Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: Prev 1 [2]   Go Down
  Print  
Author Topic: RF Concepts 2-417 2 Meter Amplifier Won't Amplify No Transmit  (Read 9708 times)
W4JST
Member

Posts: 0




Ignore
« Reply #15 on: September 17, 2012, 05:46:23 PM »

How do I test these transistors?

Everyone agrees is sounds like one or two bad transistors? Even just one would cause the amplifier to not amplify at all and actually reduce power from about 45 to 5 watts?

Thanks!
Logged
WB2WIK
Member

Posts: 20575




Ignore
« Reply #16 on: September 17, 2012, 06:37:56 PM »

The transistors are not in parallel.  If one is dead, the amp is just about dead.

The circuit is actually two independent amplifiers with a splitter driving them and a combiner coupling them together.  The circuit won't work on "one cylinder," it needs both.

What you have is a really common problem, and the reason a lot of solid state amplifiers are for sale in the first place: They've already failed, and the seller is looking to dump them someplace.  Unfortunate, but reality.

Luckily, you can likely replace both transistors for less than $100, which isn't bad.

These issues are the reason I still use tubes for VHF amplifiers. Wink
Logged
WB2WIK
Member

Posts: 20575




Ignore
« Reply #17 on: September 17, 2012, 06:42:57 PM »

Oops, forgot to mention: "Testing" these transistors is very difficult.

You might detect a pure short circuit using only on Ohmmeter, but they usually don't short; the more common problem is blowing one or more of the multiple emitter connections (internally), and an Ohmmeter test doesn't reveal that.

The manufacturer (like Motorola, now Freescale) probably had a method to test them just fine when they were manufacturing them; but that was long ago and it's doubtful they retained that ability over the decades.
Logged
W9GB
Member

Posts: 2616




Ignore
« Reply #18 on: September 17, 2012, 06:54:08 PM »

RF PARTS advises:  
http://www.rfparts.com/
SRF3883 Motorola -- Replace / Use MRF247 or 2SC2782
http://www.datasheetcatalog.org/datasheet/motorola/MRF247.pdf

RF Parts
435 South Pacific Street
San Marcos, California 92078   U.S.A.
Orders and Information
Toll-Free: (800) 737-2787
Telephone: (760) 744-0700
7:00 a.m. to 4:00 p.m. PST  Monday - Friday
Email (US Orders Only): orders@rfparts.com
==
Advanced Semiconductor has the legacy Motorola dies, and does make "runs" of various RF transistors.
http://www.advancedsemiconductor.com/index.html
The MRF 247 is one RF transistor they do offer.
http://www.advancedsemiconductor.com/pdf/mrf247.pdf

Advanced Semiconductor, Inc. (ASI)
7525 Ethel Avenue
North Hollywood, California 91605  U.S.A.
Telephone: (818) 982-1200
Toll-Free: (800) 423-2354
Fax: (818) 765-3004
Email: sales@adsemi.com
« Last Edit: September 17, 2012, 07:18:14 PM by W9GB » Logged
AC5UP
Member

Posts: 3845




Ignore
« Reply #19 on: September 17, 2012, 06:55:27 PM »

There are dozens of web pages that show how to check a transistor, if this one is lacking do a web search for others: http://www.hobby-hour.com/electronics/testing-semiconductors.php

Here's the spec sheet for a Moto MRF247:  http://www.datasheetcatalog.org/datasheet/motorola/MRF247.pdf
I had no luck finding a pinout for the SRF3883 but the MRF247 is supposed to be a sub for it.

With a multimeter in diode test mode (2k scale on Ohms) there should be continuity from B to E and B to C in one direction only. The voltage drop will typically be less from B to C than B to E. Reverse the test leads and you should see infinity in the opposite direction. From C to E there should be infinite resistance in both directions. Best results are with the transistor out of the circuit as your amplifier has a 39 Ohm resistor across the base and grounded emitter.

http://upload.wikimedia.org/wikipedia/commons/thumb/c/c2/Transistor_npn.svg/581px-Transistor_npn.svg.png
Logged

Never change a password on a Friday                
KA4POL
Member

Posts: 1969




Ignore
« Reply #20 on: September 17, 2012, 10:07:23 PM »

How do I check the transistor? Would one bad transistor cause the amplifier not to amplify any? (Power actually goes down when the amplifier is turned on, from about 40 watts to about 5 watts).
I'd start to measure the voltages at both transistors for comparison without RF being applied. This should show you already what might be wrong. Best thing to do, however, is to remove it from the circuit and check out the 'diodes' B to E, B to C. In Ohms range they should conduct in one direction only. So you need to reverse the test leads. Also check for C to E. It could happen that these are shorted. Either way there should not be a low value. I know it is a pain to completely remove the transistor. Unsoldering just the two easiest pins would be a good compromise.
I have considered the attenuation. It plain means the RF is consumed in the circuit, i.e. flowing to ground without any amplification. That's like two equal resistor parallel and one becoming zero.
Logged
W4JST
Member

Posts: 0




Ignore
« Reply #21 on: September 18, 2012, 12:59:10 PM »

Thanks everyone for the great information.

If it's going to be about $100 to replace the transistors, I'm thinking I may rather sell this than pull these out and test them and still have to put that kind of money into it.

What is this amplifier worth now? Does anyone want it?
Logged
AC5UP
Member

Posts: 3845




Ignore
« Reply #22 on: September 18, 2012, 03:04:39 PM »

As a point of reference, a new Mirage B-310 can be had for $239 plus free shipping in October:  http://www.texastowers.com/b310g.htm  Not saying that's the one to get, just sayin'.

The Mirage can manage 100 watts out with less than 8 watts in. Yours is 175 watts out with 45 in. That's less gain but with an output approximately 2 dB greater. Which at 6 dB per unit is less than half an S-Unit. It's also approximately 23 years old. Needs maybe / probably $100 worth of PA transistors plus whatever else hasn't been discovered so far. The Mirage is both a new piece and newer technology by 20+ years. Plus it carries a one year warranty.

IMHO a fair offer is whatever the buyer & seller can agree the RF Concepts amp is worth minus the anticipated repair costs. Note that I am not including any cost for labor, which might be a less than reasonable position.

If the amp is worth $150 it's a $50 amplifier... If it's worth $100 it's a freebie... If it's worth $75 you pay the buyer $25 to haul it away.

I suspect the last option is not going to be considered, but as mentioned previously, just sayin'............   Tongue
Logged

Never change a password on a Friday                
WB2WIK
Member

Posts: 20575




Ignore
« Reply #23 on: September 18, 2012, 03:38:44 PM »

How do I check the transistor? Would one bad transistor cause the amplifier not to amplify any? (Power actually goes down when the amplifier is turned on, from about 40 watts to about 5 watts).
I'd start to measure the voltages at both transistors for comparison without RF being applied. This should show you already what might be wrong. Best thing to do, however, is to remove it from the circuit and check out the 'diodes' B to E, B to C. In Ohms range they should conduct in one direction only. So you need to reverse the test leads. Also check for C to E. It could happen that these are shorted. Either way there should not be a low value. I know it is a pain to completely remove the transistor. Unsoldering just the two easiest pins would be a good compromise.
I have considered the attenuation. It plain means the RF is consumed in the circuit, i.e. flowing to ground without any amplification. That's like two equal resistor parallel and one becoming zero.

All that "might" work, but is likely to actually only reveal a shorted junction and not much more than that.  These ballasted emitter power transistors can fail in ways that DC measurements can't detect.

I recently had to replace an MRF247 in an older Mirage amp.  Out of circuit, it "passed" the "good transistor" test using an Ohmmeter just fine.  But it couldn't run more than a few Watts output (not shorted) because an internal failure prevented it from doing so.  I think what might happen in some of these cases is one or more of the internal emitter ballasts opens, still allowing some current to flow, but not much.
Logged
WB8VLC
Member

Posts: 119




Ignore
« Reply #24 on: September 18, 2012, 06:16:41 PM »

I have that exact same amplifier sitting in front of me and the overheated/problem component is not apparent because the schematic posted is not the same as your amplifier.

 My amplifier went thru a feed back modification to fix some spurious issues and it now looks exactly like yours does, originally it had a single molded choke as L9 and L10.

 These amps went thru an ECO, (Engineering change) at some time in their design cycle with changes made to the collector to base feedback circuitry, what is shown as L9 and L10 in the schematic was changed from a single molded .15 uh choke to a small wirewound choke in series with a VK200/Fb  multi hole bead choke.

The small wirewound choke is encapsulated in an almost clear shrink wrap and in the right transistor it sure looks like it is blackened near the transistors inner flange mounting screw at the junction with the multi turn bead.

It looks like the junction of the wirewound choke and the VK200 ferrite bead choke actually arced to the flange mounting screw.

Someone probably modified this amplifier with the spurious feedback fix and didn't properly dress/insulate the leads at the junction of the changes and or over drove the amp causing the feedback network to unsolder and possibly drop down and arc to the flange mounting screw.

The amplifier uses a wilkinson combiner on the input and output so its easy to disconnect the right side amplifier by removing the lower transistor but I would also open up the feed back choke where it is burnt by unsoldering at the junction of the multi hole bead and the burnt inductor under the clear shrink tubing.

I would also remove the UNELCO metal mica capacitor on the base side of the bad transistor because it looks black and possibly shorted along with removing L7 which feeds bias to the base of the suspected bad transistor, this should totally isolate the suspected bad transistor side.

next try driving it with half power or less, around 20 to 25 watts, only because the input wilkinson resistor will start to dissipate the power on the open transistor side and it will get hot during these tests.

 Unfortunately I don't have a good updated schematic of what the amp looks like with these feedback mods but this should get you started.

 Also the fact that one side is damaged there is the possibilty that someone drove it over the max drive level and damaged both transistors but still by disconnecting the lower device this will be obvious if you don't get any power out with the top amp chain connected up

Mike
Logged
KA4POL
Member

Posts: 1969




Ignore
« Reply #25 on: September 18, 2012, 09:55:31 PM »

All that "might" work, but is likely to actually only reveal a shorted junction and not much more than that.  These ballasted emitter power transistors can fail in ways that DC measurements can't detect.

I recently had to replace an MRF247 in an older Mirage amp.  Out of circuit, it "passed" the "good transistor" test using an Ohmmeter just fine.  But it couldn't run more than a few Watts output (not shorted) because an internal failure prevented it from doing so.  I think what might happen in some of these cases is one or more of the internal emitter ballasts opens, still allowing some current to flow, but not much.
Yes, you are right. I've had similar experience. The question, however, was about what to do. And my experience is, it is worth while trying to measure first. You still might exchange the transistor/s later on. But changing on suspicion is not good either. It may even be an expensive experience if the problem sits at some other place and destroys the good new transistors right away.

I am missing your proposal for what he should be doing  Wink
Logged
WB8VLC
Member

Posts: 119




Ignore
« Reply #26 on: September 19, 2012, 10:57:06 AM »

About the only thing he can do now is to disconnect the bad transistor side from the wilkinsons and run at reduced drive to just see if the other transistor is good.

 This is better than replacing both transistors without knowing if both are damaged.

 Also the SRF3883 is the correct transistors for that model amp but I don't think that RF parts has them anymore as I remember the SRF parts were tighter beat matched parts and they were made exclusively for a particular manufacturer.

  As is mentioned above the difference between the SRF3883 and the MRF247 was different beta and also tighter beta matching of the SRF3883 parts but even with the wilkinson splitter/ combiners being used the tight beta matching shouldn't be a major problem if he uses RF parts Co. MRF247 or similar replacement.

About the only thing he would need to do is set Idq for any replacement transistors, I can mesasure my amps Idq if he needs to know how to set bias which he probably will.

 
Logged
Pages: Prev 1 [2]   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!