S Meter issue ( FIXED!)

User avatar
WM4CH
Posts: 59
Joined: Wed Nov 09, 2022 1:20 pm
Location: Cape Cod, MA

S Meter issue ( FIXED!)

Postby WM4CH » Fri Dec 23, 2022 3:57 pm

I have a used 7000 DLE MKII that I just got.

The S meter reads 54 db LOWER than my IC-7300, opposite of what is expected based on the locked S meter thread.

dbm levels are very low for what sound like very strong signals. I have yet to see anything above S9 on any signal. and that is very rare.
Most signals are S2-S7

The spectrum scope signal levels are the same as the S Meter reading -dbm.

My son and I talk with 100 watts each, my signal on his ANAN-100D is S9+20DB
His signal on my ANAN-7000 DLE MKII is S4

My ANAN noise floor on 80 meters ( my nosiest band ) on the S meter is S2, on the IC-7300 it is S9

This is at 0 DB attenuation.

If I activate the attenuation the Noise level on the S Meter goes UP. It is supposed to stay the same as the radio is smart enough to compensate for the attenuation selection.

on 80m 0DB attenuation, the noise level is S2

at 30DB acttenuation, the noise level is S7.

Noise and signal levels are correspondingly similiar on other bands. ie 20 meter noise is S0-1 on the anan.

Do I need a power calibration?
or something else?
Could the receiver be damaged?

what can I look at or try?

any help is appreciated. thanks.
Last edited by WM4CH on Sat Mar 11, 2023 2:48 pm, edited 3 times in total.
w9mdb
Posts: 446
Joined: Sun Apr 09, 2017 5:53 pm

Re: S Meter issue ( old S meter thread is locked )

Postby w9mdb » Fri Dec 23, 2022 5:12 pm

First thing is you can't compare to the 7300 as it's S meter is almost useless. You'll note that the 7300 meter will change based on your RF gain which an S meter is not supposed to do. But most all rigs (non-SDR) do this.

You can't compare from one location to another either. Noise levels can be quite different.

I had a similar problem though and had to do a reset of the config to get the normal levels back.
Mike W9MDB
User avatar
WM4CH
Posts: 59
Joined: Wed Nov 09, 2022 1:20 pm
Location: Cape Cod, MA

Re: S Meter issue ( old S meter thread is locked )

Postby WM4CH » Fri Dec 23, 2022 5:36 pm

This is same location same antenna, same RF rooting.

Based on the ambitious S meter and Scots posts, the expectation is the ANAN will be more accurate and therefore show higher noise levels not substantially lower noise levels.

I had not thought of doing a reset of the configuration file. Thanks.
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: S Meter issue ( old S meter thread is locked )

Postby w-u-2-o » Fri Dec 23, 2022 6:00 pm

This sounds like a database problem, especially given your observations regarding the attenuator. Try exporting your current database in case you want to go back to it, then reset the database back to defaults.

Is this on PowerSDR or Thetis? If Thetis, then this belongs in the Thetis sub-forum and it will be moved there.

The reason the "old thread" is locked is because that is intended to be a tutorial thread, not a discussion thread.
User avatar
WM4CH
Posts: 59
Joined: Wed Nov 09, 2022 1:20 pm
Location: Cape Cod, MA

Re: S Meter issue ( old S meter thread is locked )

Postby WM4CH » Fri Dec 23, 2022 9:51 pm

My mistake.

This issue is with using Thetis
User avatar
WM4CH
Posts: 59
Joined: Wed Nov 09, 2022 1:20 pm
Location: Cape Cod, MA

Re: S Meter issue ( old S meter thread is locked )

Postby WM4CH » Fri Dec 23, 2022 9:54 pm

Resetting the database had no effect.

Listening to a strong signal on 20 meters.

and changing nothing but what radio is connected to the coax

The IC-7300 shows S9+20
Changing to the ANAN it shows S7

This is while the same ham is talking.
User avatar
WM4CH
Posts: 59
Joined: Wed Nov 09, 2022 1:20 pm
Location: Cape Cod, MA

Re: S Meter issue ( old S meter thread is locked )

Postby WM4CH » Fri Dec 23, 2022 10:03 pm

I have someone coming over with nice test equipment so we will calibrate the receiver.
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: S Meter issue ( old S meter thread is locked )

Postby w-u-2-o » Fri Dec 23, 2022 11:15 pm

If a reset database did not get you in the ballpark then something is broken.

When you calibrate with a known good, calibrated signal source if you are off by more than 10dB before calibration then something is broken.

Have you checked all the external connections?
User avatar
WM4CH
Posts: 59
Joined: Wed Nov 09, 2022 1:20 pm
Location: Cape Cod, MA

Re: S Meter issue ( old S meter thread is locked )

Postby WM4CH » Sat Dec 24, 2022 2:01 am

All external connections are exactly the same as the IC-7300.

All I do is un-connect the coax from the IC-7300 and connect it to the ANAN other than power and ethernet. The only antenna selection is #1 and the antenna is connected to 1

If the previous owner hit the start button on the calibration page with no calibrated signal input would that screw up the calibration, and would that get reset to factory setting by a database reset or only by a good calibration?

I calculate it is of by about 27 db compared to my IC-7300 which we all know is probably not that accurate.

I should be able to input a known calibrated input and frequency and see what the radio reads. Then do the calibration proceedure and see if it gets fixed or not. I am hoping that is all that is wrong.

Maybe something is broken, but based on the 2 radio bandscopes, it looks like both radios are seeing the same signals so it looks like the receiver is working properly, just that the signal strengths from noise through strong signals are being reported way too low.
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: S Meter issue ( old S meter thread is locked )

Postby w-u-2-o » Sat Dec 24, 2022 2:23 am

WM4CH wrote:All I do is un-connect the coax from the IC-7300 and connect it to the ANAN other than power and ethernet. The only antenna selection is #1 and the antenna is connected to 1

On thing you might try is to move things to ANT2 to see if the ANT1 relay might be misbehaving.

If the previous owner hit the start button on the calibration page with no calibrated signal input would that screw up the calibration

Yes.

and would that get reset to factory setting by a database reset or only by a good calibration?

A database reset would fix the bad cal. A new cal (done properly) will fix the cal.

I should be able to input a known calibrated input and frequency and see what the radio reads. Then do the calibration proceedure and see if it gets fixed or not. I am hoping that is all that is wrong.

My 8000 was off by about 6dB when I calibrated it using a known good 10MHz signal source set to -73dBm (S9).

It will be interesting to hear how this turns out. What are you going to use as a signal source and how do you know it is trustworthy?

I used a the 10MHz output of my GPSDO and padded it down to -73dBm. This was checked using a borrowed spectrum analyzer that had a certified calibration (borrowed from a commercial workplace that does regular commercial lab cal. cycles on all their equipment).
User avatar
WM4CH
Posts: 59
Joined: Wed Nov 09, 2022 1:20 pm
Location: Cape Cod, MA

Re: S Meter issue ( old S meter thread is locked )

Postby WM4CH » Sat Dec 24, 2022 2:51 am

My son is coming for the holidays and is bringing a descent signal generator that he checked with a good spectrum analyzer.

Thought about seeing if any difference with a different antenna port. I will try that in the morning.
User avatar
WM4CH
Posts: 59
Joined: Wed Nov 09, 2022 1:20 pm
Location: Cape Cod, MA

Re: S Meter issue ( old S meter thread is locked )

Postby WM4CH » Sun Dec 25, 2022 11:42 am

UPDATE:

It is fixed.

We put a -70dbm signal into the radio and it displayed as a-97dbm signal in the spectrum scope and the meter.

Hit the START calibrate button and all the values are back to what I would expect.

Noise level is now what I expect and S signals look good.

Put the -70 dbm signal in and it reads -69 dbm.


The receiver calibration settings are definitely not reset by resetting the database, which makes sense because most hams can not do the calibration themselves and would pay someone to do it, then have their calibration settings screwed up by loading a new database. That would be counter productive.
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: S Meter issue ( old S meter thread is locked )

Postby w-u-2-o » Sun Dec 25, 2022 11:51 am

WM4CH wrote:The receiver calibration settings are definitely not reset by resetting the database.
That is not correct. Reset your database now and watch what happens.

The hardware is not calibrated at the factory. Not for amplitude, not for frequency, not for amp gain, not for drive. This issue is one of the greatest failings of the design.

If you were off by 27dB then almost certainly you have a hardware problem of some sort. All you have done is put a shabby Band-Aid on it.
User avatar
WM4CH
Posts: 59
Joined: Wed Nov 09, 2022 1:20 pm
Location: Cape Cod, MA

Re: S Meter issue ( UPDATE: fixed )

Postby WM4CH » Sun Dec 25, 2022 1:32 pm

I will try resetting the database to see what happens.

I also have his database, I should try loading his and see what happens.

If the hardware is not calibrated at the factory, then the Thetis software has default calibration settings? Why isn't there a reset to default selection on the calibration page?
User avatar
WM4CH
Posts: 59
Joined: Wed Nov 09, 2022 1:20 pm
Location: Cape Cod, MA

Re: S Meter issue ( UPDATE: fixed )

Postby WM4CH » Sun Dec 25, 2022 2:00 pm

I reset the database. back to the original problem

Loaded the database after calibration, back to proper signal levels.



With calibrated database:
My noise level does not drop between the antenna connected and the antenna not connected on the S meter. Constant -84dbm on the S meter
but noise floor shows -117dbm on the spectrum scope.

I will get other numbers later.

Looks like I bought a damaged radio.

Any idea on where and what the failed components might be. Where to start looking?
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: S Meter issue ( UPDATE: fixed )

Postby w-u-2-o » Sun Dec 25, 2022 2:38 pm

Sorry to be the bearer of bad news, especially during the holidays :( But it seemed important to ensure you eventually got a good radio.

AFIAK, it's usually a bad relay somewhere on the RF board. One way to debug it is to use your signal generator to inject at the Orion RX input connector. If it's good there then you know it's on the RF board somewhere.

Do you have the schematics for the unit?

Doug, W5WC, is the US field service rep. He does excellent and quick work, if you decide to send it out. The best way to get in touch with Doug is: dwigley@apache-labs.com or support@apache-labs.com

eta: there is a reset button for the level cal...

Capture.JPG
Capture.JPG (53.34 KiB) Viewed 4775 times
User avatar
WM4CH
Posts: 59
Joined: Wed Nov 09, 2022 1:20 pm
Location: Cape Cod, MA

Re: S Meter issue ( UPDATE: fixed )

Postby WM4CH » Sun Dec 25, 2022 3:18 pm

Ok thanks for the info.

I think the schematics are on this website someplace. I will find them.

There is no reset button on my Thetis calibration page.

A bad relay I can find and replace. A bad op amp I can replace.

So it seems like the sensitivity of the receiver is down 29db since the 'calibration' also increased the noise floor with no input about 30 db, so that is consistent with the -70dbm signal being down 29 db.
User avatar
WM4CH
Posts: 59
Joined: Wed Nov 09, 2022 1:20 pm
Location: Cape Cod, MA

Re: S Meter issue ( UPDATE: fixed )

Postby WM4CH » Sun Dec 25, 2022 3:22 pm

Thought I had an old copy of Thetis, but am running 2.9.0.3

15403A1B-97D9-4191-ACB7-39D59FB57DDE.jpeg
15403A1B-97D9-4191-ACB7-39D59FB57DDE.jpeg (20.26 KiB) Viewed 4766 times
User avatar
WM4CH
Posts: 59
Joined: Wed Nov 09, 2022 1:20 pm
Location: Cape Cod, MA

Re: S Meter issue ( not fixed )

Postby WM4CH » Sun Dec 25, 2022 11:06 pm

Now I am running 2.9.0.6 and it has the reset button
User avatar
WM4CH
Posts: 59
Joined: Wed Nov 09, 2022 1:20 pm
Location: Cape Cod, MA

Re: S Meter issue ( not fixed )

Postby WM4CH » Mon Dec 26, 2022 5:32 am

doing some troubleshooting
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: S Meter issue ( not fixed )

Postby w-u-2-o » Mon Dec 26, 2022 1:21 pm

I thought I had seen a post from you saying that the RX2 path (assuming you mean the path to the second ADC) was OK, but not RX1/first ADC, that RX1 was good at the Orion board input, and that the RX1 path problem was band independent. If so then that helps a lot to narrow down the problem. It shouldn't take much more probing to find the bad component (probably a relay).
User avatar
WM4CH
Posts: 59
Joined: Wed Nov 09, 2022 1:20 pm
Location: Cape Cod, MA

Re: S Meter issue ( not fixed )

Postby WM4CH » Wed Dec 28, 2022 4:18 pm

Yes, Here is the info in case it can help anybody else.

Block diagram can be followed and that is in the Thetis Manual.

I rewrote this part because I misspoke on the original post.

receive paths that are attenuated:
ANT1,2,3, input through any filter including the bypass of the filters path through RX1 BPS1 has an attenuation of 30 db of signal strength, including noise into the RX1 Receiver. So the attenuation is not due to a filter.
EXT1 input that bypasses the T/R Relay also has a 30 db attenuation into the RX1 receiver. So the attenuation is not due to the primary T/R Relay.

RX2 port input which goes through RX2 BPF2 filters into RX2 Receiver is normal.

Here is the clincher.
Rx ByPass Port bypasses all of the T/R relays, bypasses the filters, bypasses the LPF RLP40+, bypasses the Internal PS Relay RL19 and goes straight through to one side of the External PS relay and is normal into the RX1 Receiver with no attenuation of the signal or the noise.

So that leaves only several components.

RL17 as mentioned above, even though through the BYPASS port connection it is good. no way to test the other side until I open it up.

RL19 Internal Pure Signal input Relay. Internal PS seems to work just fine. No way to test until I open it up.

LPF RLP40+. DC to 40 Mhz low pass filter.

RL22 relay to allow EXT1 path instead of ANT1,2,3 path. Both EXT1 and ANT paths are attenuated.

GBLC03C bidirectional protection diodes.

Since both the EXT1 path and the ANT1,2,3 paths are both attenuated, that excludes the primary T/R relay K5

PS seems to work fine, so for now excluding RL19 , but could have a poor contact

Could be RL22, but both contacts being bad would be rare I think.

So the primary suspects are down to 2, which will be tested first once I open it up.

LPF RLP 40+. I don’t know if this would fail and produce a consistent 30 db attenuation but it is conceivable that something in it could short or open and make it a DC to forever -30db filter instead of a low pass filter.

GBLC03C might fail and produce a 30 db attenuation but not sure how. It is opposing diodes both pos and neg so it is normally open. Failure would be a DC short to ground,( an open path would look normal but offer no protection to static ) but maybe not a complete RF short to ground and maybe cause a 30 db loss at RF, either way not sure how that could produce a 30 db attenuation in the signal path.But I had an FT-891 that had a protection diode get damaged from a strong signal and the receiver could only hear very strong signals. So this is my primary suspicion

The rest of this will have to be with the radio opened up and leads placed to see where the input signal drops 30 db.
User avatar
N3GX
Posts: 5
Joined: Wed Oct 05, 2022 6:22 pm

Re: S Meter issue ( not fixed )

Postby N3GX » Mon Jan 02, 2023 8:44 pm

On my ANAN 7000DLE MKII (latest serial number) works great but it looses the calibration and reverts to default as soon as the Thetis v2.9.0.6 x64 (4/23/22) is closed. By default Thetis reports about 7dBm in excess, (a calibrated -73 turns into -66 dBm.).
I have to run the calibration routine every time the Thetis is loaded. I have reset the database to default to see if it has to do with corrupted data but it still does not work. I use an ELECRAFT XG2 signal generator.
Anyone is having this problem?
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: S Meter issue ( not fixed )

Postby w-u-2-o » Tue Jan 03, 2023 5:11 pm

N3GX wrote:On my ANAN 7000DLE MKII (latest serial number) works great but it looses the calibration and reverts to default as soon as the Thetis v2.9.0.6 x64 (4/23/22) is closed. By default Thetis reports about 7dBm in excess, (a calibrated -73 turns into -66 dBm.).
I have to run the calibration routine every time the Thetis is loaded. I have reset the database to default to see if it has to do with corrupted data but it still does not work. I use an ELECRAFT XG2 signal generator.
Anyone is having this problem?

Windows 10 Pro, Thetis 2.9.0.6 x64, ANAN-8000DLE, Firmware 2.1.18

I cannot duplicate your problem. I just dragged out all of my calibration equipment and there is no problem with the level calibration on my installation.

When calibrating, pay attention to the S-meter, not necessarily the panadapter. Calibrate in Sig Avg mode. Make your bandwidth 500Hz.

When looking at the panadapter, ensure that the detector mode is average, the zoom is all the way zoomed in (minimum bandwidth), and that the bin width is adjusted to maximum. This should get the S-meter and the S-meter within 1dB of each other.
rdwing
Posts: 92
Joined: Sat May 21, 2022 2:05 am

Re: S Meter issue ( not fixed )

Postby rdwing » Wed Jan 04, 2023 7:11 am

N3GX wrote:On my ANAN 7000DLE MKII (latest serial number) works great but it looses the calibration and reverts to default as soon as the Thetis v2.9.0.6 x64 (4/23/22) is closed. By default Thetis reports about 7dBm in excess, (a calibrated -73 turns into -66 dBm.).
I have to run the calibration routine every time the Thetis is loaded. I have reset the database to default to see if it has to do with corrupted data but it still does not work. I use an ELECRAFT XG2 signal generator.
Anyone is having this problem?


Yes, my radio also exhibits this issue. Mine is about 6 dB high, reading -64 dBm when fed a -70 dBm signal. My radio also loses the calibration every time Thetis is closed.
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: S Meter issue ( not fixed )

Postby w-u-2-o » Wed Jan 04, 2023 11:40 am

It appears I can't duplicate the problem because it does not occur when carrying over a preexisting calibration like I am.
User avatar
DL2XY
Posts: 116
Joined: Sun Jul 30, 2017 9:47 am

Re: S Meter issue ( not fixed )

Postby DL2XY » Thu Jan 05, 2023 10:58 pm

N3GX wrote:. By default Thetis reports about 7dBm in excess, (a calibrated -73 turns into -66 dBm.)


Do not calibrate at -73dB !
This is an ADC-System, ADC's need to be calibrated near fullscale. -20dBm should be fine.

Calibrating at S9 means 1/1000 of fullscale voltage, so the software must extrapolate by a factor of 1000 meaning all measurement errors are multiplied by this factor too.
User avatar
ramdor
Posts: 1468
Joined: Wed Jul 03, 2019 3:07 pm
Contact:

Re: S Meter issue ( not fixed )

Postby ramdor » Fri Jan 06, 2023 2:34 pm

Just to add, if you performed a level calibration with 2.9.0.5 or versions before then that value will be saved in your db and used fine. However, if you run a calibration with 2.9.0.6 unfortunately the value is not saved.

This was fixed back in may/june 22, but I didn't get around to releasing a version that used the fix before I went awol. That said, a number of 2.9.0.7 releases have been made by various people and have entered the wild. As long as those versions used the last push of code to github back in mid 2022 then they should be ok as well.
Chiron
Posts: 11
Joined: Sat Feb 26, 2022 8:00 am

Re: S Meter issue ( not fixed )

Postby Chiron » Sun Jan 08, 2023 6:12 pm

ramdor wrote:Just to add, if you performed a level calibration with 2.9.0.5 or versions before then that value will be saved in your db and used fine. However, if you run a calibration with 2.9.0.6 unfortunately the value is not saved.

This was fixed back in may/june 22, but I didn't get around to releasing a version that used the fix before I went awol. That said, a number of 2.9.0.7 releases have been made by various people and have entered the wild. As long as those versions used the last push of code to github back in mid 2022 then they should be ok as well.



Can it been solved by using (going back to) 2.9.0.5 and do the calibration and save it. Then go to version 2.9.0.6 again?
User avatar
ramdor
Posts: 1468
Joined: Wed Jul 03, 2019 3:07 pm
Contact:

Re: S Meter issue ( not fixed )

Postby ramdor » Sun Jan 08, 2023 6:14 pm

Chiron wrote:Can it been solved by using (going back to) 2.9.0.5 and do the calibration and save it. Then go to version 2.9.0.6 again?


yah probably, but have not tried myself

Return to “Thetis”