-
You can now help support WorldwideDX when you shop on Amazon at no additional cost to you! Simply follow this Shop on Amazon link first and a portion of any purchase is sent to WorldwideDX to help with site costs.
-
Retevis is giving away Radios for the New Year and Every Member is Eligible. Click Here to see the details!
Co-phase amplifiers?
- Thread starter T23
- Start date
Can it be done?
Yes, you could build a input and output combiner just like some builders combine two transistor boards in multi transistor amps.... the old Solid State Varmint SS-600 amps were built using two, 5 transistor amps that were combined together.
It can be done, but you would probably be better off in the end to just get a single amp that will produce the output power you desire from the start.
What did you have in mind?
73
Jeff
Yes, you could build a input and output combiner just like some builders combine two transistor boards in multi transistor amps.... the old Solid State Varmint SS-600 amps were built using two, 5 transistor amps that were combined together.
It can be done, but you would probably be better off in the end to just get a single amp that will produce the output power you desire from the start.
What did you have in mind?
73
Jeff
Thought so, I was thinking it was like that... There would have to be some sort of impedance matching though for the inputs and outputs. But what about the output, would it be reasonable to think one could acheive the added nominal output of both or more amp's rated disapation?
T23
T23
It can be done but not easily. You need to preserve the impedance and ensure that the input and output cables are EXACTLY the same length including connectors etc.Then you need to ensure that the output combiner can handle the total power output as well as proper timing and sync of the changeover relay. Commercial broadcast transmitters do it all the time but they are designed from the ground up to do it. Lots of finesse involved and not worth it. Get a bigger amp.
BTW just a pet peeve of mine but you do not co-phase amps or antennas nor do you use a co-phase cable. Co-phase seems to be just a CB term that has been adopted over time due to popularity even though it is wrong. Drop the "co" part and just "phase " the antennas using a "phasing" cable. LIke i said it is just a looooong standing pet peeve I have.
BTW just a pet peeve of mine but you do not co-phase amps or antennas nor do you use a co-phase cable. Co-phase seems to be just a CB term that has been adopted over time due to popularity even though it is wrong. Drop the "co" part and just "phase " the antennas using a "phasing" cable. LIke i said it is just a looooong standing pet peeve I have.
If you hear that about the antenna, and the cable at the same time, is that a co-pet peeve?
Sent from my DROIDX using Tapatalk 2
It can be done. Both amps have to be identical. Otherwise expect smoke very quick. Is it worth it. Nope. Although I did have a friend that used an external splitter/ combiner for 2 64 pill/ transistor amplifiers. Yeah its insane.
I've noticed many high power VHF amplifiers are using 1/4 wave 75 ohm coaxial phasing cables to split the input and combine the outputs. If one has two high power discrete HF amps it may be wise to consider what they do at higher frequencies because it has less loss and offers the ability to have more distance between the amps.
That will work because those amps are most likely very close to 50 ohms input and output impedance whereas a typical CB amp is likely to be anything but that. Part of the function of the phasing lines is to transfer the impedance and if the two amps have different impedances then it becomes a nightmare to deal with. The problem with dealing with products that have no set technical standards is that there is no set standard to deal with them.
Stacking is simply placing two or more antennas in an array.Stacking lines are also called phasing lines. You do not stack amps unless there is not enough room for them on the desk.
That's why matching is important. You can't combine anything that is not very close in gain, impedance, and power output without running into multiple problems that usually result in smoke. Many think if they have the same two amplifiers, the parts will match. Maybe if you bought them both at the same time from the same place it would increase your chances.
Next week they may have used a different value padder cap. Different supplier for the transformers. Different lot of transistors of even type of transistor. The variables are endless from year to year on these amps and depend more on what parts were available at a good price then anything else including performance at times. Any significant differences in wire lengths that carry RF will also cause undesired phase shifts.
It's also easy to think that transistors will work good together just because they have the same part number. They should at least have the same lot number (number printed under the part number) and have been subjected to the same or no prior use so the gain will be similar. The transistor with the most gain will hog the majority of the drive power and cause it to make more output then the rest. Becoming your weakest link in the chain and a likely point of failure.
Another very important aspect is the phase angle of the "harness" or combiner used. A scaler network analyzer will be needed to ensure the two legs on the splitter harness are within 0-5 degrees of each other and the combiner legs are within 0-5 degrees of each other. Otherwise, you run the risk of one amplifier cancelling the other amplifier out, the closer to 180 degrees out of phase you get. Same principle can be applied to antennas.
For years I've been getting good results with the MFJ antenna analyzer and a precision 50 ohm carbon resistor. I took a 47 ohm carbon resistor and used a Dremmel to grind out enough carbon to make it a perfect 50 ohm resistor. Soldered it right across the inside of the RF connector on the MFJ so it sees a perfect match on its own.
Knowing a exact electrical 1/2 wavelength of coax with the far end open will reflect infinite impedance back to the source, when you parallel infinite with 50 ohms resistive load, you still have 50 ohms resistive load. This means when you connect the 1/2 wavelength of coax across the MFJ and 50 resistor, you should trim the coax for a perfect match at your frequency.
That will give you 1/2 wavelength of coax even if you are using 75 ohm cable since the cable always reflects the load impedance in a 1/2 wavelength cable (or multiple of). The load at the end of the cable is open and the 75 ohm 1/2 wave cable will still reflect an open while the MFJ should only see the resistor.
You can then cut your cable in half to form two 1/4 wave lines. An alternate method to find the 1/4 wavelength directly without cutting the cable in half would be to set the analyzer to double the desired frequency and find the 1/2 wavelength there. It will be 1/4 wave at half the frequency.
I've also wondered if one couldn't use a signal generator and a quality dual trace oscilloscope for finding the electrical 1/2 wavelength of coax? With a "T" connector on channel 1 input, RF generator applied to one side of the "T" and the 1/2 wave cable connected between the other side of the "T" and channel 2 input.
We know the when the cable is cut to exactly 1/2 wavelength it will provide a 180 degree phase shift. If we push the invert button on channel 2 input it will provide another 180 degree phase inversion. This should make both traces perfectly overlap when the cable is 1/2 wavelength at the generator frequency. Anyone tried it?
Knowing a exact electrical 1/2 wavelength of coax with the far end open will reflect infinite impedance back to the source, when you parallel infinite with 50 ohms resistive load, you still have 50 ohms resistive load. This means when you connect the 1/2 wavelength of coax across the MFJ and 50 resistor, you should trim the coax for a perfect match at your frequency.
That will give you 1/2 wavelength of coax even if you are using 75 ohm cable since the cable always reflects the load impedance in a 1/2 wavelength cable (or multiple of). The load at the end of the cable is open and the 75 ohm 1/2 wave cable will still reflect an open while the MFJ should only see the resistor.
You can then cut your cable in half to form two 1/4 wave lines. An alternate method to find the 1/4 wavelength directly without cutting the cable in half would be to set the analyzer to double the desired frequency and find the 1/2 wavelength there. It will be 1/4 wave at half the frequency.
I've also wondered if one couldn't use a signal generator and a quality dual trace oscilloscope for finding the electrical 1/2 wavelength of coax? With a "T" connector on channel 1 input, RF generator applied to one side of the "T" and the 1/2 wave cable connected between the other side of the "T" and channel 2 input.
We know the when the cable is cut to exactly 1/2 wavelength it will provide a 180 degree phase shift. If we push the invert button on channel 2 input it will provide another 180 degree phase inversion. This should make both traces perfectly overlap when the cable is 1/2 wavelength at the generator frequency. Anyone tried it?
When cutting coax lines for phasing purposes, don't forget take into consideration the velocity factor of the coax.
