LME49830 with Foldback Current Limit and 2sk1530

[mantabernd]

Hello!

I designed a LME49830 with foldback current limit and the popular 2sk1530/2sj201, the circuit is similar to the an1850 from ti.

Unfortunately I can only simulate the circuit in tina-ti because I don't have the LME in lt-spice. BUT I don't have the 2sk1530 mosfets in tina-ti So I tried to simulate it with irfp240 and irfp9240 and it nearly worked. The only thing were the low distortions on the takeover point but I think it is because of wrong bias current and the "wrong" mosfets in the simulation. The current limitier worked very well

So now, a question to the specialists here in the fomum, will this amp work??

Is it a good idea to take the expensive 2sk1530 / 2sj201 mosfets instead of the cheaper irfp240 / 9240?

Thanks in advance ;-)

regards
bernhard

 

[AndrewT]

are the sources inside the LME sufficiently protected such that when the protection triggers, your circuit does not blow up the LME?

 

[mantabernd]

Hello Andrew,

good Question, I don't know about fues inside the LME. In the datasheet is only a thermal fuse listed.

How can I protect the LME? With extra transistors for driving the mosfets?

 

[mantabernd]

No one else any suggestions/comments ?

 

[pergo]

LME's drivers are current limited to 56mA.
Datasheet shows that drivers can be shorted with a 10ohm resistor without problems.

That circuit will work.

 

[mantabernd]

Oh ok I didn't see this detail. Thanks for this information!!

So that's my first own amplifier design - I'm very excited about building it. I hope it won't fire it up ... Maybe, if I got the time for it I will test it this evening.

 

[AndrewT]

I think that both Pass and ESP websites shown a method & circuit to allow non destructive testing of current limiting circuits.

 

[mantabernd]

Sorry for the question but do you have links for the circuits? I didn't find them and I like non destructive solutions ;-)

 

[basitccp]

Fold back current limiter

i have a regulated bus supply of 28 volts now i want to design foldback current limiter circuit which limits load current when it exceeds above 5 ampere. can any one has any idea..

Thanks in advance..

 

[mantabernd]

I built it

I finally built that circuit above with only one pair of mosfets and it worked. I had some problems with oscillations at the beginning, it is very important to have only one big gnd point. After I changed this the oscillations were gone

So can't say anything about the sound quality because my testspeaker is very bad but I measured it.

The current limiter needs a bit tuning, the negative side seems to limit correctly but the positive side looks very very strange.

I tested it by putting a 2R4 Ohms resistant from output to gnd.

Supply voltage was +/- 60V. The Voltage at the gate of 2sk1530 was 1,758V and on the 2SJ201 it was -1,537. After putting the 2R4 Ohms at the output the voltage at the 2sk1530 was 1,501V and on the 2SJ201 it was -0,499V.

So it doesn't work simultaneously

I added two pictures, one of normal operating mode and another with the resistant at the output.

Anyone any suggestion what to change?

 

[pergo]

add 10nF capacitor between base and collector of Q2 and Q3

 

[mantabernd]

Yes it worked Wow, so fast!!

I didn't have a 10nF so I used 33nF.

Well done, thanks pergo!!

 

[pergo]

great! that was a huge oscillation

 

[jackinnj]

Q2 and Q3 shouldn't be necessary with the LME49830.

Some other stuff -- 1) the gate stopper resistors should be mounted as close as possible to the gates of the MOSFETs. 2)Increasing the value of the compensation capacitor will reduce the slew rate and may be necessary with some devices.

 

[mantabernd]

Yes I know, it's my own design for short-circuit protected amplifier and it works!!

No I only have to change the values of the protection because it workes too early.

 

[AndrewT]

DC and very long duration limiting must respect the DC SOA of the devices.

But, transient peaks far exceeding the DC SOA must be allowed to pass unhindered.

You need to modify your limiting to allow medium term and short term and very short term transient demand currents to pass.

 

[mantabernd]

It's a bit sad that I can't simulate it with this fets.

On the first step I will change R13 and R14 (the current resistors) from 0R22 to 0R10, then the current circuit will limit much later.

In my testcircuit I used 1N4007 for D1 and D2 because I don't have shottky diodes. Would it be better to replace them with shottky diodes? If yes, what will be the advantage?

 

[AndrewT]

It would probably be safer to keep the DC current limit low or very low while you experiment with limier component values.

That last thing you want is to blow up your amplifier because the limiter came in a bit late.
Can you arrange in your circuit for the limit to reduce as heatsink temperature rises?

 

[mantabernd]

It would probably be safer to keep the DC current limit low or very low while you experiment with limier component values.

My DC supply limits the current at 2,5A.

That last thing you want is to blow up your amplifier because the limiter came in a bit late.

Yes, that's right!

Can you arrange in your circuit for the limit to reduce as heatsink temperature rises?

Sorry I didn't understand this sentence... Does it mean that I should use a ntc in my current limit circuit?

 

[peranders]

Try to add increase R15, R16 to 220-330 ohms. I have never seen an amp with lateral fets having 22 ohm only. You problem can be found if you read this... I'll trace it down for you.