On Tuesday, August 10, 2021 at 4:20:59 PM UTC+10, boB wrote:
> On Mon, 9 Aug 2021 18:35:55 -0700 (PDT), Anthony William Sloman
> <bill....@ieee.org> wrote: 
> 
> >On Tuesday, August 10, 2021 at 7:39:45 AM UTC+10, boB wrote: 
> >> On Sun, 8 Aug 2021 21:04:49 -0700 (PDT), Anthony William Sloman <bill....@ieee.org> wrote: 
> >> >On Monday, August 9, 2021 at 1:50:37 PM UTC+10, boB wrote: 
> >> >> On Sun, 8 Aug 2021 00:46:06 -0700 (PDT), Anthony William Sloman <bill....@ieee.org> wrote: 
> >> >> >On Sunday, August 8, 2021 at 3:35:37 PM UTC+10, boB wrote: 
> >> >> >> On Fri, 6 Aug 2021 22:25:41 -0700 (PDT), Anthony William Sloman <bill....@ieee.org> wrote: 
> >> >> >> >On Saturday, August 7, 2021 at 1:56:14 PM UTC+10, palli...@gmail.com wrote: 
> > 
> >> >> >> >> > The point is that if the current is zero for one third of the time, 
> >> >> >> >> > positive for one third of the time and negative for one third of the time, 
> > 
> ><snipped Phil being silly> 
> > 
> >> >> >> >The standard modified sine wave has an off time of 33% and doesn't blow up incandescent lamps. 
> >> >> >> 
> >> >> >> At what DC input voltage ? The clamp time will depend on battery 
> >> >> >> voltage to regulate the output to the proper RMS voltage. 
> >> >> > 
> >> >> >Not in a modified sine wave converter. You typically use some other system to fix the output voltage. 
> >> >> > 
> >> >> >You can play with the off-time to regulate the output power, but as you move away from 33%, off time, the third harmonic content in the output goes up. 
> >> >> 
> >> >> I suppose you could design a Modified Square Wave inverter to do that but at some point, it might just be as easy to just make a real sinewave. 
> >> > 
> >> >It isn't. 
> >> 
> >> If you have to regulate the secondary voltage just to keep the MSW at a specific duty cycle, then it isn't too far from making a sinewave in the first place. 
> > 
> 
> >For a rather bizarre value of "isn't too far".
> 2+2 = 5 for large values of 2
> > 
> >> Most of the RE sinewave inverters these days for off-grid use just 
> >> make sinewaves at 20+ kHz and boost by way of a 50/60 Hz transformer. 
> >> > 
> >> >> The old 50Hz/60Hz MSW inverters changed the off/clamp time to regulate 
> >> >> the output voltage based on load and battery voltage. I worked on 
> >> >> them for years. 
> >> > 
> >> >The ones you worked on might have, but it's just one of the options. 
> >> > 
> >> >> Either way, the distortion is bad, around 30% THD but it is measured 
> >> >> slightly different for power waveforms that it is for audio power 
> >> >> amps. Similar though. For inverters, only the first 41 or so 
> >> >> harmonics matter in the THD specs. 
> >> > 
> >> >And the third harmonic matters a lot. Reducing it to zero helps quite a bit. 
> >> 
> >> Just reducing the third harmonic isn't enough for a LOT of things. 
> > 
> >Obviously not. The switches can frequently switch in ten or twenty nsec, so that's harmonics out to above 10MHz. 
> > 
> >You can use an output filter to block most of it.
>
> Yes and that should be a standard part of the output filter if 
> possible. Common mode filtering also but possibly a separate and 
> lesser differential inductance to reduce the capacitance that would 
> otherwise bypass the higher frequencies and let them get through.
> > 
> >Sloman A.W., Buggs P., Molloy J., and Stewart D. &ldquo;A microcontroller-based driver to stabilise the temperature of an optical stage to 1mK in the range 4C to 38C, using a Peltier heat pump and a thermistor sensor&rdquo; Measurement Science and Technology, 7 1653-64 (1996) 
> > 
> >used a pulse-width-modulation system to control the current through the Peltier heat pump, and we looking at something between 20mV and 30mV per degree out of our sensor, so our one millidegree was 20 to 30 uV, so I put a fairly fierce LC filter right on the switched output, which got the worst case ripple down to about 10mV peak to peak at 17.4kHz. Higher frequency components were much more heavily attenuated. You need ferrite beads to catch the fastest stuff. 
> >
> Will try to look at your paper.
>
> >> Might help motors a lot though. 
> > 
> >The interesting question is the frequency where the motor winding stops looking inductive.
>
> I suppose so somewhere above the resonant frequency of the motor?

I was thinking of the parallel capacitance of the motor windings. If the magnetic path through the motor goes though something conductive you've got another way for current to by-pass the windings. None of the motors I worked on - there haven't been many - threw up any problems from that, but I suppose that it is a theoretical possiblity.
 
> >> The sharp edges are very bad on many things. Try one of those 
> >> specific Duty Cycle MSW inverters on a triac light dimmer for 
> >> instance. Or try getting the buzz of the MSW out of your MW radio 
> >> receivers... Or a lot of times even out of a stereo running off of 
> >> that waveform. 
> > 
> >When I was working ins the science workshop at Nijmegen University one of my colleagues had the job of looking at the hash on the mains plugs around the university. 
> >Lots of mains to DC converters sprayed a lot of RF back into the power leads. 
> > 
> >> Just make sinewaves. No excuse these days except for the cheapest of the cheapest small inverters. 
> > 
> >Unfortunately, mass market gear is designed to be very cheap indeed. 
> > 
> >> I still use one in the car just to charge the cell phone or laptop which works OK. 
> > 
> >Of course you do. So does everybody else.
>
> Some do if they know about them. Then again, some say, "what is an  inverter?" 

They may not know what to call it. Even "wall wart" is pretty technical (and it isn't an inverter, though they tend to have one built in).

-- 
Bill Sloman, Sydney

On Mon, 9 Aug 2021 18:35:55 -0700 (PDT), Anthony William Sloman
<bill.sloman@ieee.org> wrote:

>On Tuesday, August 10, 2021 at 7:39:45 AM UTC+10, boB wrote:
>> On Sun, 8 Aug 2021 21:04:49 -0700 (PDT), Anthony William Sloman <bill....@ieee.org> wrote: 
>> >On Monday, August 9, 2021 at 1:50:37 PM UTC+10, boB wrote: 
>> >> On Sun, 8 Aug 2021 00:46:06 -0700 (PDT), Anthony William Sloman <bill....@ieee.org> wrote: 
>> >> >On Sunday, August 8, 2021 at 3:35:37 PM UTC+10, boB wrote: 
>> >> >> On Fri, 6 Aug 2021 22:25:41 -0700 (PDT), Anthony William Sloman  <bill....@ieee.org> wrote: 
>> >> >> >On Saturday, August 7, 2021 at 1:56:14 PM UTC+10, palli...@gmail.com wrote: 
>
>> >> >> >> > The point is that if the current is zero for one third of the time, 
>> >> >> >> > positive for one third of the time and negative for one third of the time, 
>
><snipped Phil being silly>
>
>> >> >> >The standard modified sine wave has an off time of 33% and doesn't blow up incandescent lamps. 
>> >> >> 
>> >> >> At what DC input voltage ? The clamp time will depend on battery 
>> >> >> voltage to regulate the output to the proper RMS voltage. 
>> >> > 
>> >> >Not in a modified sine wave converter. You typically use some other system to fix the output voltage. 
>> >> > 
>> >> >You can play with the off-time to regulate the output power, but as you move away from 33%, off time, the third harmonic content in the output goes up. 
>> >> 
>> >> I suppose you could design a Modified Square Wave inverter to do that but at some point, it might just be as easy to just make a real sinewave. 
>> > 
>> >It isn't.
>>
>> If you have to regulate the secondary voltage just to keep the MSW at  a specific duty cycle, then it isn't too far from making a sinewave in  the first place. 
>

>For a rather bizarre value of "isn't too far".

2+2 = 5  for large values of 2


> 
>> Most of the RE sinewave inverters these days for off-grid use just 
>> make sinewaves at 20+ kHz and boost by way of a 50/60 Hz transformer.
>> > 
>> >> The old 50Hz/60Hz MSW inverters changed the off/clamp time to regulate 
>> >> the output voltage based on load and battery voltage. I worked on 
>> >> them for years. 
>> > 
>> >The ones you worked on might have, but it's just one of the options. 
>> > 
>> >> Either way, the distortion is bad, around 30% THD but it is measured 
>> >> slightly different for power waveforms that it is for audio power 
>> >> amps. Similar though. For inverters, only the first 41 or so 
>> >> harmonics matter in the THD specs. 
>> > 
>> >And the third harmonic matters a lot. Reducing it to zero helps quite a bit.
>>
>> Just reducing the third harmonic isn't enough for a LOT of things. 
>
>Obviously not. The switches can frequently switch in ten or twenty nsec, so that's harmonics out to above 10MHz.
>
>You can use an output filter to block most of it.


Yes and that should be a standard part of the output filter if
possible.  Common mode filtering also but possibly a separate and
lesser differential inductance to reduce the capacitance that would
otherwise bypass the higher freqeuencies and let them get through.



>
>Sloman A.W., Buggs P., Molloy J., and Stewart D. &#4294967295;A microcontroller-based driver to stabilise the temperature of an optical stage to 1mK in the range 4C to 38C, using a Peltier heat pump and a thermistor sensor&#4294967295; Measurement Science and Technology, 7 1653-64 (1996)
>
>used a pulse-width-modulation system to control the current through the Peltier heat pump, and we looking at something between 20mV and 30mV per degree out of our sensor, so our one millidegree was 20 to 30 uV, so I put a fairly fierce LC filter right on the switched output, which got the worst case ripple down to about 10mV peak to peak at 17.4kHz.  Higher frequency components were much more heavily attenuated. You need ferrite beads to catch the fastest stuff.
> 

Will try to look at your paper.

>> Might help motors a lot though. 
>
>The interesting question is the frequency where the motor winding stops looking inductive.

I suppose so somewhere above the resonant frequency of the motor ?


>
>> The sharp edges are very bad on many things. Try one of those 
>> specific Duty Cycle MSW inverters on a triac light dimmer for 
>> instance. Or try getting the buzz of the MSW out of your MW radio 
>> receivers... Or a lot of times even out of a stereo running off of 
>> that waveform. 
>
>When I was working ins the science workshop at Nijmegen University one of my colleagues had the job of looking at the hash on the mains plugs around the university.
>Lots of mains to DC converters sprayed a lot of RF back into the power leads.
> 
>> Just make sinewaves. No excuse these days except for the cheapest of the cheapest small inverters.
>
>Unfortunately, mass market gear is designed to be very cheap indeed.
>
>> I still use one in the car just to  charge the cell phone or laptop which works OK. 
>
>Of course you do. So does everybody else.

Some do if they know about them.  Then again, some say, "what is an
inverter ?"

boB

On Mon, 09 Aug 2021 21:31:17 -0700, jlarkin@highlandsniptechnology.com
wrote:

>On Mon, 09 Aug 2021 14:39:36 -0700, boB <boB@K7IQ.com> wrote:
>
>>On Sun, 8 Aug 2021 21:04:49 -0700 (PDT), Anthony William Sloman
>><bill.sloman@ieee.org> wrote:
>>
>>>On Monday, August 9, 2021 at 1:50:37 PM UTC+10, boB wrote:
>>>> On Sun, 8 Aug 2021 00:46:06 -0700 (PDT), Anthony William Sloman
>>>> <bill....@ieee.org> wrote: 
>>>> 
>>>> >On Sunday, August 8, 2021 at 3:35:37 PM UTC+10, boB wrote: 
>>>> >> On Fri, 6 Aug 2021 22:25:41 -0700 (PDT), Anthony William Sloman 
>>>> >> <bill....@ieee.org> wrote: 
>>>> >> 
>>>> >> >On Saturday, August 7, 2021 at 1:56:14 PM UTC+10, palli...@gmail.com wrote: 
>>>> >> >> The Slowman fool puked: 
>>>> >> >> ===================== 
>>>> >> >> >> 
>>>> >> >> > The point is that if the current is zero for one third of the time, 
>>>> >> >> > positive for one third of the time and negative for one third of the time, 
>>>> >> >> ** Not how real "mod sine" inverters operate. 
>>>> >> >> There are two requirements on such devices: 
>>>> >> >> 
>>>> >> >> 1. Output the normal supply peak voltage. 
>>>> >> >> 
>>>> >> >> 2. Output the normal supply RMS voltage. 
>>>> >> >> 
>>>> >> >> Using these two, the "off" time must be 50%. 
>>>> >> >> 
>>>> >> >> Using only 33% would result in incandescent lamps failing in seconds. 
>>>> >> > 
>>>> >> >The standard modified sine wave has an off time of 33% and doesn't blow up incandescent lamps. 
>>>> >> 
>>>> >> At what DC input voltage ? The clamp time will depend on battery 
>>>> >> voltage to regulate the output to the proper RMS voltage. 
>>>> > 
>>>> >Not in a modified sine wave converter. You typically use some other system to fix the output voltage. 
>>>> > 
>>>> >You can play with the off-time to regulate the output power, but as you move away from 33%, off time, the third harmonic content in the output goes up.
>>>>
>>>> I suppose you could design a Modified Square Wave inverter to do that 
>>>> but at some point, it might just be as easy to just make a real 
>>>> sinewave. 
>>>
>>>It isn't.
>>
>>
>>If you have to regulate the secondary voltage just to keep the MSW at
>>a specific duty cycle, then it isn't too far from making a sinewave in
>>the first place.
>>
>>Most of the RE sinewave inverters these days for off-grid use just
>>make sinewaves at 20+ kHz and boost by way of a 50/60 Hz transformer.
>>
>>
>>> 
>>>> The old 50Hz/60Hz MSW inverters changed the off/clamp time to regulate 
>>>> the output voltage based on load and battery voltage. I worked on 
>>>> them for years. 
>>>
>>>The ones you worked on might have, but it's just one of the options.
>>> 
>>>> Either way, the distortion is bad, around 30% THD but it is measured 
>>>> slightly different for power waveforms that it is for audio power 
>>>> amps. Similar though. For inverters, only the first 41 or so 
>>>> harmonics matter in the THD specs. 
>>>
>>>And the third harmonic matters a lot. Reducing it to zero helps quite a bit.
>>
>>Just reducing the third harmonic isn't enough for a LOT of things.
>>Might help motors a lot though.
>>
>>The sharp edges are very bad on many things.   Try one of those
>>specific Duty Cycle MSW inverters on a triac light dimmer for
>>instance.    Or try getting the buzz of the MSW out of your  MW radio
>>receivers...   Or a lot of times even out of a stereo running off of
>>that waveform.
>>
>>Just make sinewaves.  No exceuse these days except for the cheapest of
>>the cheapest small inverters.  I still use one in the car just to
>>charge the cell phone or laptop which works OK.
>
>
>We are now designing an AC source that will convert 48 DC to AC up to
>240 RMS, 100 watts or so. An H-bridge and a transformer will boost 48
>to 400 DC, and a second H-bridge will make the AC output. That will
>use one small planar transformer and eight d-pak mosfets.
>
>Three of those makes 3-phase AC.

100 watts isn't too bad to use a 60Hz transformer, weight wise.  How
much weight are you saving by going HF ?  I have seen your product
page on this inverter I think but don't know what the market wants for
what you are making.  Price is no object maybe ?

My new inverter does basically what you are doing in that it takes
48VDC and boosts it up to 200 to 300 VDC and makes the two 120V
sinewave on the HV side.

3-phase is nicely done with a fairly simple sync/stacking connection
with 120 degree phase shift.  I assume that is what you do as well.

boB

On Tuesday, August 10, 2021 at 2:31:28 PM UTC+10, jla...@highlandsniptechnology.com wrote:
> On Mon, 09 Aug 2021 14:39:36 -0700, boB <b...@K7IQ.com> wrote: 
> 
> >On Sun, 8 Aug 2021 21:04:49 -0700 (PDT), Anthony William Sloman 
> ><bill....@ieee.org> wrote: 
> > 
> >>On Monday, August 9, 2021 at 1:50:37 PM UTC+10, boB wrote: 
> >>> On Sun, 8 Aug 2021 00:46:06 -0700 (PDT), Anthony William Sloman 
> >>> <bill....@ieee.org> wrote: 
> >>> 
> >>> >On Sunday, August 8, 2021 at 3:35:37 PM UTC+10, boB wrote: 
> >>> >> On Fri, 6 Aug 2021 22:25:41 -0700 (PDT), Anthony William Sloman 
> >>> >> <bill....@ieee.org> wrote: 
> >>> >> 
> >>> >> >On Saturday, August 7, 2021 at 1:56:14 PM UTC+10, palli...@gmail.com wrote: 
> >>> >> >> The Slowman fool puked: 
> >>> >> >> ===================== 
> >>> >> >> >> 
> >>> >> >> > The point is that if the current is zero for one third of the time, 
> >>> >> >> > positive for one third of the time and negative for one third of the time, 
> >>> >> >> ** Not how real "mod sine" inverters operate. 
> >>> >> >> There are two requirements on such devices: 
> >>> >> >> 
> >>> >> >> 1. Output the normal supply peak voltage. 
> >>> >> >> 
> >>> >> >> 2. Output the normal supply RMS voltage. 
> >>> >> >> 
> >>> >> >> Using these two, the "off" time must be 50%. 
> >>> >> >> 
> >>> >> >> Using only 33% would result in incandescent lamps failing in seconds. 
> >>> >> > 
> >>> >> >The standard modified sine wave has an off time of 33% and doesn't blow up incandescent lamps. 
> >>> >> 
> >>> >> At what DC input voltage ? The clamp time will depend on battery 
> >>> >> voltage to regulate the output to the proper RMS voltage. 
> >>> > 
> >>> >Not in a modified sine wave converter. You typically use some other system to fix the output voltage. 
> >>> > 
> >>> >You can play with the off-time to regulate the output power, but as you move away from 33%, off time, the third harmonic content in the output goes up. 
> >>> 
> >>> I suppose you could design a Modified Square Wave inverter to do that 
> >>> but at some point, it might just be as easy to just make a real 
> >>> sinewave. 
> >> 
> >>It isn't. 
> > 
> > 
> >If you have to regulate the secondary voltage just to keep the MSW at 
> >a specific duty cycle, then it isn't too far from making a sinewave in 
> >the first place. 
> > 
> >Most of the RE sinewave inverters these days for off-grid use just 
> >make sinewaves at 20+ kHz and boost by way of a 50/60 Hz transformer. 
> > 
> > 
> >> 
> >>> The old 50Hz/60Hz MSW inverters changed the off/clamp time to regulate 
> >>> the output voltage based on load and battery voltage. I worked on 
> >>> them for years. 
> >> 
> >>The ones you worked on might have, but it's just one of the options. 
> >> 
> >>> Either way, the distortion is bad, around 30% THD but it is measured 
> >>> slightly different for power waveforms that it is for audio power 
> >>> amps. Similar though. For inverters, only the first 41 or so 
> >>> harmonics matter in the THD specs. 
> >> 
> >>And the third harmonic matters a lot. Reducing it to zero helps quite a bit. 
> > 
> >Just reducing the third harmonic isn't enough for a LOT of things. 
> >Might help motors a lot though. 
> > 
> >The sharp edges are very bad on many things. Try one of those 
> >specific Duty Cycle MSW inverters on a triac light dimmer for 
> >instance. Or try getting the buzz of the MSW out of your MW radio 
> >receivers... Or a lot of times even out of a stereo running off of 
> >that waveform. 
> > 
> >Just make sinewaves. No exceuse these days except for the cheapest of 
> >the cheapest small inverters. I still use one in the car just to 
> >charge the cell phone or laptop which works OK.
> We are now designing an AC source that will convert 48 DC to AC up to 
> 240 RMS, 100 watts or so. An H-bridge and a transformer will boost 48 
> to 400 DC, and a second H-bridge will make the AC output. That will 
> use one small planar transformer and eight d-pak mosfets. 
> 
> Three of those makes 3-phase AC. 

But John can't be bothered to tell us how he uses the eight MOSFets to make his approximation to AC. He's done his boasting, but not been specific enough about how he's doing it to earn any admiration.

-- 
Bill Sloman, Sydney

On Mon, 09 Aug 2021 14:39:36 -0700, boB <boB@K7IQ.com> wrote:

>On Sun, 8 Aug 2021 21:04:49 -0700 (PDT), Anthony William Sloman
><bill.sloman@ieee.org> wrote:
>
>>On Monday, August 9, 2021 at 1:50:37 PM UTC+10, boB wrote:
>>> On Sun, 8 Aug 2021 00:46:06 -0700 (PDT), Anthony William Sloman
>>> <bill....@ieee.org> wrote: 
>>> 
>>> >On Sunday, August 8, 2021 at 3:35:37 PM UTC+10, boB wrote: 
>>> >> On Fri, 6 Aug 2021 22:25:41 -0700 (PDT), Anthony William Sloman 
>>> >> <bill....@ieee.org> wrote: 
>>> >> 
>>> >> >On Saturday, August 7, 2021 at 1:56:14 PM UTC+10, palli...@gmail.com wrote: 
>>> >> >> The Slowman fool puked: 
>>> >> >> ===================== 
>>> >> >> >> 
>>> >> >> > The point is that if the current is zero for one third of the time, 
>>> >> >> > positive for one third of the time and negative for one third of the time, 
>>> >> >> ** Not how real "mod sine" inverters operate. 
>>> >> >> There are two requirements on such devices: 
>>> >> >> 
>>> >> >> 1. Output the normal supply peak voltage. 
>>> >> >> 
>>> >> >> 2. Output the normal supply RMS voltage. 
>>> >> >> 
>>> >> >> Using these two, the "off" time must be 50%. 
>>> >> >> 
>>> >> >> Using only 33% would result in incandescent lamps failing in seconds. 
>>> >> > 
>>> >> >The standard modified sine wave has an off time of 33% and doesn't blow up incandescent lamps. 
>>> >> 
>>> >> At what DC input voltage ? The clamp time will depend on battery 
>>> >> voltage to regulate the output to the proper RMS voltage. 
>>> > 
>>> >Not in a modified sine wave converter. You typically use some other system to fix the output voltage. 
>>> > 
>>> >You can play with the off-time to regulate the output power, but as you move away from 33%, off time, the third harmonic content in the output goes up.
>>>
>>> I suppose you could design a Modified Square Wave inverter to do that 
>>> but at some point, it might just be as easy to just make a real 
>>> sinewave. 
>>
>>It isn't.
>
>
>If you have to regulate the secondary voltage just to keep the MSW at
>a specific duty cycle, then it isn't too far from making a sinewave in
>the first place.
>
>Most of the RE sinewave inverters these days for off-grid use just
>make sinewaves at 20+ kHz and boost by way of a 50/60 Hz transformer.
>
>
>> 
>>> The old 50Hz/60Hz MSW inverters changed the off/clamp time to regulate 
>>> the output voltage based on load and battery voltage. I worked on 
>>> them for years. 
>>
>>The ones you worked on might have, but it's just one of the options.
>> 
>>> Either way, the distortion is bad, around 30% THD but it is measured 
>>> slightly different for power waveforms that it is for audio power 
>>> amps. Similar though. For inverters, only the first 41 or so 
>>> harmonics matter in the THD specs. 
>>
>>And the third harmonic matters a lot. Reducing it to zero helps quite a bit.
>
>Just reducing the third harmonic isn't enough for a LOT of things.
>Might help motors a lot though.
>
>The sharp edges are very bad on many things.   Try one of those
>specific Duty Cycle MSW inverters on a triac light dimmer for
>instance.    Or try getting the buzz of the MSW out of your  MW radio
>receivers...   Or a lot of times even out of a stereo running off of
>that waveform.
>
>Just make sinewaves.  No exceuse these days except for the cheapest of
>the cheapest small inverters.  I still use one in the car just to
>charge the cell phone or laptop which works OK.


We are now designing an AC source that will convert 48 DC to AC up to
240 RMS, 100 watts or so. An H-bridge and a transformer will boost 48
to 400 DC, and a second H-bridge will make the AC output. That will
use one small planar transformer and eight d-pak mosfets.

Three of those makes 3-phase AC.




-- 

John Larkin      Highland Technology, Inc

The best designs are necessarily accidental.

On Tuesday, August 10, 2021 at 7:39:45 AM UTC+10, boB wrote:
> On Sun, 8 Aug 2021 21:04:49 -0700 (PDT), Anthony William Sloman <bill....@ieee.org> wrote: 
> >On Monday, August 9, 2021 at 1:50:37 PM UTC+10, boB wrote: 
> >> On Sun, 8 Aug 2021 00:46:06 -0700 (PDT), Anthony William Sloman <bill....@ieee.org> wrote: 
> >> >On Sunday, August 8, 2021 at 3:35:37 PM UTC+10, boB wrote: 
> >> >> On Fri, 6 Aug 2021 22:25:41 -0700 (PDT), Anthony William Sloman  <bill....@ieee.org> wrote: 
> >> >> >On Saturday, August 7, 2021 at 1:56:14 PM UTC+10, palli...@gmail.com wrote: 

> >> >> >> > The point is that if the current is zero for one third of the time, 
> >> >> >> > positive for one third of the time and negative for one third of the time, 

<snipped Phil being silly>

> >> >> >The standard modified sine wave has an off time of 33% and doesn't blow up incandescent lamps. 
> >> >> 
> >> >> At what DC input voltage ? The clamp time will depend on battery 
> >> >> voltage to regulate the output to the proper RMS voltage. 
> >> > 
> >> >Not in a modified sine wave converter. You typically use some other system to fix the output voltage. 
> >> > 
> >> >You can play with the off-time to regulate the output power, but as you move away from 33%, off time, the third harmonic content in the output goes up. 
> >> 
> >> I suppose you could design a Modified Square Wave inverter to do that but at some point, it might just be as easy to just make a real sinewave. 
> > 
> >It isn't.
>
> If you have to regulate the secondary voltage just to keep the MSW at  a specific duty cycle, then it isn't too far from making a sinewave in  the first place. 

For a rather bizarre value of "isn't too far".

> Most of the RE sinewave inverters these days for off-grid use just 
> make sinewaves at 20+ kHz and boost by way of a 50/60 Hz transformer.
> > 
> >> The old 50Hz/60Hz MSW inverters changed the off/clamp time to regulate 
> >> the output voltage based on load and battery voltage. I worked on 
> >> them for years. 
> > 
> >The ones you worked on might have, but it's just one of the options. 
> > 
> >> Either way, the distortion is bad, around 30% THD but it is measured 
> >> slightly different for power waveforms that it is for audio power 
> >> amps. Similar though. For inverters, only the first 41 or so 
> >> harmonics matter in the THD specs. 
> > 
> >And the third harmonic matters a lot. Reducing it to zero helps quite a bit.
>
> Just reducing the third harmonic isn't enough for a LOT of things. 

Obviously not. The switches can frequently switch in ten or twenty nsec, so that's harmonics out to above 10MHz.

You can use an output filter to block most of it.

Sloman A.W., Buggs P., Molloy J., and Stewart D. &ldquo;A microcontroller-based driver to stabilise the temperature of an optical stage to 1mK in the range 4C to 38C, using a Peltier heat pump and a thermistor sensor&rdquo; Measurement Science and Technology, 7 1653-64 (1996)

used a pulse-width-modulation system to control the current through the Peltier heat pump, and we looking at something between 20mV and 30mV per degree out of our sensor, so our one millidegree was 20 to 30 uV, so I put a fairly fierce LC filter right on the switched output, which got the worst case ripple down to about 10mV peak to peak at 17.4kHz.  Higher frequency components were much more heavily attenuated. You need ferrite beads to catch the fastest stuff.

> Might help motors a lot though. 

The interesting question is the frequency where the motor winding stops looking inductive.

> The sharp edges are very bad on many things. Try one of those 
> specific Duty Cycle MSW inverters on a triac light dimmer for 
> instance. Or try getting the buzz of the MSW out of your MW radio 
> receivers... Or a lot of times even out of a stereo running off of 
> that waveform. 

When I was working ins the science workshop at Nijmegen University one of my colleagues had the job of looking at the hash on the mains plugs around the university.
Lots of mains to DC converters sprayed a lot of RF back into the power leads.

> Just make sinewaves. No excuse these days except for the cheapest of the cheapest small inverters.

Unfortunately, mass market gear is designed to be very cheap indeed.

> I still use one in the car just to  charge the cell phone or laptop which works OK. 

Of course you do. So does everybody else.

-- 
Bill Sloman, Sydney

On Sun, 8 Aug 2021 21:04:49 -0700 (PDT), Anthony William Sloman
<bill.sloman@ieee.org> wrote:

>On Monday, August 9, 2021 at 1:50:37 PM UTC+10, boB wrote:
>> On Sun, 8 Aug 2021 00:46:06 -0700 (PDT), Anthony William Sloman
>> <bill....@ieee.org> wrote: 
>> 
>> >On Sunday, August 8, 2021 at 3:35:37 PM UTC+10, boB wrote: 
>> >> On Fri, 6 Aug 2021 22:25:41 -0700 (PDT), Anthony William Sloman 
>> >> <bill....@ieee.org> wrote: 
>> >> 
>> >> >On Saturday, August 7, 2021 at 1:56:14 PM UTC+10, palli...@gmail.com wrote: 
>> >> >> The Slowman fool puked: 
>> >> >> ===================== 
>> >> >> >> 
>> >> >> > The point is that if the current is zero for one third of the time, 
>> >> >> > positive for one third of the time and negative for one third of the time, 
>> >> >> ** Not how real "mod sine" inverters operate. 
>> >> >> There are two requirements on such devices: 
>> >> >> 
>> >> >> 1. Output the normal supply peak voltage. 
>> >> >> 
>> >> >> 2. Output the normal supply RMS voltage. 
>> >> >> 
>> >> >> Using these two, the "off" time must be 50%. 
>> >> >> 
>> >> >> Using only 33% would result in incandescent lamps failing in seconds. 
>> >> > 
>> >> >The standard modified sine wave has an off time of 33% and doesn't blow up incandescent lamps. 
>> >> 
>> >> At what DC input voltage ? The clamp time will depend on battery 
>> >> voltage to regulate the output to the proper RMS voltage. 
>> > 
>> >Not in a modified sine wave converter. You typically use some other system to fix the output voltage. 
>> > 
>> >You can play with the off-time to regulate the output power, but as you move away from 33%, off time, the third harmonic content in the output goes up.
>>
>> I suppose you could design a Modified Square Wave inverter to do that 
>> but at some point, it might just be as easy to just make a real 
>> sinewave. 
>
>It isn't.


If you have to regulate the secondary voltage just to keep the MSW at
a specific duty cycle, then it isn't too far from making a sinewave in
the first place.

Most of the RE sinewave inverters these days for off-grid use just
make sinewaves at 20+ kHz and boost by way of a 50/60 Hz transformer.


> 
>> The old 50Hz/60Hz MSW inverters changed the off/clamp time to regulate 
>> the output voltage based on load and battery voltage. I worked on 
>> them for years. 
>
>The ones you worked on might have, but it's just one of the options.
> 
>> Either way, the distortion is bad, around 30% THD but it is measured 
>> slightly different for power waveforms that it is for audio power 
>> amps. Similar though. For inverters, only the first 41 or so 
>> harmonics matter in the THD specs. 
>
>And the third harmonic matters a lot. Reducing it to zero helps quite a bit.

Just reducing the third harmonic isn't enough for a LOT of things.
Might help motors a lot though.

The sharp edges are very bad on many things.   Try one of those
specific Duty Cycle MSW inverters on a triac light dimmer for
instance.    Or try getting the buzz of the MSW out of your  MW radio
receivers...   Or a lot of times even out of a stereo running off of
that waveform.

Just make sinewaves.  No exceuse these days except for the cheapest of
the cheapest small inverters.  I still use one in the car just to
charge the cell phone or laptop which works OK.

boB





>
>> >This doesn't seem to worry Phil, but it does throw away one of the few good points of a pretty crude system.

On Monday, August 9, 2021 at 1:50:37 PM UTC+10, boB wrote:
> On Sun, 8 Aug 2021 00:46:06 -0700 (PDT), Anthony William Sloman
> <bill....@ieee.org> wrote: 
> 
> >On Sunday, August 8, 2021 at 3:35:37 PM UTC+10, boB wrote: 
> >> On Fri, 6 Aug 2021 22:25:41 -0700 (PDT), Anthony William Sloman 
> >> <bill....@ieee.org> wrote: 
> >> 
> >> >On Saturday, August 7, 2021 at 1:56:14 PM UTC+10, palli...@gmail.com wrote: 
> >> >> The Slowman fool puked: 
> >> >> ===================== 
> >> >> >> 
> >> >> > The point is that if the current is zero for one third of the time, 
> >> >> > positive for one third of the time and negative for one third of the time, 
> >> >> ** Not how real "mod sine" inverters operate. 
> >> >> There are two requirements on such devices: 
> >> >> 
> >> >> 1. Output the normal supply peak voltage. 
> >> >> 
> >> >> 2. Output the normal supply RMS voltage. 
> >> >> 
> >> >> Using these two, the "off" time must be 50%. 
> >> >> 
> >> >> Using only 33% would result in incandescent lamps failing in seconds. 
> >> > 
> >> >The standard modified sine wave has an off time of 33% and doesn't blow up incandescent lamps. 
> >> 
> >> At what DC input voltage ? The clamp time will depend on battery 
> >> voltage to regulate the output to the proper RMS voltage. 
> > 
> >Not in a modified sine wave converter. You typically use some other system to fix the output voltage. 
> > 
> >You can play with the off-time to regulate the output power, but as you move away from 33%, off time, the third harmonic content in the output goes up.
>
> I suppose you could design a Modified Square Wave inverter to do that 
> but at some point, it might just be as easy to just make a real 
> sinewave. 

It isn't.
 
> The old 50Hz/60Hz MSW inverters changed the off/clamp time to regulate 
> the output voltage based on load and battery voltage. I worked on 
> them for years. 

The ones you worked on might have, but it's just one of the options.
 
> Either way, the distortion is bad, around 30% THD but it is measured 
> slightly different for power waveforms that it is for audio power 
> amps. Similar though. For inverters, only the first 41 or so 
> harmonics matter in the THD specs. 

And the third harmonic matters a lot. Reducing it to zero helps quite a bit.

> >This doesn't seem to worry Phil, but it does throw away one of the few good points of a pretty crude system.

-- 
Bill Sloman, Sydney

On Sun, 8 Aug 2021 00:46:06 -0700 (PDT), Anthony William Sloman
<bill.sloman@ieee.org> wrote:

>On Sunday, August 8, 2021 at 3:35:37 PM UTC+10, boB wrote:
>> On Fri, 6 Aug 2021 22:25:41 -0700 (PDT), Anthony William Sloman 
>> <bill....@ieee.org> wrote: 
>> 
>> >On Saturday, August 7, 2021 at 1:56:14 PM UTC+10, palli...@gmail.com wrote: 
>> >> The Slowman fool puked: 
>> >> ===================== 
>> >> >> 
>> >> > The point is that if the current is zero for one third of the time, 
>> >> > positive for one third of the time and negative for one third of the time, 
>> >> ** Not how real "mod sine" inverters operate. 
>> >> There are two requirements on such devices: 
>> >> 
>> >> 1. Output the normal supply peak voltage. 
>> >> 
>> >> 2. Output the normal supply RMS voltage. 
>> >> 
>> >> Using these two, the "off" time must be 50%. 
>> >> 
>> >> Using only 33% would result in incandescent lamps failing in seconds. 
>> > 
>> >The standard modified sine wave has an off time of 33% and doesn't blow up incandescent lamps.
>>
>> At what DC input voltage ? The clamp time will depend on battery 
>> voltage to regulate the output to the proper RMS voltage. 
>
>Not in  a modified sine wave converter. You typically use some other system to fix the output voltage.
>
>You can play with the off-time to regulate the output power, but as you move away from 33%, off time, the third harmonic  content in the output goes up.

I suppose you could design a Modified Square Wave inverter to do that
but at some point, it might just be as easy to just make a real
sinewave.

The old 50Hz/60Hz MSW inverters changed the off/clamp time to regulate
the output voltage based on load and battery voltage.  I worked on
them for years.

Either way, the distortion is bad, around 30% THD  but it is measured
slightly different for power waveforms that it is for audio power
amps.   Similar though.  For inverters, only the first 41 or so
harmonics matter in the THD specs.

boB



>
>This doesn't seem to worry Phil, but it does throw away one of the few good points of a pretty crude system.