Re: IRF510 amplifier failures

M Garza <mgarza896@...>
 

Jerry,
I do not disagree, I have a fuse inline with both power lines.  When a fuse blows, you dont know if there has been damage or not.
In my opinion, it would be better to prevent the situation from being able to happen.  This idea seems to have worked for the builder, 
since he made specific mention of it. 
"With the current limit circuit the amplifier now survives transmitting into any SWR from an open circuit to a short."

Marco - KG5PRT


On Thu, Jan 25, 2018 at 11:37 PM, Jerry Gaffke via Groups.Io <jgaffke@...> wrote:
A suitable fast-blow fuse or polyswitch in the PA-PWR line into the board should be sufficient
to protect the IRF510's from too much current.
The polyswitch is like a fuse, except it resets once it cools down after you remove power.
A thermal sensor on the IRF510 heatsink that at least turns on a front panel LED might be a good idea.

This old post here from Allison is informative, well worth re-reading every few months:
    https://groups.io/g/BITX20/message/22597
As she says, lots of things can kill an IRF510, but the most common is heat if you have a good amp design.
Powering at 28v for 55W from a push-pull pair, she didn't see trouble when driving improper antenna impedances.

I'd think that if powered at 12v, and if they have a reasonable heatsink, and if you don't screw up 
when fiddling with the gate bias pots at rv2 and rv3, or drop a screwdriver into the wrong spot,
the IRF510's should not blow.  Even when driving a non-existent antenna, the RF voltages at the drains
go up some without a load, but not drastically, T11 is just a 1:2 voltage transformer.
Driving a 0 ohm antenna might be a bit worse since the currents will go up, but probably survivable.

Jerry, KE7ER


On Thu, Jan 25, 2018 at 09:11 pm, M Garza wrote:

There is a schematic for the current limiting circuit that is used.  This might be something to incorporate into the design.  It is only 7 more parts.

 


Join BITX20@groups.io to automatically receive all group messages.