The 1200w solid state amp gets a homebrew LP filter - pics

[K1JJ]

Well, finally got most of the bugs outa the new (16) MRF-150 amp and now added a homebrew  air-core low pass filter for both 40M and 75M.

The old LP filter from CCI choked up at 300w, so had to roll my own. Thanks to Frank/GFZ for the design. I swept both bands last night and the roll off above 4.5 mhz (75)  or  8mhz (40M) very well.

The LP filter is switchable between 40/75M using two input/output toggle switches.

I bought a 2nd 48V 34A power supply from Frank and now have each module getting power from its own supply. Now the amp will put out serious power without voltage sag.

* The reason the coils are spread out is cuz I adjusted them for best plateau and roll off specs according to the RF sweep using 50 ohm loads at both ends. *

I marked the component values in case anyone wants to duplicate the filters.

[K1JJ]

Here's a pic of the spectrum of the SS amp. Huz brought over a nice spec analyzer to check out all my amps here. I was very impressed with performance.

Shown below is an integrated (30 seconds) shot of my voice going through high and low frequency excercises. The result (LSB) is about a 4kc TOTAL bandwidth waveform that drops off 40db from its peak. I'm using 2.8kc filters in the rig. That's pretty decent and I'm very happy.  

Tom, K1JJ

[WA1GFZ]

Tom,
Keep an eye on that big dork knob for temperature rise. There is a lot of current there. I'll be interested in how they hang in. The filters will radiate a bit so if you have any RF in the shack consider mounting them in a box.
Nice room heater for the winter operating. I think you need a bigger heat sink though. Bias will be cut back as it heats up so IMD will come up as the heat sink warms up.

[K1JJ]

OK Frank -

I just ran a power test thru them.  I socked yallos for a full minute at 500-600w pep. The coils and capacitors ran stone cold. I couldn't detect any rise in temp at all.

I'm idling each module at about 5.5A and the fan keeps the fets and heatsink slightly warm but not hot after a transmission.

The better news:  Remember the 75M problem I always had where there was less output than the other bands for a given drive level?  Well now on  75M with the LP filter in line the amp acts exactly as the other bands and puts out full power.  It appears it likes the load of the LP filter and probably settles down the parasitic or whatever it had. 

Interestingly, if I switch on the 40M LP filter for 75M operation, the problem comes back. Same with no LP filter of course.  So I won't play with success and call it basically completed.

T

[WA1GFZ]

IMD should be fairly good at over 1 amp per FET. Great on the filter. I was also planning to use door knobs so good you ran the test for me. A simple LP filter puts a weird load on the amp above cut off. I guess it helps in this case making things happy. Looks like you finally have a solid state amp you like after, what 18 years.
Maybe my 15 meter problem with erb 2 is related to no filter.

[K1JJ]

Yep, it's a good amp now.

Another thing I notice is the waveform on the scope has cleaned up. Without the LP filter I cud see the fuzzy signs of harmonics in there. Now the waveform looks clean, like it is mainly composed of fundamental energy, as you'd expect.

It's running it very conservatively now and unless I do something stupid, shud last a long time.   I still have no swr, overdrive or overcurrent protection, (except for a B+ breaker) so there is risk.  The 50 ohm pad for the FT-1000D is a great load and form of drive protection. I have accidently dumped near to 100 w into the pad and had no problems.... lucky it was there.

I'll wait until the sunspots climb up to make a LP for 10/20M.

T

[WA1GFZ]

At 40 volts it will take a beating to hurt the MRF150s Going lower with voltage will make it tuffer. At 30 volts it should do 400 watts out.  So when are you going to add two more EB104s.