My Motors got me to Tap into Radiant Energy

[Ein~+ein]

You are not right feeling this thread is dead. It is the preliminary foundation for education preparing memebrs for motor design.

@JohnStone/UFOPolitics: For those of us new here, perhaps you might want to sum up the consensus among contributors to this thread and where you see it heading. If radiant energy is nothing more than the energy of electromagnetic waves, which particular waves is UFOPolitics referring to (light, heat, radio, etc), what's the effect and how do you rule out other possibilities (mains power surges, spikes, etc)?

[JohnStone]

hello,

First of all, great work to UFOpolitics and all those who were able to replicate. Does not look as easy as it seems..
I have done a couple of oscillating circuit with all the mosfet and diode as mentionned but the best I got was was 60w input to barely light a 23watt CFL.....
I have been searching for the schematic (circuit) of the dual N-channel mosfet and p-channel mosfet. I was not able to find it on this web site or IN UFO picture database. Is it possible to post it ? or show me a link to where about it is. Thanks

Joel

As far I remember the push pull stage was an option for possible enhancements. But P-FETs suffer from serious drawbacks. For serious use N-FETs are recommended afor high and low side. Drivers are cheap in current times. We possibly go to push pull stages if required - no requirements up to now.
This thraad led to understand radiant in small extent - education! No commercial use intended.
JS

[JohnStone]

Hi John,
What is the purpose having multiple signal. You are sending a signal when the previous one is already on high... wouldn't that prevent radiant from coming in ?
This thread seems to be dead !!

J

It is a requirement for testing motors at the other thread of UFO. (4 pol stator -> 4 channel PWM -> delayed) Sorry for misuse of this thread and confusion.
I felt that electronic discussion could fit more in this area.


You are not right feeling this thread is dead. It is the preliminary foundation for education preparing memebrs for motor design. Hence it is the solid ground for the other thread.
JS

[Lightworker1]

Thanks a million, Indiana Boys
regarding

http://octopart.com/partsearch#search/requestData&q=SFH617A

many regards
light

[lerameur]

Hi John,
What is the purpose having multiple signal. You are sending a signal when the previous one is already on high... wouldn't that prevent radiant from coming in ?
This thread seems to be dead !!

J

[JohnStone]

Programming delayed signals

Hi friends!
regarding the the generation of 4 or 6 signals for UFO motor I'd like to propose 2 sequences for Arduino. Unfortunately I will not be able to program myself hence I ask any of you to continue.

Requirements:
1. Signal (A) with adjustable duty cycle and frequency
2. Adjustable delay for subsequent signals
2. Signal B being a copy of A but delayed
4. Signal C being a copy of B but delayed
5. signal D being a copy of C but delayed

For Cornboy (MAG3) - sequence continued:
6. Signal E being a copy of D but delayed
7. Signal F being a copy of E but delayed

Programming approach:

a)

Unfortunately ardiuno has only one 16 bit timer hence we can not use one individual one for every output stream.
The most elegant way would be to feed a data array.

The idea behind is to disassemble the sequence to a plurality of time slots while a single time slot starts at a transition of any signal and ends at subsequent transitition of any signal (see distance of dotted lines).

The timeslots will be represented by timer values in the the data array.
To any timeslot there are associated two bit patterns while a bit represents an output channel at an output port. One bit pattern represents the output state and the other represents a mask determining bits not to modify. Hence we can select 4 or 6 bit out of 8 bit of an output byte by bool calculation.

This sequence will run very fast as timer interrupt sequence only.
At any timer interrupt the timer will be fed with the next value out of the array and at same time the outputs (4 or 6 bit) will be transfered to the output ports.

The drawback is that any change needs to be done by sophisticated calculation of new timer values.
The calculation may take some time but furtunately it will not affect signal generation at all.

There are some alerts for programmers:

• There could be an issue if two transitions of different signals are very close to each other.
• Depending on time delay and ON-time there are changes in bit patterns necessary as well.

b)

Much more slower but easier to program - usage of SW timers.
A main signal (A) is being generated and all other signals will be a coppy of A - but delayed.

A HW main clock counts a set of timers (decrement) in the timer interrupt.
The main loop will perform some tasks if any of timers is 0.

Main Signal A:

• SW timer A will be loaded alternately with high and low time after being zero. At same time the main output bit for signal A will be inverted.

Service for signal B:

• The output value for signal A - if changed - needs to be shifted in a byte in order to control sequence for Signal B.
• Load delay timer for B timer with signal delay and enable counting

Slave Signal B:
If delay timer B = 0 then

• Disable this delay timer from further decrement
• Get bit value out of shift register B and forward it to output bit for signal B

Service for signal C:

• The output value for signal B - if changed - needs to be shifted in a byte in order to control sequence for Signal C.
• Load delay timer for C timer with signal delay and enable counting

...........

Hint:

• The shift registers are necessary in order to deal with times for high time and delay time being similar or crossing. This resolves issue from method a)
• If shift registers not being preloaded at initialisation sequences B...D will start dealyed. foll signal state wil occure after after 24 cycles of A.

-----------------------------------
Command interface:

• Please expand thos4 4 control switches from NICO's program to 6 switches in order to add delay control.
• Please prepare program for controlling values from PC. it will ease later procedures considerably. I can provide a a draft program for command interpreter.

Keep it running
JohnS

[lerameur]

N and P channel circuit

hello,

First of all, great work to UFOpolitics and all those who were able to replicate. Does not look as easy as it seems..
I have done a couple of oscillating circuit with all the mosfet and diode as mentionned but the best I got was was 60w input to barely light a 23watt CFL.....
I have been searching for the schematic (circuit) of the dual N-channel mosfet and p-channel mosfet. I was not able to find it on this web site or IN UFO picture database. Is it possible to post it ? or show me a link to where about it is. Thanks

Joel

[iankoglin]

Hi Ian, So we could get a Monster driver PCB made here? so when they say per service, You will get 50-300pcs, 4-layer PCB boards per service @$0.31 ??? Is that a mistake? Would it be a 10x10cm roughly?

G'day sampojo
I have not used them but I have bought products from them on EBay
One of their product description is here
2LAYER GREEN PCB 5CM X 10CM MAX [IM120418002] - $22.00 :

It looks like you get 10 PCB's for $20
You have to provide the Circuit they will build/make them I am sure that you need to know what the circuit you are giving them will work as you want it to prior to getting them to manufacture them

I am sure it does not have components loaded But there again I do not know exactly what they are providing
Description
Overview

This service provides quick and high quality PCB fabrication experience for worldwide hobbyists.

You will get 10 pcs, 1-layer or 2-layer, 100% e-tested green PCB boards per service. If you need more than 10 boards, you may add an integral multiple quantity to shopping cart.



I am intending to get my smaller machines running first using boards I make my self and when I succeed I will then try them

Kindest Regards

Kogs still at it

[sampojo]

Small Batch PCBs

Spelling mistake
G'day John
I today received an email from Bellevue2046 I purchased some small circuit boards like voltage reduction 40 down to 12v and Square wave oscillators and they were very good and cheap.
The email was soliciting business from me by way of offering me a transistor tester I looked up their Webb page and saw that they do small runs PCB's here is their Webb page here

Small Batches PCB :
I thought you may like to look them up
l

Hi Ian, So we could get a Monster driver PCB made here? so when they say per service, You will get 50-300pcs, 4-layer PCB boards per service @$0.31 ??? Is that a mistake? Would it be a 10x10cm roughly?

[JohnStone]

.....

Small Batches PCB :
I thought you may like to look them up.....l

Thanks for this hint. Very interesting source for lots of devices and custom PCB as well.
Unfortunately no real source for Europe. To the price you need to add transport - on top 20% customs - on top 19% VAT. They want to prevent American products to enter our market - and vice versa of course

All those being located in Noth America - visit the site!
JS

[iankoglin]

Spelling mistake

Unfortunately they were not interesed at all.
Got severe flu - makes me silent :-(
rgds John

G'day John
I today received an email from Bellevue2046 I purchased some small circuit boards like voltage reduction 40down to 12v and Square wave oscillators and they were very good and cheap.
The email was soliciting business from me by way of offering me a transistor tester I looked up their Webb page and saw that they do small runs PCB's here is their Webb page here

Small Batches PCB :
I thought you may like to look them up

I thank you for all the time and effort you are putting to help us with the Monster driver and Tutorials I really appreciate you for what I am learning
I also purchased 20 Infineon IPW60R041C6's

Kindest regards


Still trying to be helpful

[JohnStone]

.....

Renesas - Super Junction MOSFET
......

ST Microelectronics - MDmesh MOSFET
......

Thanks for hints:
Comparing data:
(rough overview - I did not dig into minute details)

Infineon: ...IPW60R041C6
ST:...........STY145N65M5
Renesas: ...RJK60S8DPK-M0

Voltage at drain: .....I 650V /.. S 710V / ..R 600V
rise time: ...............I 10ns / ..S 11ns / ..R 56ns
fall time: ................I 7ns / ...S 82ns / ..R 64ns
gate capacitance: ...I 6.5nF / S 18.5nF / R 5.3nF
current max.: .........I 77.5A / S 138A / ...R 55A
current realistic: .....I 49A / S 87A / .......R 34A
RDS-on: ................I 41m / S 12m / R 45mOhm

Discussion:
- ST FET was built to have slower turn off time in order to require less protection means. We require to have as fast switching as possible BUT these are theoretical values. In practical use other influences will override it in many cases.
- In genereal: Power FETs are built from a plurality of tiny FET cells connectd in parallel. Therefore high current -> more FET cells -> more gate capacitance. Take i.e. two Infineon FETs in parallel and you have about same capacitance like ST.
- Similar with RDS-on: two Infineon in parallel -> 20mOhm while one ST 12mOhm (but ST is here superiour) Practical use will be twice RDS-on - compute this for heat calcualtion.

Summed up:
- Give Infineon and ST a try but not Renesas (Japan owns other superiour qualifications - but none for high end FETs)
- Estimate roughly 1/2 max. current / twice RDS-on / 2...4 times switching time as realistic values for practical use.
- Please understand we refer here to high end and power applications where we suffer on severe heat losses. We strongly need to munimize them. For lower current applications there are lots of other and cheaper FETs available.


BTW: I just reviewed CREE SiC FETs again. They are superiour at voltage above 600V and still FAST!! But none with RDS-on below 80mOhm. Hence those FETs from Infineon and ST discussed above are recommended as current technology for our application.
JS

[prochiro]

Yes, Robert is fine.

@UFOPolitics
Yes, he is fine. I think spending this cold winter without much heat made California look good.

I am progressing with the Monster driver. What I am doing is slowly adding fet units to the system until I can get enough watts to do the job. I calculated that 36 volts and an average of 30 amps is 1080 watts. This is the target power I (think)??? which will do the job and not heat the fets. What I want to do is get enough so there is only a few degrees above ambient temp to start with and hit 1080 watts. I have some more diodes and fets coming . In a few days, I will be populating the 5th and 6th fet board which has two fets per board. Yes, I am slow but will never stop. I am building another small office now but work on electrics at nite. JohnStone has given us the best of circuits and we must just get enough fets behind it to run with little or no heat. Bro Donald drew a revised position of components and is less of a issue to build and repair, but the same circuit. THANKS GUYS...

Dana

[Ufopolitics]

Thanks Dana!

Hello UFOPolitics

This is probably the quickest intro and circuit related to motor control for the Triac.

http://sg.renesas.com/media/applications/motor_control/triacdrive/TriacAppnote.pdf

Dana

Hello Dana,

Thanks my friend!

On the other topic...


When are we going to see that Imperial being oulsed by the Monster driver my Dear Friend>??!!


Pleease We are eagerly waiting!

And by the way, send my regards to Bob French, long time I do not see Him around...hope everything is fine...

and Regards


Ufopolitics


Edit 1:
After I finished writing to you this post...I went to check my email...and guess what?..a mail from Mr Bob French...

He is fine, moving to CA...

Regards again

[Ufopolitics]

Thanks John!

Hello John,

No problems at all, is always great to know advantages-disadvantages...

I am sure the Monster driver would give a nice fight to the Beast...


Thanks!


Regards



Ufopolitics