I was very kindly donated an RM KLV200 valve linear amplifier some weeks ago from a station not far from me. Sadly they didn't work, which wasn't a major issue, as there so simple inside. The EL519 output pentode had gone pop. Valves don't usually go pop as-such. There's many stages of failure, most typically being weak emissions. However this did go out with a spectacular light show. The Anode and cathode had an internal short-circuit, which is another common fault along with open filaments. I was in company with Pete 26CT3855 at the time when we power up the amp on the bench for testing. I can honestly say it made us both jump back.

Picture below of front panel.



You may have already noted from reading the post title and looking at the image, the amp is indeed labelled as the KLV350. Obviously RM reused the front panel from that particular version. Looking for information on the internet revels that the KLV350 was discontinued not long after released.

Picture below of rear panel.



The rear panel has the basic minimum connections for wiring the amplifier inline. Notice how the RTX and ANT labelling runs as one continuous word. I can't help to think perhaps some people wiring this up backwards.(yes it's common, I hear it on the bands a lot.)

Picture of internal view.



As you can see from the photo, these amplifiers are incredibly simple inside, theres also a sticker label on the transformer confirming the true model number of the unit. If it had been the KLV350, there would be 2x EL519's in parallel configuration.

So why did it fail, well thats probably one of 3 possible reasons.

1st being the plate output loading not being correctly set, this in-time will have a massive effect on longevity, as the valve has more-than-likely been "Red Plating" due to the massive mis-match on it's output.

2nd being the valve has been subject to many hours of standby "without use". This is another contributing aspect to premature failure, as the cathode emission surface will just evaporate, rendering the valve useless and likely to have internal shorts.

3rd and finally is the valve may had an undetected fault which was destine to fail at somepoint after a few hours or thermal cycling. Another contributing factor is excessive shock during transit from supplier to customer. This could have also happened to the valve in place as the whole unit was subject to excessive shock treatment.

Final conclusion

The amplifier now works absolutely fine, and has done ever since I replaced the EL519. I do remember noting that the original owner commented on the relay "Chattering" whilst in transmit. First of all this in not a relay fault. if you look closely at the anode PCB on the final picture, your see there is a very poor attempt as making a "Meander" or "PCB Track Inductor" for the APC or "Anti Parasitic Choke". Those who have spent many years working with valve equipment like myself for just over 30 years, will instantly recognize this problem. You can either replace the valve that has a slightly different spec on performance, or simply add a 6 turn 9mm dia APC inductor in series with a 2 watt 10 ohm carbon resister to stop this issue.

Here's a quick look at a "BPF" for the 11mtr band. This particular Band Pass Filter was built out of necessity for the current WebSDR I have running at my QTH.

This "BPF" has a total of 10 Elements (sometimes called poles) in the system, which has a very good rejection for unwanted signals appearing on the SDR waterfall. When I originally setup the WebSDR, there was not only "Birdies" but "Harmonic" images appearing, along with 2 very powerful FM Broadcast stations around 26.500 and 27.415 Mhz. This wasn't going to make a very popular SDR with those there, so I had to make a filter.

Photo of Home-Made BPF Below



As you can see from the photo, the filter is constructed in a very typical fashion, with each section being isolated from the next "well almost". Anyhow this is a very important thing to do, as if we was to omit the ground screening from each stage, we would have a very poorly working filter with unwanted break-thru likely to occur from the very signals we are trying to filter out.

There are many types of band pass filters we can use eg. "Butterworth" or "Bessel" to name a couple, but this particular filter is based on the "Chebyshev" design, which has one of the steepest cutoffs.

In basic terms the filter is merely nothing more than individual tuned stages, both in "Series and Parallel" configurations at the precise pass band center frequency. Which in this case has a center frequency of "27.250" Mhz.

Below showing the cutoff range of the Chebyshev BPF.



As we can see from the above graph, this filter not only cuts off the FM Broadcast band, but also removes all unwanted signals from AM Broadcast Stations too. From 520Khz to 1.8 MHz the dB loss is approximately -130dB and for FM Broadcast 87.MHz to 108 MHz the cut of is approximately -65dB to -80dB respectively.   

What we have effectively is to different pass filters working in unison. "RTL-SDR" do sell filters for removing unwanted Broadcast stations for both AM and FM, but they are packaged and sold as individual items. Whereas in this case we've achieved both in one simple Band Pass Filter.

Schematic Layout



Above shows the circuitry layout for the "BPF" should you wish to construct your own. You can either choose one or both methods of tuning to calibrate your filter. The method I decided on was to make the inductors variable, as opposed to variable capacitors.  The type of tuning is essentially "Permeability Tuning". I prefer this way to calibrate the filter more so than variable capacitors. But you can easily use both methods for extra adjustment if you wish.

Below is the listed component values required to build the filter. Note that some of the calculated values are not going to be easy to obtain as in fixed disc capacitors, or of the peg pre-wound inductors. So your going to need to add capacitors in either series or parallel to make-up the required values (or near as possible) to the calculated values.

Components Values

L1   0.3133 uH   
C1   147.93 pF   
L2   0.4422 uH   
C2   104.81 pF   
L3   0.1819 uH   
C3   254.77 pF   
L4   0.4422 uH   
C4   104.81 pF   
L5   0.3133 uH   
C5   147.93 pF

Calibration

To calibrate (or align) the filter you can either use a "Vector Network Analyzer" or "VNA" for short with a signal injected at the input end, and the VNA connected to filters output. You will be able to have a visual picture of the filters cutoff, and fine tune these with the adjustments methods you decided to use when constructing the filter.

The other option if a VNA is not available to you, is to just simply place the filter inline between a normal CB radio that's tuned to the CF of the filter, with a power meter and dummy load connected to its output. Tune the filter until maximum forward power is achieved. Set the radio to no more than 4watts or lower when making the adjustments, as the lower the input power, the easier it be to see on the power meter.

Little SSTV Sked tonight on 27.295 USB. In total there was 4 Stations participating 26JS07, 26CT3699, 26AM200 and myself.

Here's a few of the SSTV images received at my Home QTH.









Hopefully more stations for the next SSTV Sked.

Excellent mate  👍👍

Hopefully now from my laptop it will attach. Just to note for forum that this is RTTY on fldigi. Will & I worked PSK31 this eve before switching to RTTY
....73
Rob