level shifters and mosfet drivers

[Pafi]

they cost around 4-5 dollars excluding VAT

The same as here, but how much time do you have to work for this money?

 

[Tekko]

I can afford them but it feels wrong paying for IC´s that will most likely blow up a few times before it gets right.

 

[Danko]

Yes, it feels bad, paying for a small smoke. But it is good, paying for knowing, hot NOT to connect the IC, how NOT to use, how NOT to ......

If you know, why did the IC (or anything in a circuit) blow up, then it's worth the money, you spent on the IC.

Uh. I hope you understand my sentences, i'm not a native english speaker

 

[Tekko]

I have a schematic on a class d amp that works in a schematic simulator (borrowed from a guy here @ the forum) Tho it havent been built in reality if i understood the thread right, but i was gonna give it a shoot anyway when i have gotten some IR2010 MF drivers.

 

[Tekko]

Phew. I was able to cancel the samples order from them MAX628CPA Now i dont need to worry about getting IC´s i dont need.

 

[Tekko]

I have ordered three IR2010 IC´s now, costed about $20 tho

I hope i can replace 1N914 in the schemo with 1N4148 or some other small high speed diode from a smps.

 

[Tekko]

Now i have got all the parts and an begin the build.

Here is an idea on layout. Suggestions on changes are welcome.

Layout example

 

[Danko]

The webserver says, I don't have permissions to access the requested URL. :-(

 

[Tekko]

Keep trying.

 

[Tekko]

Now i have tested the TL494 PWM part and it appears to work.

 

[Tekko]

Now i have completed the amp, but theres a problem, the 74HCT14N doesent invert the signal, like it should.

What happens ? Both fets get turned on at the same time = the fets get hot as well as the driver and the bootstrap diode.

The 74HCT14N is driven by 12 volts DC.

 

[Tekko]

Bah 74HCT14 is 5V only and MM74C14 is nowhere to be found

Looks like i have to come up with an own solution as the output from the pwm is 0-12v while 74CHT14 can only take 5V and the IRF2010 needs 12v input. Damn.

Does it have to be a HEX inverted Scmitt trigger or could i make an inverter with just a transistor and a resistor ?

 

[Ouroboros]

Instead of the 74C14, you could use the 40106 from the old first-generation CMOS range which is still available.

 

[Tekko]

Nevermind.

I´ll change the Tl494 to run as pushpull instead so i wont need inverters at all. If that works i dunno.

 

[Tekko]

First IR2010 blown.

 

[Eva]

Any 4000-series NAND, NOR, XOR or XNOR CMOS gate or inverting buffer will work. These parts are usually named CD4xxx, HCF4xxx, HEF4xxx, MC14xxx, etc... (use google if you don't know them yet). They are somewhat slower than the 5V CMOS TTL-compatible 74HCxxx or 74HCTxxx series but they come very handy anyway because they will handle 15V or even 18V depending on part suffix.

Also, the IR2010 has a "Vdd" logic supply pin that you can connect to +5V if you are using TTL levels, or to any other supply voltage up to 25V in order to get proper input logic thresholds. Read the datasheet.

 

[Tekko]

I have read the datasheet over and over.

I have flushed the idea of using inverters on the inputs of the gate driver, i use the TL494 ability to give two split signals instead.

 

[kASD]

Hi Tekko,

You Cannot use TL494 to split the signals in different polarities because the maximum duty cycle obtained from either output of TL494 is limited to 50% only , In order to get full or near 95% duty cycle, you have to wire the outputs in "OR" configuration....

The best way is to use a fully differential comparator with differential input/Output for inverted/non-inverted signals simunteanously such as LM360 from National Semiconductor....

www.national.com

K a n w a r

 

[Tekko]

One word: BAH.

Using inverters are out of the game anyway, the output from the tl494 is 12v so 5v inverters don´t suffice.

Is it possible to make homemade inverters, like a resistor and transistor or does that just make a level shifter ?

 

[Pafi]

You Cannot use TL494 to split the signals in different polarities because the maximum duty cycle obtained from either output of TL494 is limited to 50% only , In order to get full or near 95% duty cycle, you have to wire the outputs in "OR" configuration....

You forgot OutputControl pin! You can get inverted signal from 494 simply: Just wire one internal transistor to common emitter, other to common collector! The problem is there is no place for DT adjust. But paralelled diode with gate-resistor can do it.