mcu circuit problem

Go To Last Post
68 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hello all , I am new to use A T 9 0 P W M
my circuit meet problem ,hope some one can help me .

attached the related schematic , the PB0,PB1 & out0 ,out1 of mcu are two pairs of PWM ports,the 17th and 18th feet is AVCC and AGND . the A01, A02 ,ADC2, Adc3,ADC6 are all ADC ports. when I powered the MCU by ISP( con6) ,(not connected device to AC220V) the both two PWM work good in testing ;
But when I connect the board to power AC220V,(not connected to ISP programer) there are no PWM signal output, but the voltage of C22 and C23 is near 3.6V in testing, I feel strange and can not understand what happened. could you please give me some tips? thanks a lot.

Attachment(s): 

Last Edited: Sun. May 3, 2009 - 10:40 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Here's a tip. If you want to be silly enough to power a circuit like this at least have the decency to properly label the AVR i/o pins in the schematic.

Watch you don't leave the ISP wired to the board when powering it with mains otherwise you're in for a harsh lesson.

(maybe you didnt' connect VCC to the mcu)

oddbudman

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Please stop experimenting with this ciruit
and 230 Volts immediately. Otherwise we
probably will not see many postings from
you here !

The look how other people power thei AVR and
how an 7805 is used properly.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks Oddbudman and Ossi.

Its hard to write text on the component, so I attached a new picture with all ports description.

the J3 solded on PCB . and the ADC ports connected to other parts on board to test the voltage. the VCC is connected as I tested voltage on C23 is about 3.6V.

May I ask Ossi , where can I find the shematic of good ways to power AVR and how to use 7805 properly? my device must connect to 220V AC at last.
Thank you very much.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

where can I find the shematic of good ways to power AVR and how to use 7805 properly? my device must connect to 220V AC at last.

As you don't seem to know what you are doing why not just get a "wall wart" (say 6V) and then use the regulator beyond that?

As others have said DO NOT make direct connections to 220V as you are in serious danger of being killed.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks clawson, I have no choice but to connect to 220V AC , as the main part of device must work on condition of 220VAC or 110V AC, the MCU just used as control purpose.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Whoa! That circuit will kill somebody...

To anybody: As others have suggested, please use a proper transformer for your projects! Mine is a regulated 9V "wall wart". It's a really cheap life insurance.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks Tuomas , man will not touch the device except in test.

the PWM signal will sent to half bridge driver to drive power switch mosfet under about 400V , the picture just show part of schematic

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

So what most likely be the problem?
1. power of VCC
2. disturbed by other circuit.
3. ADC problem
4. Software problem?
Unfortunately, I still not found.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

As stated above, having a circuit tied directly to the Mains is potentially lethal.

Your Resistor zener first stage voltage regulator does not provide any isolation from the Mains.

A small power transformer from 120/240 to 12 V, then to the bridge rectifier, then to the 3-terminal linear regulator, would be MUCH safer.

That said, I do not see that your Reset\ is pulled high. I do not see a 0.1 uC and 1uF cap across the uC V+ to Gnd pins as by-pass caps.

I would test the circuit with a 9 V battery feeding the linear regulator to get the uC part of the circuit up and running.

Then decide if you wish to use an isolated power supply or a floating supply.

Isolated supplies are the norm for safety reasons. You need to be very careful when doing floating circuits. It is easy to hurt or kill yourself or others with floating circuits.

JC

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The 200k resistor in front of the regulator does not give you much current to work with. Perhaps Vcc is not stable as a result.

Tom Pappano
Tulsa, Oklahoma

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You should add a charge capacitor to the primary side of the volatage regulator.

And since your circuit isn't galvanic isolated from the mains power, for testing purposes insert a 230VAC mains isolation transformer and plug the mains socket into this and you should be safe.

Last Edited: Sun. May 3, 2009 - 10:15 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

This is really dangerous. And why a bridge rectifier? A single diode will still give enough voltage to kill you.

Stealing Proteus doesn't make you an engineer.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Before working any further with your circuit I would suggest reading some application notes about non-isolated power supplies:

Daycounter, Inc. - Transformerless AC to DC power Supply http://www.daycounter.com/Circui...
Microchip AN954 - Transformerless Power Supplies: Resistive and Capacitive http://ww1.microchip.com/downloa...
Microchip TB008 - Transformerless Power Supply http://ww1.microchip.com/downloa...
Supertex AN-D30 - Off-Line 5.0V Output Non-Isolated Linear Regulator http://www.supertex.com/pdf/app_...
A Capacitor-Fed, Voltage-Step-Down, Single-Phase, Non-Isolated Rectifier http://www.hamill.co.uk/pdfs/acf...
ON Semiconductor AND8146/D - High Current LED - Capacitive Drop Drive http://www.onsemi.com/pub/Collat...
Transformerless Power Supply http://www.zen22142.zen.co.uk/Ci...
Sixerdoodle Electronics - AVR Switch http://www.jenrathbun.com/Electr...

[Edit (more application notes added from STMicroelectronics ):
STMicroelectronics AN1476 - LOW-COST POWER SUPPLY FOR HOME APPLIANCES http://www.st.com/stonline/produ...
STMicroelectronics AN2300 - An alternative solution to Capacitive power supply using Buck converter based on VIPer12A http://www.st.com/stonline/produ...

Last Edited: Mon. May 4, 2009 - 12:55 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
The 200k resistor in front of the regulator does not give you much current to work with. Perhaps Vcc is not stable as a result.

That was my first reaction...still, this circuit is so scary...what is the ultimate purpose? What does the device do?

John

Just some guy

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi

Quote:
my device must connect to 220V AC at last
How did you land this job?
Quote:
I have no choice but to connect to 220V AC
Why, a simple 'No' & pass the job onto a professional
Quote:
man will not touch the device except in test
How do you figure that out when it break down later

I do not know what background you have
By judging your responses, it show clearly that you have a very limited knowledge of electronics & the safety of electrical theory.

But this does concern other members in this forum who want to keep you alive at the end of the day.
They are offering advices ensuring that you work in a safe environment.
But if you chose not to follow then we cannot be held responsible for any damage or death you cause to yourself.

Ken

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks Doc JC, I ve tried to connect the 4.5V battery to the input pin of regulator ( plug connect to AC220V at same time) and found PWM work good. so I think problem from VCC,may be it is not steady enough,although I test there are 3.6V on C23.

And thanks TOM, yes I guess may be the current is too small, but what I feel strange is there are 3.6V on C23. if current not enough I think the voltage should be lower,do you think so?

And thanks ArnoldB, this is only the part of circuits related to mcu. I ve said most of other components( not show on picture) need to be run in high voltage, so the isolation of power for mcu will not protect me much .

Thanks so much for AndersAnd, yes, I use isolated transformer of 220V for test. I connect a 10u/25V cap to input end of regulator but still have the problem.

and thanks for your application notes, that is really helpful, but I heared use capacitor to supply power will be more dangerous, the L and N of AC power input seem can not be swap in connection.

And thanks johnrk, it is used for lamp .

At last thanks pykedgew, thank you and all friends' remind. I do not like to work in dangerous too, hope I will still alive tomorrow.

Best wishes to everyone! thanks all your help again.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

For safety, you should be in an "earth-free" area, with all the mains-supplied test equipment supplied from an isolation transformer, as well as the DUT. If your employer has not provided such a facility, both he and you are breaking the law. That would be the case in the UK, at any rate.

Leon

Leon Heller G1HSM

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

If you want to control mains-connected lamps
I would use opto-isolated sold-state relays
ore a similar technique to isolate the
high-voltage part from the control part.
That makes life a lot easier.

And only if I want to produce thousands of circuits
I would eliminate the isolation in order
to reduce cost.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

are there any good way to make the VCC more steady? I change the Rvcc to 100K ( V3.6 on C23), but the mcu still not work good.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

thanks Leon and Ossi again, I must say control communication is opto-isolated, so that is safe to user.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You might want to put some bulk capacitance on the input to the regulator - 220uF might be a good start. I designed light dimmers professionally for many years - when testing, I'd not use a design like you have - way too dangerous. I'd use optocouplers then when I'd sorted out the code then move towards having a 'hot' circuit. My 'scope probes had many burn marks on them!

Try reverse engineering a commercial product to see how they do it - might be worth your life. Also, ST (www.st.com) have a few example designs of light dimmers with everything 'hot'. A bit of research might save a few electrocutions.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

iamhere wrote:
thanks Leon and Ossi again, I must say control communication is opto-isolated, so that is safe to user.

I don't see any isolation in the schematic!

Leon

Leon Heller G1HSM

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

;-)

There are pointy haired bald people.
Time flies when you have a bad prescaler selected.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

iamhere wrote:
are there any good way to make the VCC more steady?

Read Microchip AN954 and STMicroelectronics AN1476 and design a transformerless power supply similar to the ones suggested in these application notes.
There's both a Resistive Power Supply and a Capacitive Power Supply in both application notes.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks AndersAnd again, I ve studied the Capacitive Power Supply application notes, but what I confused is the electric potential , the mcu can not be separate with main load circuit, the electric potential of mcu may too high compare to GND of main load. will the mcu be damaged when connect to main load? I must say the mcu need to control the main load and test ADC information from main load.

attached picture.

Attachment(s): 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Let us analyse the electric potential of above circuit, if the L=200V ,N=0V, then Vout=200V, Vd2=195V, if the right end of D2(195v) connect to GND of main circuit(0V), what will happened? if the right end of D2 not connected to the GND of main circuit, how can we test the ADC of main LOAD?

Last Edited: Tue. May 5, 2009 - 09:10 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You have to account for minimum and maximum current into the circuit.

This is obviously a very cost-sensitive application, but designing for direct line power is not trivial. Power line transients will kill the design you've presented, and of course you must provide adequate isolation so as not to injure the user.

I would not try to develop this while line powered. It's not necessary, and it's significantly dangerous. Just power it from a simple power supply, and get the code right.
Treat the power supply and interface to the triac as a separate exercise, and use programmed chips there.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

If I connect a 0.1u Capacitor parallelly to Rvcc, will it add current supply ability?

Attachment(s): 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

It is inadvisable for hardware beginners to do this sort of design, which unfortunaly is often done to keep the price of products down.
I would suggest that if it has to be done, that it be done at low AC voltage initially Ie. use a transformer with secondary volatge of say 12V. When the software works, get a hardware persons to do the remaining part of the design.

MOC3020 optocoupled TRIAC drivers do a great job of switching AC.

DO NOT READ PAST THIS IF YOU ARE INEXPERIENCED OR THE NERVOUS TYPE

On the other hand, if the 240VAC connects straight to the bridge & some dropping resistor of say 30-40 K is used, you are really quite safe if you connect your self on the microcontroller side of the 30-40 K resistor. I DID NOT SAY THAT YOU WONT GET A SHOCK, BUT IT WILL NOT KILL YOU (UNLESS YOU HAVE A HEART CONDITIONS OR YOU TRY AND DO THIS ON THE TOP OF A LADDER ETC).
The current will be limited to about 10 mA and is not likely to cause ventricular fibrilation.
Remember "It is the volts that Jolts, but the mills that Kills"

Any way if you are lucky you will work in a organization where the OH&S people( the kill joys) will
stop you working on such equipment.

It is however a fact that most servicemen face it daily in just about every piece of equipment that they
have to work on.

If you know what you are doing it is quite OK, if not give the job to someone else!

Lee de Vries

Charles Darwin, Lord Kelvin & Murphy are always lurking about!
Lee -.-
Riddle me this...How did the serpent move around before the fall?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Iamhere - did you try my suggestion? Why are you just blindly trying anything?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

It's getting worse & am feeling very very sick.
Cannot really look at this as it is a hit & miss arrangement.

As someone said "SCARY" & still is.

Notice the black mark on the cat nose.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I hope his company has product liability insurance, they are probably going to need it.

Leon

Leon Heller G1HSM

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I believe iamhere might turn into iwashere pretty soon. :)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Jay - someone just had to say it!!!!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hello everyone, glad to back here alive, thanks for all your attention and help. I replace the Rvcc to 20K and C22 to 47u, it is working now, but power waste too much in Rvcc. Also I get a isolated power supply module which will not waste much of energy , but it cost about Euro1.5, I hope I can limit all the costs of device under euro 6, so may be it is a little expensive. I'd like to know if you have other good ways to power the mcu, with input AC 85-265V or DC140-340V; output 5V or 3.6V, 20ma; power consume less than 0.5W; and cost under 1 euro. I must say capacitive way is not acceptable as the GND of mcu must connected to - of bridge in main circuit.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

In any proper electronics project, the power supply usually counts for about 25%+ of the cost of the device. As projects get more complicated, so does it's power supply.

Do yourself a favor and just buy a ready made 220v -> 5v power supply module, at least for the 5V logic, then route the 220v to your 220v devices properly. Just the routing properly part will probably take you a couple of hours.

Don't rush things, you will only get accidents/problems, and in the end your design will probably have to be redone completely by someone else after your employer gives you your notice.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
I must say capacitive way is not acceptable as the GND of mcu must connected to - of bridge in main circuit.

You may still use a capacitive method. The RC ballast network just goes on the input side of the bridge. A shunt regulator like a zener should be place across the output side of the bridge.

Tom Pappano
Tulsa, Oklahoma

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Oh, Tom bring me good information again, thank you very much. could you show me the shematic?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Change resistor "N1" to a 150 ohm 2-watt metal oxide

In series with resistor "N1", insert a .68 uf/630 volt *metallized polypropylene* capacitor.

Instead of resistor "LOAD", place a 220 uf capacitor.

Replace resistor "Rvcc" with a direct connection.

Connect the *bottom end* of capacitor "Cvcc" to ground.

You may need to increase the voltage rating of "DV0" a little so that the input to the regulator stays above the minimum allowed, to allow for input ripple.

Tom Pappano
Tulsa, Oklahoma

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

thanks Tom again, but the "LOAD" here is not a resistor, it is a symbol of main load in circuit, the mcu will control the "LOAD" and test ADC from "LOAD" , the "LOAD" consume about 80W.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
the "LOAD" here is not a resistor, it is a symbol of main load in circuit, the mcu will control the "LOAD" and test ADC from "LOAD" , the "LOAD" consume about 80W.

Well, *that* is a new wrinkle! What I described will only supply a couple dozen milliamps, not drive an 80 watt mystery load. We simply must have more useful information about exactly *what* you are trying to do, *how* you want to do it, and *why*.

Tom Pappano
Tulsa, Oklahoma

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Iamhere - seems like you're trying to 'design by proxy'- getting us to design a circuit so you can make money. If you're designing a circuit for sale in the EU, you'll have to comply with CE - this is going to give you more problems than you're currently facing. Mains harmonics will be be an issue if you're running some form of power conversion at 80W.

Nevertheless, if you want a wide range, low loss power supply for low current, I can suggest a simple circuit - but I'll leave you to figure out the schematic.

Basically you have a high voltage mosfet that is biased on. A circuit consisting of a couple of transistors and a zener diode turns the mosfet off when the input voltage exceeds a certain value (as set by the zener diode). The circuit is fed by the raw full wave rectified AC. Therefore the mosfet is on for the first few degrees of the AC waveform then turns off past this then back on again for the last few degrees. The output of the mosfet goes into an electrolytic capacitor for smoothing then into your regulator. Low power and near lossless - just what you wanted. Downside is that it takes around 10 components - the most expensive part being the high voltage mosfet, the rest being resistors, small signal transistors and a zener.

Remember it's one thing to quote a dollar price but that means little unless you specify a quantity. There's a big difference in making 10 items and 100,000 items.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks TOM, you advise me the following circuit, right?

In 5V mcu power supply circuit, the GND of mcu connect to N of input power, right?

and the GND of mcu need to connect to the GND of "LOAD"' in main circuit so it can test ADC ,no problem?

then , the the N of input power will connect to GND of LOAD in main circuit, THAT IS THE PROBLEM!!!!!!!!!!!!!!!

Attachment(s): 

Last Edited: Thu. May 7, 2009 - 09:22 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks Kartman, you suggest way seems interesting, if you have time ,could you please show me the schematic ? I can not understand very well a new circuit not experienced, thank you so much.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hello every one, I have new design of power supply.

please give your suggestion.

the Q1 will stop work when the pfc running.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Kartman's idea is a good one, especially since it can work downstream of the bridge with pulsating DC.

But still,

We (or at least "I") simply must have more useful information about exactly *what* you are trying to do, *how* you want to do it, and *why*.

And now appears "PFC". Private First Class? Peripheral Faraday Conflapulator? Power Factor Canoodler?

Tom Pappano
Tulsa, Oklahoma

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Why not just use the design suggested by Supertex I posted at page 1?
Supertex AN-D30 - Off-Line 5.0V Output Non-Isolated Linear Regulator http://www.supertex.com/pdf/app_...
This is a design using a N-Channel Depletion-Mode (normally-on) Vertical DMOS FETs.
Using a Depletion-Mode (normally-on) MOSFET instead of the much more commn Enhancement-Mode (normally-off) MOSFET makes this circuit simple, only need one Zener, a bias resistor and a bulk capacitor. And you don't have to connect any high voltage high ohm resitors to the high DC voltage at the rectifier bridge.
This solution also makes cooling simple as only the FET in TO-2220 house needs cooling by a heat sink (disspipates 6.1 W in the shown example).
Please note they used a wrong symbol for the MOSFET in the circuit. The symbol is for an enhancement MOSFET, but it should have been a depletion MOSFET symbol. The suggested DN2535N5/DN2540N5 are in fact enhancement MOSFETs and Supertex use the correct symbol in the DN2535/DN2540 datasheets
DN2535 http://www.supertex.com/pdf/data...
DN2540 http://www.supertex.com/pdf/data...

VCC of 140-340 V?
Do you know what VCC, VDD, VSS and VEE means?

What is the purpose of D1 and DV1 in your circuit? And what is PFC connected to?

Attachment(s): 

Last Edited: Thu. May 7, 2009 - 06:20 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Also please tell me which regulator you will use that will accept 340v at input and output 3.6v.

The moment you plug this in the wall, either you will die, or you will set fire to the house and kill your family/neighbors.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

A few schematics ago, you had a circiut with a fuse between neutral and ground. Don't to that; it's illegal, and potentially fatal...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

n1ist wrote:
A few schematics ago, you had a circiut with a fuse between neutral and ground. Don't to that; it's illegal, and potentially fatal...

That schematic is figure 4 from this document:
Microchip TB008 - Transformerless Power Supply http://ww1.microchip.com/downloa...
Microchip TB008 wrote:
FIGURE 4:

Precautions:
1. As mentioned earlier, the neutral should be connected to earth ground through a fuse.
This would insure protection in case of improper wiring.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I'll drop Microchip a note. It's still wrong...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

n1ist wrote:
I'll drop Microchip a note. It's still wrong...

I don't know the laws in this area, they are probably also different from country to country.
But to me it also seems very strange to connect Neutral directly to Ground through a fuse. Please do contact Microchip about this and let us hear their answer.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
This would insure protection in case of improper wiring.

The fuse might blow but the circuit would still work if moved to a properly wired outlet since neutral and ground are usually bonded together at the breaker panel. Odd...

Tom Pappano
Tulsa, Oklahoma

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The ONLY place neutral should connect to GND is at the junction box in the house. Neutral can carry a potentially lethal current when an appliance is connected in the socket.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks all of you again, PFC is "Active power factor correction" , most of devices which consume above 50 watts have such circuits. it has a transformer which can supply some power, once the PFC running , it will charge C22 to 5V so the Q1 will stop work,then the R3 will not consume power any more.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

And I note AndersAnd's N-Channel Depletion-Mode way , it is similar to mine , and it is look more simple , but it will be much more expensive than my way( the 6.5W Depletion mosfet is not cheap ).
May be I'd better change DV0 and DV1 to 8V. the function of DV1 is to prevent C22 charged too high by PFC. D1 can be canceled.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Do you have a bulk capacitor after the bridge rectifier to keep the voltage stable at peak level?
Otherwise your NPN woill only turn on at peak volatages unlike the depletion-mode MOSFET which is tunred on almost the whole perio except at the very bottom of the rectified sinus curve.

iamhere wrote:
And I note AndersAnd's N-Channel Depletion-Mode way , it is similar to mine , and it is look more simple , but it will be much more expensive than my way( the 6.5W Depletion mosfet is not cheap ).

Digi-Key has IXYS IXTP02N50D 500V depletion-mode N-channel MOSFETS in TO-220 pacakge.
Price: 0.61576 @ 500 units.
http://search.digikey.com/script...
This is less than half the price of Supertex DN2540N5-G at Mouser:
http://www.mouser.com/Search/Pro...

Don't know what type of NPN power transistor you plan to use and what the price is compared to IXYS IXTP02N50D?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

My circuit suggestion sounds like it works much the same as the Supertex one except the transistors are used to invert the mosfet operation and to use a enhancment mode rather than depletion mode mosfet. Tradeoff- less parts but more expensive ones or more cheap parts.

Iamhere - this is getting silly - now you're telling us about a PFC. How can you expect us to design you a circuit when you're not telling us the whole story? Here's a hint - go purchase a commercial version of what you're trying to design then reverse engineer it. Find out how someone else solved the problem. It might save a lot of time and money on your part. Why did I not give you a schematic? If I did, you'd be asking simple questions about it. If you can't cope with ohms law, basic circuit theory and electrical safety, then you'll have 'fun' designing power electronics.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks Andersand and Kartman again,
Regarding PFC, sorry I did not mention about PFC before, but PFC is neccessary in most high voltage big watts device (or it will against laws), every one have experiences know that. and I ve told you my device consume about 80W, so you can guess my circuit will have PFC module.

My design working now,I changed Q1 to two NPN, and DV0 to 7.5V, DV1 to 9.1V, and C22 to 100u/25V, it run good, (about 6.5V was tested on C22 without connect to PFC, about 8.8V tested on C22 when connect to PFC) I think my way is much better than Depletion mosfet way :

1. More safe, my design will be more health than Depletion way, with R3 10K 5W resistor ,the circuit's peak current only about 30ma, so even the Q1 go short or regulator and mcu go short,the circuit will still keep safe. now let us see Depletion mosfet, the peak current may be big, it may easy to damage some components . and if there are any components go short, what will happened? also if meet thunder?
2. Consume less power, with connection to PFC inductor, the Q1 will not work any more when the device running. so it will not waste much of energy.
3. much more cheap, let me tell you my costs:
two 500V NPN miniamp transistor (to make a Darlington): 0.01x2
one 10K 5W resistor: 1x 0.02
one regulator: 1x0.07
one cap 100u/25V: 0.01
other parts: 0.02
total costs about: 0.14 ( euro)

And I noted if the DC power go very low the Q1 will stop work, but as the mcu only need 10ma current, so that is not a problem.

I share my design ,may be you can try my way if you meet similar situation, also if somebody have better ways do let me know ,thanks.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

your design is a simple series regulator! you complained about temperature rise so we suggested means that had very low power disipation. The technique I suggested and the one using a depletion mode mosfet are two ways to do the same thing - it has extremely low power disipation.

How well does your circuit work across the input voltage range you specified? How about temperature rise?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hello Kartman, would you like to show the schematic? maybe then it will be easier to analyse the power dissipation.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Kick that troll.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

@iamhere As you are concerned about PFC, this is obviously for a commercial design. Based on your responses in this thread, this design is far beyond your abilities. STOP NOW and hire a qualified engineer. If you do not, you will surely injure or kill someone!

Writing code is like having sex.... make one little mistake, and you're supporting it for life.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The idea was that you come up with the schematic and then you can analyse the power disipation. Design a simple circuit that turns off a mosfet when the input voltage rises above a certain value. You've got mosfet switching losses - next to nothing, mosfet on resistance losses - next to nothing, voltage drops across resistors - considering the currents are in the uA region, 1/4W resistors are sufficient except for the voltage ratings.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I would love to see the YouTube video when this puppy pops!!

Please, for your own safety and for the the safety of the poor schmuck that wants this 'thing' STOP!! This circuit is lethal.

Jim

I would rather attempt something great and fail, than attempt nothing and succeed - Fortune Cookie

 

"The critical shortage here is not stuff, but time." - Johan Ekdahl

 

"Step N is required before you can do step N+1!" - ka7ehk

 

"If you want a career with a known path - become an undertaker. Dead people don't sue!" - Kartman

"Why is there a "Highway to Hell" and only a "Stairway to Heaven"? A prediction of the expected traffic load?"  - Lee "theusch"

 

Speak sweetly. It makes your words easier to digest when at a later date you have to eat them ;-)  - Source Unknown

Please Read: Code-of-Conduct

Atmel Studio6.2/AS7, DipTrace, Quartus, MPLAB, RSLogix user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Jim - we all know what the 'bang' sounds like and the tell tale smell of molten silicon! Personally, I think the youtube videos of people putting nails into LiPo battery packs is more interesting.........

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Kartman,
Good point!!

JIm

I would rather attempt something great and fail, than attempt nothing and succeed - Fortune Cookie

 

"The critical shortage here is not stuff, but time." - Johan Ekdahl

 

"Step N is required before you can do step N+1!" - ka7ehk

 

"If you want a career with a known path - become an undertaker. Dead people don't sue!" - Kartman

"Why is there a "Highway to Hell" and only a "Stairway to Heaven"? A prediction of the expected traffic load?"  - Lee "theusch"

 

Speak sweetly. It makes your words easier to digest when at a later date you have to eat them ;-)  - Source Unknown

Please Read: Code-of-Conduct

Atmel Studio6.2/AS7, DipTrace, Quartus, MPLAB, RSLogix user