Jump to content
Electronics-Lab.com Community

Audiogurus opinion of inverter


audioguru
 Share

Recommended Posts

Hi Ante,
This inverter will heat with about 100W so it will try to draw 600W. That is 50A from a 12V battery.
As I said before, then the output transistors will try to conduct 25A each! But their max rating is only 15A. Do you think they will blow open or they will short?

If the load is only 250W then the output transistors might survive, until the load is turned off. It is too bad there are no protection diodes.
Do you think the output transistors would be destroyed immediately by the inductive voltage spike?

Then without a fuse there will be fireworks! Dangerous??


MODERATOR EDIT:

Link to comment
Share on other sites

Here we go again!
The original 500W inverter project didn't work because it had many errors, so Rhonn and I corrected it. The corrected circuit has been built and proven over the years!

The project had the corrected schematic and the defective one for a long time until recently, but now the corrected schematic has been deleted and the original schematic has been "corrected" by MP. The thread about the corrections done by Rhonn and me has also been recently deleted.

Now we have a 500W inverter that blows up! It also uses a timing capacitor with an extremely high value for a non-polarized one, and it uses two quad opamps instead of a single dual opamp.

Here is the project: http://www.electronics-lab.com/projects/power/033/index.html
Here is the schematic corrected by Rhonn and me and the schematic "corrected" by MP.

post-1706-1427914316837_thumb.gif

post-1706-14279143169779_thumb.png

Link to comment
Share on other sites

Hi Mik3ca,
Each side of the output of the inverter has a three transistor darlington circuit for its high current gain. The final output transistors are paralleled to increase their max current rating.

It is a 500W inverter. It isn't perfect, it is cheap so it uses common 2N3055 transistors with a 15A max current rating instead of using 75A Mosfets. The transistors heat with about 100W so its total is 600W at full 500W load.

600W from a 12V battery is 50A. The 2N3055 paralleled output transistors on each side must conduct that 50A alternately through the transformer.

MP's version has two output transistors in parallel on each side= 25A each. Their max rating is only 15A and they don't have much current gain above 10A. They are very overloaded and will fail.
My version has four output transistors sharing the 50A on each side= 12.5A each. No problem.

In MP's circuit the 2N3055 driver transistors are also overloaded. Maybe 20A each?
In my version, the output transistors have a minimum current gain of 5. Therefore their total base current on each side is 12.5A/5 x 4= 10A. The 2N3055 driver transistors can handle it.

The max output current from the opamps is only about 25mA so the C1061 transistors amplify it to the 2A max base current for the driver transistors.

That is why there are so many transistors.

All transistors should have something across their base-emitter to turn them off and to bypass leakage current. Most circuits use resistors like in my version. MP's circuit doesn't so will probably have "thermal runaway" where a transistor doesn't completely turn off when it should be resting which makes it get hot. Then the heat increases its leakage current which creates more heat which creates more current, around and around until it melts. When you buy darlington transistors then the resistors are inside.

My circuit has diodes to arrest the inductive voltage spikes from the transformer. MP's circuit doesn't have them.

My circuit has a fuse and a power switch. MP's doesn't.

Nobody has made MP's circuit, not even him. My circuit has been made by many people with good results. It is too bad my circuit has been deleted from the project.

Link to comment
Share on other sites

Mik3ca,
First, you should understand that I did not design this inverter. I only corrected the mistakes in the schematic. There were lines that were crossed and connected, and decimal places wrong in values. They were simple fixes. The design is still the work of the original author.
Secondly, note that you do not need the current limiting resistors in a switched application such as an inverter. Such resistors are needed for a linear amplification such as an audio power amp. Look at the many designs all over the web for a square wave inverter circuit and you will see they are all missing the resistors. Square wave inverters are all basically designed the same. There is nothing different about this circuit than the many all over the world presently in service. I have actually had emails telling me the one that audioguru helped design works much better if the power resistors are removed from the circuit.
Note that audioguru admits that he has never built an inverter.
This is why I moved this thread to the Theory section.

BTW...it is my understanding that Rhonn's design is not deleted. It is just being moved to it's own page. Audioguru is making a big deal out of nothing.

MP

Link to comment
Share on other sites

I used 0.1 ohm emitter resistors on the output transistors to help match them so the one with the highest gain doesn't "hog" all the current then fail first. Then the remaining transistors will also fail one after the other in order of their amount of gain and Vbe.

Transistors all have different amounts of current gain and different Vbe voltages. Small transistors like the BC547 can be purchased as BC547A for low gain, BC547B for medium gain and BC547C for high gain. They can also be purchased without the letter then have any amount of gain like 2N3055 power transistors.
Inverter manufacturers can match transistors if needed. Kit manufacturers can also match transistors. I don't think a hobbiest would want to sort through a bucket full of transistors to try matching them when emitter resistors can do it.

With a 500W load then each 0.1 ohm emitter resistor has 12.5A through each output transistor plus up to 2.5A from the base for a total of 15A and a loss of only 1.5V. The "12V" battery is 13.8V anyway so the small loss is not serious.
A high gain transistor will also have a low Vbe and will clamp the base voltage to its value if the paralleled transistors didn't have emitter resistors. Then low gain transistors would not have a base voltage high enough for them to conduct much current. There is a huge difference of base voltage from "typical" to "max".
The emitter resistors reduce the gain of high gain transistors to match low gain transistors for more equal sharing of the total current.

My corrected circuit worked well. Why is it deleted from the project?

Mik3ca, here is an inverter with the same few corrections as my larger circuit but for only 100W:

post-1706-1427914317254_thumb.gif

Link to comment
Share on other sites

It has been a practice to keep emmiter resistors to limit th current under failures before a fuse could blow. i suggest few more  additions.
1. an electrolytic cap of considerable value proportion to the load crt across the +ve and -ve very near to the transformer mid point and common ground of the O/p emitters .  also it is better to add a series resistor of low value and an electrolytic to act as a filter while feeding power to  the CMOS IC. 

one point is not clear how the device cd4047b is adopted for such large power supply, while a serios thought is to be given to dead zone ( a small time lag between return to zero of the plulse on Q output and raise of pulse on Q* output.  let us recollect that lack of this deadzone would bring bottom pair of transistors on before the current in top ser dies down.  modern ICs like sg3524 have taken care of this phenominun while working at high frequencies and few manufacturers adopt 3524 at 50 Hz output .  perhaps someone can enlighten me further.

sarma

Link to comment
Share on other sites

Hi Sarma,
I agree that an electrolytic supply bypass capacitor would be needed if the battery is far away. The battery has a huge amount of very low impedance capacity, like a huge capacitor.
The Cmos IC would probably be destroyed by alternator spikes if the battery is being charged in a car while the inverter is running. A series resistor with a zener diode would fix it.

This inverter is used in The Philippines where the battery is charged in a far away different city! A guy on a motorcycle carries the batteries back and forth.

The 2N3055 transistors switch very slowly so I don't think a dead zone is necessary. It is used with high switching frequency Mosfet circuits.

Link to comment
Share on other sites

You are really a paranoid person, talking about secret emails and replacing projects. The inverter project that I corrected was on this site long before you first appeared. It was still on this site when I fixed the lines which were drawn badly by the author. You should seek professional help instead of trying to instill unrest on this site because you have a beef with me.
Here in this forum you are doing the same thing that causes me to delete your posts in the forums that I moderate. Your posts have nothing to do with Theory or anything else. Just bitterness, grumbling and insult. I didn't pull your precious audio amp inverter off the website. I don't even have access to that area of this site. But you already know that since you used to be a moderator here and lost it. Smoke and mirrors. Just trying to get people angry to help your cause. Why don't you go peddle this crap on Aaron's website where they will take it.

MP

Link to comment
Share on other sites

MP.
Google's cache of this inverter project shows my corrected inverter schematic there a few days ago. It was recently replaced by yours that has problems. You ruined the project.

After you deleted my reply to the person who enquired about making the defective schematic, you had my corrected schematic deleted and replaced with yours then sent him a secret e-mail.

You say here that you received e-mails about the project when it had my corrections working well for for two and a half years.

Why did you barge into my explanation of the project to the person, which is what he enquired about, then cause so much trouble??
What is your problem??

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
 Share

×
  • Create New...