Broadcast TV Preamplifier Evaluations

[ocw]

Due to my wanting a new RF preamplifier for off air broadcast TV reception, I decided to compare some old model preamps with some new ones that I had made and/or purchased.  I prepared the attached report to show the results of my tests.

I previously liked the performance of the Winegard AP-8700 preamp.  Their AP-8275 unit had extra gain which made it easy to overload and I was beginning to wonder about the AP-8700 as well.  At first, I thought about making a comparison of their IMD figures but decided to take advantage of some equipment that I have and measure the preamplifiers’ gain, weak signal capabilities and find at what point that an increase in their input level would decrease an excellent receive signal MER of 45 to 30.  MER is similar to SNR; with MER counting IMD and other spurious unwanted emissions as noise.

I used my Rohde & Schwarz SFQ TV test transmitter to generate the DTV signal.  It has a MER of 45 on its output.  A Tandberg E5710 encoder fed a test pattern AIS signal to the SFQ.  Its output fed the DUT preamplifier.  The output of that fed my R&S EFA TV test receiver which measured the MER (and many other parameters) of its input.  I thought that I might need a higher level than the SFQ could provide for these tests, so I purchased a Chinese made board with a TQP3M9103 amplifier IC.  While rated for one watt/30 dBm output, I found that it had a higher output level with the low IMD and harmonic figures that I was looking for as compared to the common 3-watt eBay special amplifiers.  Those have bad IMD and harmonics above ½ watt output.  The TQP3M9103 amplifier has a narrower frequency range.  I retuned it to be centered around 600 MHz.  I was trying for closer to 500 MHz but kept having efficiency problems going lower than 600.  I decided that I could live with that and had my tests near the top of our current broadcast channels, on channel 35.

I measured the performance of preamps using 15 different amplifier IC’s/transistors.  Many of the circuits were hand built recently using promising looking IC’s.  I built and tested more than one using certain model IC’s.  Qorvo was found to offer the greatest number of models having the gain and performance that I was after.  For each I measured its gain, the minimum signal level which would have an output with a MER of 20 or above (a MER of 15 or above is needed for reception), and the maximum level which would produce an output with an MER of 25 or above.  The difference between the last two was calculated to show the amplifier’s dynamic range.

Besides an excessive input signal hurting reception of the strong signal, the IMD produced by it can hurt reception of the adjacent and even second adjacent channels.  An even stronger signal can block reception of a still wider range.  To demonstrate this, I show how an AP-8700’s output to a spectrum analyzer doesn’t normally hurt the quality of a TV signal with a MER of 45.  But when overdriven its output is much worse due to all of the IMD which was being created.  My SA pictures on the attachment show the difference.

I color coded some of the figures on the attached results.  As expected, the Winegard preamps did not look too good.  Since I recently modified my BT-1000 amplifier, I could not complete the DTV tests on it.  Based on prior regular IMD tests on it, I approximated the figures.

The LTC6430 preamps had the best ability to handle strong inputs.  However, they were not quite as good on weak signal reception.

Changing the bias inductor used with my PGA-122-75+ amplifier corrected a bad looking amplifier.

The Qorvo QPB and MAAL-011139 amplifiers have a feedback circuit keeping the gain fairly constant over a fairly wide frequency range.  The Qorvo TQP and most other amplifiers do not have the feedback which make them easier to build.  But they have decreasing gain at higher frequencies.

I added a mild high pass filter to a few of the units to decrease the gain on the VHF TV and FM frequencies.  They had little effect on the UHF channels.

I have probably already made this explanation too long.  I hope that some find the attachment useful.