Anan 7000DLE Mk-II PA output.

User avatar
Gary_Sitton
Posts: 10
Joined: Mon Mar 04, 2019 6:35 pm

Anan 7000DLE Mk-II PA output.

Postby Gary_Sitton » Sun Mar 31, 2019 8:23 pm

Hello:
I just received a new Mk-II without the internal NUC (I have my own which I use externally) and I am having a problem with the power amplifier (PA) output levels on some bands. First let me say that I'm no expert but I do know the fundamentals of RF amplifiers. With this in mind I have observed that on some bands, 17, 10 and 6 meters especially, PA drive attenuation must be significantly lowered to reach 100 watts of output into a 50 Ohm dummy load. I have been using an old Anan 100-D for years and never had a PA output issue with it or using Pure Signal (PS).
What I have found is that one may indeed compensate for the mysterious low PA output with additional drive but at the expense of increased IMD distortion. This increased distortion is probably due to driving the PA into its non-linear operating range. If it weren't for wanting to use SSB with PS rather than CW only, there would be no drive output issue. The "rub" is that you get plenty of SSB output in this over-driven PA situation but with noticeably increased IMD distortion. So, here comes PS to the rescue, but in order to have enough operational head-room to reduce the distortion, PS reduces the final pre-compensated output level to well under 100 watts.
In some sense, Apache's claim to reduce IMD levels to < -50 dB is correct but always requires using PS. This only presents a problem for any bands which may have low output requiring over-drive to reach the nominal, advertised 100 watt output. My suspicion is that these low-level bands have some sort of impedance matching issues with the output circuitry. And by the way, a set of schematics sure would help to at least confirm the possible source of this problem. I certainly hope that my 7000DLE Mk-II is not typical but that is one purpose for this post as I do not know anyone yet who owns one other than myself.
73, Gary K5AMH

P.S. No other issues yet to report and I did update to the 2.6 firmware.
73, Gary K5AMH
User avatar
w-u-2-o
Posts: 1617
Joined: Fri Mar 10, 2017 1:47 pm

Re: Anan 7000DLE Mk-II PA output.

Postby w-u-2-o » Mon Apr 01, 2019 1:44 am

Gary_Sitton wrote:With this in mind I have observed that on some bands, 17, 10 and 6 meters especially, PA drive attenuation must be significantly lowered to reach 100 watts of output into a 50 Ohm dummy load.
This seems to be a very common observation by most people.
The "rub" is that you get plenty of SSB output in this over-driven PA situation but with noticeably increased IMD distortion. So, here comes PS to the rescue, but in order to have enough operational head-room to reduce the distortion, PS reduces the final pre-compensated output level to well under 100 watts.
That is probably not correct. If you are achieving 100W with two-tone and PureSignal active, then you are also achieving it with voice modulation. See this post: https://apache-labs.com/community/viewtopic.php?f=20&t=2862
In some sense, Apache's claim to reduce IMD levels to < -50 dB is correct but always requires using PS.
Yes, certainly. The amp's in the Apache hardware, without PureSignal, don't appear to be particularly different than any other non-linearized 12V amp. This also includes the 50V amp in the 8000. Non-linearized I only see IMD performance from the latter in the -35dBc'ish range.
My suspicion is that these low-level bands have some sort of impedance matching issues with the output circuitry.
That's very possible.
And by the way, a set of schematics sure would help to at least confirm the possible source of this problem.
Schematics are available from the factory on request.
I certainly hope that my 7000DLE Mk-II is not typical but that is one purpose for this post as I do not know anyone yet who owns one other than myself.
I'd have to say it's pretty typical.
dl4zbg
Posts: 42
Joined: Sun Apr 09, 2017 5:33 pm

Re: Anan 7000DLE Mk-II PA output.

Postby dl4zbg » Mon Apr 01, 2019 6:13 pm

Gary,

I did not expect this from the new MKII. I have been suffering from low output/ high IMD (without PS) on 10 an 17m with my 7000 DLE right from the beginning.

Good luck

Volker
User avatar
Gary_Sitton
Posts: 10
Joined: Mon Mar 04, 2019 6:35 pm

Re: Anan 7000DLE Mk-II PA output.

Postby Gary_Sitton » Tue Apr 02, 2019 8:05 pm

Volker:
I suspect this is just a sloppy tolerance issue with the impedance matching filter components for 17 and 10 meters. The distortion is the result of having to over-drive the PA into its non-linear region in order to increase the output. When I get a chance later, I'm going to put a scope on the output section and see where the problem lies. The issue usually is in the toroids whose ferrite characteristics or number of windings may have changed from the original design. For now I'm be happy that there seems not to be any other hardware issues. The RX is definitely much cleaner with fewer major birdies, and the TX cooling seems to be improved as well. Of course I'm comparing this all to my old Anan 100-D.
Gary
73, Gary K5AMH
User avatar
w9ac
Posts: 55
Joined: Sun Apr 09, 2017 4:01 pm

Re: Anan 7000DLE Mk-II PA output.

Postby w9ac » Mon Apr 22, 2019 2:20 am

Gary_Sitton wrote:Volker:
I suspect this is just a sloppy tolerance issue with the impedance matching filter components for 17 and 10 meters. The distortion is the result of having to over-drive the PA into its non-linear region in order to increase the output. When I get a chance later, I'm going to put a scope on the output section and see where the problem lies. The issue usually is in the toroids whose ferrite characteristics or number of windings may have changed from the original design. For now I'm be happy that there seems not to be any other hardware issues. The RX is definitely much cleaner with fewer major birdies, and the TX cooling seems to be improved as well. Of course I'm comparing this all to my old Anan 100-D.
Gary

FWIW, same here on 17m only. By backing off drive to 90W output, there's a substantial improvement with PS engaged as it goes from -55 dBc to -65 dBc. With PS disengaged, I'm seeing about -32 dBc at 100W and about -35 dBc at 50W. That's pretty much in line with the competition.

Paul, W9AC
User avatar
w-u-2-o
Posts: 1617
Joined: Fri Mar 10, 2017 1:47 pm

Re: Anan 7000DLE Mk-II PA output.

Postby w-u-2-o » Mon Apr 22, 2019 3:02 am

In looking at the design, the final amplifier stage, using dual RD100HHF1 FETs, has remained essentially identical since the original Munin amplifier board for the Hermes. The only thing that has changed in the leap into the 7000 are the driver stages. Abhi has posted that instead of the two driver stages used on the 200D and older designs, where there was a driver on the SDR board and another driver stage on the PA board, now there is only a single, higher gain driver on the Orion MKII. The driver stage on the PA board is gone.

This strongly suggests the problem is marginal gain at the higher frequencies in the driver stage that is on the Orion MKII board, since the final stage is a proven design at this point with well understood gain (although I can't say I know what the gain of each stage is, myself). The final stage is also proven to be reasonably well behaved into a variety of impedances (unlike the amp in the 8000, which gets crazy into high impedances).

It is possible that slightly changing the bias voltages on the driver and/or final stage (they are all adjustable) might squeak out that last missing dB (the difference between 80W and 100W is 1dB). But if this is happening to a lot of people it tends to suggest marginal gain in the driver stage on the Orion MKII.
User avatar
artcomm
Posts: 8
Joined: Fri Nov 30, 2018 8:23 pm

Re: Anan 7000DLE Mk-II PA output.

Postby artcomm » Wed Apr 24, 2019 1:58 am

I just sent an email to Anan support about this subject and very eagar to hear thier reply...
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Good Day, Anan Support

I want to address an issue with my new (Non-PC) 7000DLE MK-II ser#002 purchased 3 weeks ago. It has come to my attention after extensive testing with either Pro1 v2.4 w/ OPSDR Mrx 3.4.9 and or Pro2 v1.9 w/ Thetis 2.6.6 and setting 38.8 in PA Setting/PA Gain the radio does well over 100 watts on every band, except 17m is only 82 watts and 10m is only 74 watts max with CW Tune MOX carrier.

For this test, I have a brand new RG8X high-quality coax jumper 3ft, LP 100-A Digital Wattmeter, Palstar DL5K Dummy Load and an Astron 70M Power Supply @ 14.97 volts at 70 AMPS... Here are my findings with 4 Radios one at a time hooked up to the same test equipment >>>>

Yaesu FT-991A -- 104 to 106 Watts from 6m to 160m

Flex 6400 -- 103 to 109 Watts from 6m to 160m

Anan 200D -- first with Pro1 v2.4 and Mrx 3.4.9 and then Pro2 vNP1.8 and Thetis 2.6.6 with PG Gains set at 38.8 Highest output was 20m 148watts but all the other band well over 100 watts including (17m 114 watts) and (10m 109 watts.)

My last test with my new Anan 7000DLE MK-II (non-pc) PA Gain 38.8 Well over 100 watts on every band EXCEPT (17m only 82 watts) and (10m only 74 watts)...

So please do not blame my Equipment with three other radios testing well over 100 watts on every band, But my newest and Most Expensive radio, the 7000 is not TX'ing 100 watts on 2 bands as advertised...

So my questions are... Is this a Firmware issue? or a physical mechanical calibration that needs to be done on the new 7000 MK-II ????

Thank you for your time and please advise me how to address this issue and hoping other users can test the way I did and report their findings as well.

Art C. N4QNT '73
w3ub
Posts: 8
Joined: Fri Dec 14, 2018 6:32 pm
Location: Lunenburg, NS/Florida

Re: Anan 7000DLE Mk-II PA output.

Postby w3ub » Thu May 09, 2019 7:20 pm

Any reply from them? My relatively new 7000DLE (non-mkII) also has low power on those two bands.
/Doug
User avatar
Gary_Sitton
Posts: 10
Joined: Mon Mar 04, 2019 6:35 pm

Re: Anan 7000DLE Mk-II PA output.

Postby Gary_Sitton » Sat Jun 15, 2019 8:57 pm

OK, here are my results of adjusting the PA bias voltages and thus the PA idle or no-signal current. The total no-signal (idle) PA current is about 2.2 amps using the 7000DLE voltage/current display option. This condition can be simulated in the SSB mode by setting the microphone input to zero and using the MOX or PTT to key the amplifier. About 0.2 amps can be accounted for by the PA driver and other ancillary circuits leaving 1 amp for each side of the class AB push-pull MOSFET PA. I found by reducing the idle current separately by adjusting R5 and R6 on the PA board that I could achieve a total current level of 1.7 amps or 0.75 amps for each MOSFET. This small change improved the total PA current head-room by 0.5 amps which reduced the idle power dissipation by 7 watts = 14VDC*0.5amps. I found also that this previously wasted power is now available for added RF output power since I can now obtain > 100 watts on all bands except 10 mtrs having only 95 watts. I was able to run Pure Signal with excellent results showing < -50 dB IMD3 improving the -30 dB IMD3 without Pure Signal. I am assuming no cross-over distortion was introduced by lowering the PA bias currents since the distortion without Pure Signal was not increased from before. Apparently the factory bias current setting of 1 amps each for the PA is overly cautious and unnecessarily reduces the PA efficiency and increases PA heat dissipation. I would also expect that a more rigorous method using an oscilloscope, etc. for setting the bias would safely achieve even lower bias current values and efficiency.

73, Gary K5AMH
73, Gary K5AMH

Return to “Everything Else: Antennas, Relays, Switches, Power, Grounding, Cooling, etc.”