Login  Register

Re: KPA500 Questions about power and efficiency

Posted by KQ5S-Tom on Jan 17, 2018; 1:45am
URL: http://elecraft.85.s1.nabble.com/KPA500-Questions-about-power-and-efficiency-tp7637476p7637482.html

I did it again.  I meant to send this to the group but sent it to Jack
direct.  Sorry Jack.

"The most recent firmware version is V1.54, releasing in mid December."

I went to the software page and it says the most recent beta release is MCU
1.52 Nov 17, 2017 and production release is 1.47.

-----------------
73,
Tom - KQ5S

On Tue, Jan 16, 2018 at 7:18 PM, Jack Brindle <[hidden email]> wrote:

> Hey Richard;
>
> I’m glad you are enjoying our creation. The KPA500 is truly a fine piece
> of hardware and firmware
> that continues to evolve. The most recent firmware version is V1.54,
> releasing in mid December.
>
> Others have answered many of your questions, let me give my shot at a few.
> I usually run my
> KPA500 in contests at 600 watts. Briefly lighting the 700 watt LED is not
> unusual, but I don’t
> consistently leave it lit. The amplifier is quite comfortable at 600 watts
> for long periods of time.
> Mine is one of the first units and was hand-built, so I can’t claim to
> have the original PA Deck
> installed (it is still sitting on my work bench however), but my current
> PA Deck is one of the
> original shipping versions. After about 10 years, it is working just fine
> for me.
>
> As for calibration, we perform the cal at the factory for factory-built
> units, but not for kits.
> The reason is that we need to marry the front panel board with the PA Deck
> to perform these
> adjustments, something that is not done until well into assembly. The
> adjustment values are
> kept in memory on the front panel board’s processor. This also means that
> if you swap the
> PA Deck you will need to recalibrate the measurements. Feel free to
> perform your own
> calibration to a known standard. You will probably find that you will need
> to vary the values
> by some small amount to get really accurate measurements. In general they
> are pretty close
> to start off with, but won’t be “right on.”
>
> Enjoy the amplifier, and by all means use it at the power you want. If you
> have any questions
> or need any help, we are here, just ask. You will find customer support to
> be very helpful.
>
> 73!
>
> Jack, W6FB
> Elecraft Engineering
>
>
> > On Jan 16, 2018, at 3:14 PM, Richard Stutsman <[hidden email]> wrote:
> >
> > I've had my KPA500 for a couple of weeks, now, and I have questions
> > relating to its efficiency as a function of power output and band. A
> friend
> > of mine recommended that I run only 300 watts output most of the time in
> > order to preserve the life of the FETs. I told him that I didn't spend
> > $2300 on an amp just to get less than 5 dB of gain in my signal (over my
> > 100-watt exciter) instead of the 7 or 8 dB I paid for. Am I wrong?
> >
> > I took a lot of data this morning that indicates that there is an inverse
> > relationship between output power and efficiency - to the extent that,
> down
> > to about 300 watts, I find that the amount of heat the finals have to
> > dissipate is less, the higher the power output. This is rather
> > anti-intuitive.
> >
> > And if that is true, then wouldn't it be better for the finals if I
> > routinely ran the amp at the higher power, so long as voltage, current,
> and
> > temperature remain within safe limits? Is it safe to run 600 watts on CW,
> > especially on 80 and 160 meters, where the efficiency appears to be so
> good
> > that the finals run cool at 600 watts - even cooler than at 500 watts?
> >
> > The following data was taken while transmitting into a dummy load. I
> > adjusted the drive power to obtain one of several predefined output power
> > levels and then measured input volts and amps and calculated input power
> > (HV x Current), power dissipation (Input - Output), and efficiency
> > (Input/Output). Each measurement was taken after about one second of key
> > down in CW mode. Doing the calculations in between each test allowed the
> > finals and dummy load to cool off. I never heard the fan come on. Drive,
> > Output, Input and Dissipation are in Watts. I have shown voltages falling
> > slightly below 60.0 in red, but I had no problem getting 600 watts output
> > with less than 40 watts of drive, even in those cases.
> >
> > *Band  Drive Output  Volts  Amps  Input Dissipation Efficiency*
> >
> > 1.8    23    500    64.3  10.2   654     154         76%
> >        30    600    63.9  10.9   697      97         86%
> >
> > 3.5    23    500    65.1   9.4   612     112         82%
> >        30    600    63.2  10.5   664      64         90%
> >
> >  5     24    500    63.7  11.2   707     207         71%
> >        30    600    62.8  12.0   754     154         80%
> >
> >  7     27    500    62.3  12.5   779     279         64%
> >        35    600    61.2  13.8   845     245         71%
> >
> > 10     32    500    61.2  13.4   820     320         61%
> >        39    600    59.8  14.6   873     273         69%
> >
> > 14     15    225    63.4   9.6   609     384         37%
> >        20    310    62.9  11.5   723     413         43%
> >        24    400    61.5  12.8   787     387         51%
> >        33    500    59.0  14.3   844     344         59%
> >        40    600    58.7  15.8   927     327         65%
> >
> > 18     26    490    60.6  12.9   782     392         63%
> >        33    600    59.6  13.8   822     222         73%
> >
> > 21     26    500    60.0  13.9   834     334         60%
> >        35    600    59.8  15.2   909     309         66%
> >
> > 24     23    500    61.4  12.4   761     261         66%
> >        30    600    61.0  13.5   824     224         73%
> >
> > 28     22    500    61.4  12.7   780     280         64%
> >        27    600    60.6  13.8   836     236         72%
> >
> > Note that in every case there is lower power dissipation (and therefore
> > less heat generated), the higher the power. I decided to test several
> lower
> > power outputs on 20 meters, and I got the maximum dissipation at 310
> watts,
> > and the dissipation at 225 watts was nearly the same as it was at 400
> watts.
> >
> > I should point out that my HV during standby or idle hovered between 79.5
> > and 80.0 volts. I am using the Red tap wire on the 115-volt input.
> >
> > And, yes, I know not to use the amp on 60 or 30 meters.
> >
> > If anybody is going to yell at me for testing a 500-watt amp at 600
> watts –
> > and for proposing to use that much power routinely on 80 meter CW – let
> me
> > know your reasoning and why or how you think that would stress the amp.
> >
> > Oh, and does anybody know if Elecraft would have adjusted the Power
> > calibration (per band) of the KPA500 kit version, as mentioned in the
> User
> > Manau? I found all the settings to be at their default 100%, FWIW. When
> > comparing at 100 watts to a cheap external watt meter (which agreed with
> my
> > TS-590sg power output setting), I found I had to increase power on almost
> > every band to 104 watts to trip the 100-watt indication on the KPA500's
> > wattmeters. I decided to leave the settings at their default.
> >
> > Rick N6IET
> > ______________________________________________________________
> > Elecraft mailing list
> > Home: http://mailman.qth.net/mailman/listinfo/elecraft
> > Help: http://mailman.qth.net/mmfaq.htm
> > Post: mailto:[hidden email]
> >
> > This list hosted by: http://www.qsl.net
> > Please help support this email list: http://www.qsl.net/donate.html
> > Message delivered to [hidden email]
>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email]
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]