-
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.
Texas Star DX500 and high SWR
- Thread starter 2RT307
- Start date
Hi Gentelman,
Sorry this reply is so late. I have this experement on my to-do list I just haven't had the time lately. I have been very busy with famiy issues. I plan on getting this done in the near future. I have some parts that I need to order up for it yet. I will first try it just seperating the bias from the Rf path. Then I will try it with the regulated feed to see the results of both.
I will keep you informed of the results as soon as it is completed. And from what I can see it should have a positive outcome.
73's
Eric
Sorry this reply is so late. I have this experement on my to-do list I just haven't had the time lately. I have been very busy with famiy issues. I plan on getting this done in the near future. I have some parts that I need to order up for it yet. I will first try it just seperating the bias from the Rf path. Then I will try it with the regulated feed to see the results of both.
I will keep you informed of the results as soon as it is completed. And from what I can see it should have a positive outcome.
73's
Eric
The diodes used for regulated bias, so you have thermal tracking, I have used a 1n4997, drill a hole just large enough you can seat into heatsink to track temperature better. Another way is to carefully file 1 side down slightly and mount ontop of transistor ( this is good enough for Kenwood). Another way, is to lift emitter, carefully notch out board slightly and then install diode so it sits on heatsink the pull tab back over and solder. This way diode makes contact with heatsink and tracks heat best. Tab holds it in place. May need a little piece of heat shrink tubing on one side to keep it insulated. Just a few suggestions that may help.
Thanks for the suggestion. I am not sure if I will have the time to totally break down the unit for such a modification, although thermal tracking would be the best way to go. I am not sure if I would put the diode so close to the transistor. In my area where the winter tremps drop as low as -5 degrees f, I think I wouldnt have the tempreture tracking react so quickly. I run almost 50 cfm worth of cooling on the heatsink of this amplifier and it gets only lukewarm in the dead of summer after a 5 minuite keydown. I have a 115 degree thernal switch mounted to the heatsink where the driver would be located in the 667v and it seems to a good job of switching the fans on at the appropiate time.
At this point I would just like to see if this bias seperating project will cure this thorn in my side issue that has plagued myself and many others for so long. These units are a good choice for use , But could be much better if they didn't do so much corner cutting in their designs. In the distant future I would like to take some time and add a thermal tracking design to it.
Don't worry Brett. As soon as I get this done and make sure the results are what they need to be, Then you can send both of them to me to have the issue corrected.
73's
Eric
At this point I would just like to see if this bias seperating project will cure this thorn in my side issue that has plagued myself and many others for so long. These units are a good choice for use , But could be much better if they didn't do so much corner cutting in their designs. In the distant future I would like to take some time and add a thermal tracking design to it.
Don't worry Brett. As soon as I get this done and make sure the results are what they need to be, Then you can send both of them to me to have the issue corrected.
73's
Eric
Not sure if you have access to the motorola rf device data handbooks. But application note AN1035 has a really simple yet stable bias circuit. Only modification I did to circuit was add an 8V regulator inline with bias circuit so it is very stable. Usually I just mount the didiode right on top of transistor. Easiest way and as I said. Kenwood has done this on there radios's that have 2x 2290 for final. Just to test circuit, I just touched diode with tip of soldering iron and watched the Vdrop .05 almost instantly. Also usually I bias a little harder. Up to about .7V. Looking for cleaner signal not more watts. I know its CB ( I only talk CB). But rather have cleaner signal. In the circuit, they use a BD135 I think, but I use a TIP31C and get very good results. Might check it out unless you have a different design in mind.
Best of luck.
Best of luck.
By the way, the transistors (pills) are almost $40 each now. Better to er on the side of caution and have a fast tracking setup than a setup that tracks slower and run the risk of blowing. Expensive lesson, trust me on that one.
I am aware of what they cost nowadays and have multible sets on hand. I have had the amplifier more than ten years without having a heat related failure.
It helps to be mindfull of your transmissions when using it. 50 cfm's worth of cpu cooling fans helps as well.
The most important thing to remember is to run it at, OR below it's rated specifications.
It helps to be mindfull of your transmissions when using it. 50 cfm's worth of cpu cooling fans helps as well.
The most important thing to remember is to run it at, OR below it's rated specifications.
You have to watch how much you bias. To much and you add more heat before your even modulate it. I never go by the bias voltage, But rather the idle current. I have found that it best set between 120 to 130 ma. per transistor pair. The trick with active temp bias tracking is to keep it in the sweet spot between clenliness and efficiency. To much positive bias and your adding heat, And to little and you risk slipping into the class c realm. If I had to err I would rather have it by way of clenliness. But that is just me.
I wonder just how benifical and active bias circuit would be for these amplifiers? To much range wouldn't be good and If it can be kept out of the class c operation then maybe. If not then perhaps it would be best just to use a regulated bias feed and set it for the optimum balence an call it a day.
Not sure on the 500. They already have it packed. For me, the higher bias V is not an issue. As amp is fully encased. With fan blowing on top then thru heatsink. And I used the 120mm fans that are rated over 2" backpresssure. LOL. Overkill, but it stays cool no matter what. The axial fan draws 5A. Anyways to each their own. Just my preference.
Update!
Well, I finally got around to poking around with this thing some more. I discovered that my Bencher filter was not playing right.... like it's owner, it has a scew loose. :blink: Since Bencher decided to pop rivet the whole stinkin' case, and I didn't feel like drilling all of them out, I picked up an old 1kw rated Nye Viking TVI filter from a trusted source. Slapped it in line after the amp and before the meter. Swr's without the amp running a Grant XL and 5' Firestick = 1.5:1 on 27.385. With the Texas Star on and sitting at about 100 watts dead key, Swr = 1.5:1. So it appears to me that the high SWR's were indeed caused by harmonics feeding the meter a false reading, and the filter is doing it's job. I have no idea why the Texas Star started doing this, but since it only cost me 20 bucks for the Nye Viking, I'm going to continue to run it that way.
73,
RT307
Well, I finally got around to poking around with this thing some more. I discovered that my Bencher filter was not playing right.... like it's owner, it has a scew loose. :blink: Since Bencher decided to pop rivet the whole stinkin' case, and I didn't feel like drilling all of them out, I picked up an old 1kw rated Nye Viking TVI filter from a trusted source. Slapped it in line after the amp and before the meter. Swr's without the amp running a Grant XL and 5' Firestick = 1.5:1 on 27.385. With the Texas Star on and sitting at about 100 watts dead key, Swr = 1.5:1. So it appears to me that the high SWR's were indeed caused by harmonics feeding the meter a false reading, and the filter is doing it's job. I have no idea why the Texas Star started doing this, but since it only cost me 20 bucks for the Nye Viking, I'm going to continue to run it that way.
73,
RT307
- No one is chatting at the moment.
