Since relay switched low-pass filters are a thing, and relay switch ATUs are a thing, it makes sense that an amplifier could have a relay switch input matching network instead of a rotary band switch.

Does such a thing exist?  Can you buy a reasonably priced board or kit to build that will handle at least hundred watts and could be substituted into an amplifier instead of a hard to find and expensive ceramic wafer band switch?  I guess component values would have to be individually tailored to a specific amp design, but it seems like this would be a useful kit if it were made available.

My Elecraft KPA-500 uses pin diodes for bandpass switching.

There used to be at least one supplier who offered an input matching board for the homebrewer. The method of tuned circuit selection was left to the builder.

If you use relays or PIN diodes, you're still going to need band-signaling information in the form of low-voltage inputs to the control elements...and the easiest way to do it is via a rotary switch.

I think most solid state amps use a relay or pin diodes for selecting tuned circuit for the band.

Are you talking about Solid State or tube type amplifiers.

Solid state do not require band switching on the input side. Input side is a wide band fixed impedance. The only reason for a band switch on a solid state amplifier is on the output side for the bandpass filters. Solid State amplifiers without output bandpass filters created multiple harmonics at high levels. You want and need to eliminate those harmonics and focus the output on the frequency of the input. Relays are used for bandpass filter selection be you are switching RF energy that is considered different than HIGH Voltage DC. Small relays can easily handle multiple Kilowatts of RF energy. Simply because the relays are seeing a very low impedance ( 50 Ohms). Where you get into trouble is when you are feeding the amplifier with bandpass filters into an antenna with anything other than a 50 Ohm impedance. This creates high standing waves and causes increased power dissipation requirements of Bandpass filter components and switching systems. Usually the transformers and capacitors and sometimes the relay contacts. Usually software in the amplifier prevents Hot Switching the Input and Output antenna relays and the Bandpass Relays. All relays change state when no RF is applied.

Tube Amplifiers are totally different animals. They are cranky old men and are fussy about what is fed into them. Tubes are narrow banded inputs and require tuned matching circuits to provide a 50 ohm INPUT impedance. These circuits are HIGH DC Voltage High Current with RF in them. As in Kilovolts. To switch these types of circuits would require very large and very specialized relays for each band of interests. It has been done but not very dependable. Much easier to use a band switch usually a ceramic wafer type switch with large low resistance contacts. Switching while RF applied will usually result in catastrophic destruction.  The load and tune circuit adjustments match the output impedance of the tube to the antenna impedance. Because these are tuned outputs they do not require bandpass filter outputs. The tuned output circuit acts as a harmonic filter.

Pin Diodes as has been referenced earlier by others are simple RF switches. They are great for switching RF energy and are excellent for Input and Output switching. They are rather expensive devices that are not simple to turn on and off. Each Pin Diode requires it's own circuit to control it. Could you use them in a solid state amplifier to select the bandpass filter on the output side? Yes but way overkill and over complicated for the usage. Can they be used with tube amplifiers to control the tuned input? No because of the high voltage and current requirements. Ceramic Wafer Switches aren't a bad thing. Like anything mechanical will wear out over time. They are not massed produced and only a few good manufacturers still make them. Yes they have been poorly cloned by the Chinese and are worse than a bad case of crabs.

Sorry but we are stuck with them for tube amplifiers input circuits.

Maybe I should not have had that 2nd pot of coffee this morning!

thanks for the explanation. I guess that is why my setup in my truck between the turner and the coil using 12 volt automotive relays to switch coil taps with 100 watts works. I tried it and use it but never knew why they could switch 100 watts ( not hot switching ) without a problem.

There is no bandswitch on my solid state amp. There is NO tuned input sections at all. The power coming from the exciter goes into a 50 ohm resistor "dummy load". So its input always looks like 50 ohms to the exciter. A small amount of power (somewhere around a watt) goes to the input of the amplifier, and a small amount is feeding a frequency counter that determines the frequency of the input signal so that the OUTPUT bandpass circuitry is automatically switched to the correct bandpass components before leaving the amps output connector.

My amp is the Mercury IIIs

Fairly common relays can switch 1500W without a problem. You'll find them in many HF antenna switches. For instance this switch from KK1L. At HF freq's you can get away with the strange impedance of a common relay. Things start to get more demanding as frequencies increase, of course.

https://www.w6pql.com/using_inexpensive_relays.htm

Just ordered a bunch of these. I lost one of the Surplus coaxial relays I was using on my 6m KW amp for the imput side. I am now going to a relay system. I also got the high power versions for 2m for the other harris.

Tohatsu makes board mounted coaxial relays as well that are good up to about 450mc but not at high power. As other mentioned as the frequency goes up there is stray reactance that happens in the relays and cause impedance changes. Also the insertion loss get very high and the relays suffer on the isolation side. One reason as you cross the UHF boundaries into SHF you start to see expensive dow key coaxial relays.

Yes, I planned on just using a commonly available 5 position rotary switch in place of the wafer bandswitch.  Hopefully one that has an identical degree of motion so it will match the band labels on the front panel.  You could even use a 7 position switch and use a relay for standby mode as well.  Just rotate it down below 80m or up above 10m (whichever is closer at the time) to put the amp in standby mode.  

This way you'd avoid drilling any new holes in the front panel and it would still look factory.

Yes, it's a tube amp.  I was just trying to brainstorm a way around using ceramic wafer switches that are hard to obtain.

I would try a rotary switch that switches relays in and out. It would be simple to do and you would not have to modify the case.

I think it is a good idea.

I know the outboard band switch wafers are available but inboard is tough to find. Pretty disappointed to see that the band switches on Harbach’s web page have jumped up from 45 dollars to $129.99 since they have been purchased and many of the parts have gone up a lot. Guess the Amp repair guy is making some changes. I just got my switch just before they were taken over so I guess I lucked out.

OP, have you looked at the band switches on RF Parts?

https://www.rfparts.com/switches/switches-band.html

There is another website that had them for a little over $80/per wafer but I don't remember the name.  Should show up with a web search.

I think I poked around there before, but I am unfamiliar with the site layout.  Looks like a good resource though.

Relays that can handle voltages in the multi Kilovolt range are very large and very expensive.

Talk to this guy he custom makes wafer switches rated for 5 kilovolts. Plus he makes some of the best tuners in the U.S.
https://www.blueridgeamateurradio.com/misc-parts/wafer-switch

I really lusted after one of these tuners bit wanted multiple outputs. I found a TT 238B that does everything I need.
73,
Rob

This is only needed for the output side right?  The input band switching could be handled with the relays as they would only need to handle circa 100W from the exciter.  It's the rear wafer (which are still available, though limited) for the SB220 that handles stuff on the HV side.  

I really, really wish there was a cheap high temp 3D printer that would run glass-filled nylon great right out of the box.  I'm not am engineer though so I wouldn't know if glass-filled nylon wafers would be able to handle the voltage.  Still, it would be easy to make replacement wafers using 3D printing and laser cutting for the contacts.  You could probably get set up to make new band switches for around $10k or less, but it would take a long time to recoup your investment and you'd have to design them for multiple hard to find switches to get enough demand to make it worthwhile.  

I suspect if there was enough demand, there would already be someone filling this role.

Pull up the diagram of the heathkit and see what the voltages are on that switch. There is some high voltages from the power supply on that switch.

Looking at the schematic, unless I'm missing something (and I very well could be), it looks like only the rear wafer handles any kind of high voltage.  It really looks like all the input switching occurs on the front wafer.

Since there has been so many diagrams over the years let's reference a common diagram.
http://www.wedophones.com/Manuals/Scanners/Heathkit%20Schematic-SB-220_sch.pdf

What that diagram calls wafer2 has B+ voltage on it.

Wafer 1F handles the tuned input circuit. At most it only has 60 VDC on it.

Yeah that's what I'm seeing.  Wafer 2 in the amp is inside the tube chamber with the 2 big inductor, and the wire coming from wafer 2 and L7 heads down through the bottom of the chassis to the T/R relay.  So if I wanted to control band switching on the HV side using something other than the wafer, I'd need those expensive relays.  

The good news is that the wafer most likely to suffer arcing damage is going to the the rear (wafer 2) and that's the one that's still available through Harbach, albeit for $124.00.