shorted. I may arrange a solenoid operated disk brake against the saw blade under the table.

2. tcmtech said: "A [...] (incandescent light dimmer) is too simple easy and cheap therefore it can't possibly work." It turns out he was right. I thought it worth the experiment, as the unit was so cheap I didn't stand to lose much. The first time I tried it, I ramped up the motor speed fairly slowly and it worked. I reasoned that the triac-based circuit would only ever be used under no load while the saw was speeding up, and any time it was under load (cutting wood), the triac would be in full conduction, so wouldn't be dissipating much heat. However, when I tried it again today, I ramped up the speed a bit quicker - and the dimmer circuit popped. Oh, well - back to the drawing board.

Of the two circuit types I mentioned when opening this thread, I think I'll go with the IGBT variant, not because I think it better in any way, but because I can see how I might control it with low-voltage electronics. I think I'll use an 8-pin AVR processor, and I'll use a zero crossing detector to synchronise the ON-pulses with the mains so I can do away with the smoothing capacitor. I was originally going to use one so that I could run the mark-space circuit asynchronously with the mains.
ss2.png
An asynchronous pulse might come at point A or B or C - so it would be necesary to smooth the rectified mains, but big capacitors like that are so expensive that I'd prefer to avoid them. If I synchronise the pulses, they will all come at point A, getting longer and longer, till the IGBT is in full conduction.

If I used the Triac-based idea, the circuit is well known. Here's one I just pinched off the web.
ss3.png
I feel in my bones that this would be a more elegant solution, but I simply can't envisage a simple electronic replacement for the pot, which has fairly high voltages on its terminals.

I'd like to thank those who contributed to this thread, but I think it can be closed at this point. The design will probably take me ages, and I may need help before it's done - but that will be the subject of another thread.
 

Thread Starter

cognas

Joined Feb 24, 2017
58
There are online archives of old TV schematics - around the 80s, many TVs had discrete component SMPSUs, a lot of them had soft start.

It could be as simple as studying a few schematics and adapting the concept.
Your input is appreciated, Ian. I also saw soft start for amplifiers - but I think the requirements to soft start a TV or amp will be quite different from soft-starting a strongly inductive load (a motor).
 
Top