Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Next revision
Previous revision
mrf458 [2013/03/18 18:56]
n1eu created
mrf458 [2018/10/29 00:13] (current)
n5na
Line 1: Line 1:
 +====== MRF-458 and MRF-454 Final Replacement ======
 //(comments posted by W9GB to qrz.com forum)// //(comments posted by W9GB to qrz.com forum)//
  
-TEN TEC OMNI-V Final AMP MRF-458 burned out several days ago, so I replaced them with 2SC2879, readjusted bias.+AG6JU commented: ​TEN TEC OMNI-V Final AMP MRF-458 burned out several days ago, so I replaced them with 2SC2879, readjusted bias.
 It worked OK, except at the 10 meter band the RF output was 75 watts instead of 100 watts. ​ It worked OK, except at the 10 meter band the RF output was 75 watts instead of 100 watts. ​
 Just in case somebody have same problem. MRF-458 is difficult to obtain. Just in case somebody have same problem. MRF-458 is difficult to obtain.
-Niko - 
  
-The Toshiba 2SC2879 is a well known RF transistor as well ....+ 
 +W9GB responded: ​The Toshiba 2SC2879 is a well known RF transistor as well ....
 The difference between the Toshiba 2SC2879 and the 2SC2879A is that the “A” version is RoHS compliant, ​ The difference between the Toshiba 2SC2879 and the 2SC2879A is that the “A” version is RoHS compliant, ​
 and uses an Aluminum Nitrite insulator (color gray) on the header vs. Berillium Oxide (color white). and uses an Aluminum Nitrite insulator (color gray) on the header vs. Berillium Oxide (color white).
Line 24: Line 25:
  
 IN ADDITION, RF PARTS stocks both the Motorola and M/A-COM MRF-454 ($43-$45) and Motorola MRF-458 ($24) parts !! IN ADDITION, RF PARTS stocks both the Motorola and M/A-COM MRF-454 ($43-$45) and Motorola MRF-458 ($24) parts !!
-http://​www.rfparts.com/​transistors_MRF-TP.html+https://​www.rfparts.com/​catalogsearch/​result/​index/?​limit=all&​q=mrf
 == ==
 The 1983 Motorola data book shows that the MRF-454 is a somewhat more rugged transistor than the MRF-458. The 1983 Motorola data book shows that the MRF-454 is a somewhat more rugged transistor than the MRF-458.
Line 36: Line 37:
 The MRF 454 has a higher gain at 30 mhz: 90 watts for 5 watts of drive while the MRF-458 is spec'd at 70 watts for 5 watts of drive. ​ The MRF 454 has a higher gain at 30 mhz: 90 watts for 5 watts of drive while the MRF-458 is spec'd at 70 watts for 5 watts of drive. ​
 The 1994 data book lists the MRF-454 as a direct replacement for the MRF-458. The 1994 data book lists the MRF-454 as a direct replacement for the MRF-458.
 +
 +===Soldering===
 +
 +while every tech has his or her own style, here's how I do it.
 +
 +you're 35 watt iron should be fine. You don't want to lift the copper from the board!
 +
 +while the soldering iron is warming, draw, photograph or take some kind of notes so you know what goes where. Most likely you'll need to unsolder feedback capacitors and resistors, so make damn sure you know where they go.
 +
 +you'll need some, make that lots, of solder wick. Get the good stuff and not from china. You want it at least 1/4 in wide.
 +
 +
 +suck up as much as you can and then try and get even more off the pcb.
 +
 +then when a sharp knife (I use an xacto) slip under the tabs and with heat applied to loosen any left over solder, gently lift the tabs up from the pcb
 +
 +the collectors have the tip of their tabs cut at an angle. Make sure you document how the transistor (S) go in. They can easily be flipped 180 deg and you'll never know it.
 +
 +I don't recall what they used for the pa, but get some thermal goo and apply a thin layer onto the heatsink and device. Thin is better than a big glop.
 +
 +I go for  the base first. I can then use a VOM to do a resistance check on the remaining leads. Bet you'll find a base to collector short.
 +
 +And while I'm in there, I'll lift the base leads of the drivers and look at them as well.
 +
 +I've had great success using my weller WTCP iron with a 35 watt element and a large chisel tip.
 +
 +and you need lots of solder wick.
 +
 +the two emitter leads are the hardest as they are mounted to the largest amount of copper on the pcb
 +
 +mike, wb8vge
 +-------------------------
 +A bigger (higher power) iron means more heat capacity and faster heating, so less time to destroy things. A 35 watt is fine. I would not try a 15 watt. Even if it is temperature controlled, the big wide tabs and PCB copper will sink the heat from the iron and drop the temperature below the set value of the iron. But you do not want a higher temperature,​ which will burn things up, like the PCB material, or separate the copper from the fiberglass. What you want is a chisel tip (sometimes called screwdriver tip) as wide as the collector tab on the transistor, on a temperature controlled soldering iron. It does not need to be an adjustable temperature iron, but it should be a temperature controlled iron. The difference is that with a temperature controlled iron, you select a tip (I'm thinking of Weller WTCP and similar irons) that gives you a fixed temperature. With an adjustable one you have a knob (or up/down buttons) and a readout that lets you set it. In either case you want about 700 to 750 Fahrenheit. If you use too low a temperature you mess around too long and damage things. If you use too high a temperature you burn stuff up.
 +
 +You want to get in, melt the solder, and wick it away quickly, and get out before you cause a lot of damage.
 +
 +I suggest you find some scrap piece of equipment that has large pieces of copper (similarly sized as the power transistor tabs) soldered to the PCB and practice a bit unsoldering and resoldering on the expendable piece of scrap, before you get into your PA.
 +
 +DE N6KB

QR Code
QR Code mrf458 (generated for current page)