[Top] [All Lists]

Re: [Amps] How to read the 3-500Z spec sheet?

To: Jeff Blaine AC0C <>
Subject: Re: [Amps] How to read the 3-500Z spec sheet?
From: Larry <>
Date: Sun, 23 May 2010 12:49:41 -0700
List-post: <">>
Well Jeff,

I both agree and disagree with your statements below. I agree with your 
statement about guys not measuring at all. But I strongly disagree with 
telling these same guys they can do "adequate" measurements with an 
incorrect setup.

The guys you are targeting are not competent enough to discern the 
difference between sound card setup issues, exciter issues, and 
amplifier issues. The average guy needs to have an idea what is required 
to do real testing of an amplifier BEFORE you give him the easy way out. 
And you need to detail the "easy" setup including the tests at each stage.

I strongly disagree with proliferating half ass measures which will 
invariably produce even more junk science than we are already dealing with.

For the average ham, it's far better to measure PO1dB (1 dB output power 
compression level) and always operate linear modes below that point. 
That guarantees the best compromise between power out and linearity on 
any amplifier with minimal test equipment and experience.


Larry - W7IUV
DN07dg - central WA

On 23-May-10 12:11 PM, Jeff Blaine AC0C wrote:
> Larry,
> You are 100% right if an absolute accurate measurement is needed. But if
> you simply want to know if you are in the 20's or the 30's, then the
> method I describe is fully adequate to the task.
> I just completed protos of a crystal modified Wentzel design with some
> tweaks added by Bill Carver that is for high DR3 range receiver testing.
> We are talking about IMD levels for the entire system of sub -110dbc.
> Right on all counts - LPFs needed - ferrites needed - lots of added
> return loss via pads sprinkled. I'm fully on board your claim here. But
> just not for Joe's basic ham shack needs. For the basic "is my tx/amp
> combo generating crap" kind of test, the 10dbc better rule of thumb for
> the sources is good enough and the modern sound card without being
> overdriven is fully up to the task.
> The problem as I view it is that guys are not measuring at all. Or if
> they are it's with something like an SB610 or whatever. To my thinking,
> better to have some minimal indication even if it's "gross" in
> resolution than none at all. This solution provides that - it's better
> than the SB610 at least. And given the number of "I've never built
> anything before and don't really know what I'm doing" guys comments we
> see on the boards, this has got to be a step in the right direction.
> Even if it's filled with significant shortcomings.
> 73/jeff/ac0c
> --------------------------------------------------
> From: "Larry" <>
> Sent: Sunday, May 23, 2010 1:51 PM
> To: "Jeff Blaine AC0C" <>; "AMPS"
> <>
> Subject: Re: [Amps] How to read the 3-500Z spec sheet?
>> On 23-May-10 10:45 AM, Jeff Blaine AC0C wrote:
>>> Of course, a real IMD test requires a 2 -tone source and you can
>>> download
>>> those applications on the web free of charge.
>> Wrong answer for amplifier testing. While you can approximate proper
>> "two-tone" testing with a sound card tone source when evaluating a
>> transceiver, a LOT more is required for accurate and technically
>> correct amplifier testing.
>> To do accurate IMD (and more correctly OIP3) testing on an amp, you
>> NEED two separate clean signal sources. This means two exciters that
>> are known to be very clean and free from harmonics (much greater than
>> 60 dB down) plus a high power combiner and usually a set of bandpass
>> filters at each of the tone frequencies. Plus, just to keep everything
>> happy, a bunch of attenuators sprinkled around the bench.
>> Anything less and you are only testing the source(s) and not the amp.
>> Some time back, W8JI detail the correct test setup for amplifier
>> testing and I think it's on his web site someplace. Anything less and
>> you are just fooling your self.
>> 73,
>> Larry - W7IUV
>> DN07dg - central WA
Amps mailing list

<Prev in Thread] Current Thread [Next in Thread>