ALC power spike info

  • 1
  • Question
  • Updated 4 weeks ago
Hi All,

Over the last few weeks I have managed to kill two BLF188XR LDMOS devices for no obvious reason.

The first was using a Flex 3000 and the second using a Flex 6400 a week later. Both radios are only good for 100W max o/p. The amps had 16dB of input attenuation so only about 2.5w hitting the 188XR, 1.8v bais and 51v on PA, hence well within spec.

On both instances the input gate NP junctions had failed (low ohms), most probably due to overdrive, ALC overshoot, TX spike etc. so my question is has anybody used a fast trace scope to observe and record the initial few seconds of TX on a 6400/6600?

Other people must be using LDMOS 188/189 devices with these radios so what experience is there out there about this issue please?

73 Steve G1XOW
Photo of Steven G1XOW

Steven G1XOW

  • 301 Posts
  • 134 Reply Likes

Posted 1 year ago

  • 1
Photo of Steven Linley

Steven Linley

  • 447 Posts
  • 78 Reply Likes
I have drive (TX setting) limited to 12Watts. Using a 13dB attenuator.
(Edited)
Photo of Neal Pollack, N6YFM

Neal Pollack, N6YFM

  • 299 Posts
  • 151 Reply Likes
What model amp was this?   2.5 watts, even if it briefly spiked, like from an Icom 7300,
to almost twice that, would not damage the devices with proper input protection design.
But if this was an ebay RF brick, with unknown bias, filtering, feedback, or protection
circuits, then all bets are off.   Even with the 16dB input attenuation, the LDMOS devices
themselves have good gain and high frequency response, requiring good design to control
oscillation.   Oscillation in an LDMOS device can puncture the junction in milliseconds.
But SWR, coax shorts, and the like don't tend to bother LDMOS, as the main output 
channel is amazingly tough.

Reference:  https://www.youtube.com/watch?v=8ziYqjMQGEQ

Also Ref:  https://www.youtube.com/watch?v=NZQHlAyLczQ

Notice that while it is hard to kill a properly designed LDMOS amp with output malfunctions,
the portion left unsaid is just how fast you can destroy them during the design phase if a
run-away oscillation occurs.  Or, in the field, if the design is marginal.

Even if this is a commercial amp, I have a hard time thinking that a Flex, Yaesu, or Icom
transceiver could harm it.   Most all ham transceivers are well known, due to design of the
circuits used, to have an initial surge on key up, amount varied by brand/design.  The amp
makers have known this for decades, and take that into account.
It is likely some other fault that developed with the amplifier, or else ALL the solid state
amps would be blowing up, including mine.

Neal
Photo of Michael Walker

Michael Walker, Employee

  • 1264 Posts
  • 375 Reply Likes
Steve

I am using the same RF-KIT amp you are with 13db pads and have yet to have an issue.  

Have you discussed this with Reinhardt?  Could it be with the method the LMDOS chips are installed as there was a lot of discussion on the RF Kit forum regarding that?

Mike
Photo of Doug Hall

Doug Hall

  • 225 Posts
  • 61 Reply Likes
Steven,

I don't have a 6400, but below is a shot of my 6300 at 100W. As you can see there is no discernible overshoot. I also measured it at lower power levels (50W, 10W, 1W) and the result was the same. Likewise if I use AM or FM and use the PTT instead of the Tune button I get the same result. No overshoot in TX at all. The Flex is actually the best radio I have owned in this regard. I don't know if the 6400 is the same, but I suspect it is.
73,
Doug K4DSP


Photo of Dee Hendricks

Dee Hendricks

  • 1 Post
  • 0 Reply Likes
Please note if your ldmos devices is not soldered down or clamped with special custom compression clamps ,Clamps must push your entire LDMOS device  flat against the copper heat spreader. Your thermal paste must be bare minimum used if not soldered in.

Your devices will fail if you don't do this. your bottom of ldmos devices must seat as flat as possible to your copper spreader or it will blow up  from thermal runaway in a short time. Your input drive should not push your device more than 1000 watts per device on a bird meter with correct matching element for accurate watt readings.

I have been using both the BLF188XR and the MRFE6VP61K25H.  devices in pairs and singles for 5 years with no issues . Also source your devices from reputable suppliers as some of the chinese versions seem to fail although the odd purchase from them worked for some builders.

ZS1NAZ  you may contact me on nasri.7ng@gmail.com  73
(Edited)
Photo of Tom W3FRG

Tom W3FRG

  • 16 Posts
  • 2 Reply Likes
I have been using the RF 2K+ Kit for over 2 years with a 13 dB pad installed and driving it with 5-6 watts typically from a Flex 6600M or 6300 without issue.
I just changed this week to a 16dB pad to insure I cannot exceed the 2.5w max limits.

Using the LP-700 I have never observed a Tx spike.
I have replace a device however, reason unknown?

Tom W3FRG