Success Stories

Discussion about this major DATV Project. See https://wiki.batc.org.uk/The_Portsdown_Transmitter
g0mjw
Posts: 2327
Joined: Sat Sep 20, 2014 9:15 am

Re: Success Stories

Post by g0mjw » Sat Jun 03, 2017 10:03 pm

G8GKQ wrote:Mike

Are you using a new Serit tuner? You might want to check that your phase noise problems are not in the receiver first. Put a 10dB pad on the tuner input and see if it improves matters at 70 cm and 146.

Yes, I'm about to need a bezel. Flat to an Aluminium panel. Has anyone on the Forum got a successful 3D print design for this?

Dave
Hi Dave,

I am but I had 20dB on the input - I had thought of that but this time no, I don't think its the receiver. I suspect it is noise on the supply from the DC-DC converters. Easy to check. Less easy to fix!

For the panel - if these 7" screens work as well as the recent reports, it would seem a much better solution. In my view, any screen intended for the PI that doesn't have mounting options is a bad design. I spent about an hour cutting out a hole for the screen and fabricating some clamps - but any clamping system is likely to put stress on the LCD.

Mike

g0mjw
Posts: 2327
Joined: Sat Sep 20, 2014 9:15 am

Re: Success Stories

Post by g0mjw » Sat Jun 03, 2017 10:09 pm

g8gtz wrote:Hi Mark

All looks great - it would make a good CQ-TV article :-)

Don't know how I'd missed the PGA144 from Sam, looks just what we need on 146 - I run the DG8 from GM3SEK but is no longer available - I'll update the wiki.

We can put the filter designs on the wiki for all to enjoy.

Noel
If anyone still has an old Mutek 2m preamp they have built in filtering and work very well in this application. There were several variants.

g0mgx
Posts: 31
Joined: Sat Dec 31, 2016 7:57 pm

Re: Success Stories

Post by g0mgx » Sun Jun 04, 2017 5:03 am

G8GKQ wrote: Has anyone on the Forum got a successful 3D print design for this?
I ordered one of these; dont have it yet so can't comment on how good or otherwise it is:

https://www.thingiverse.com/thing:1920486

Mark
G0MGX

M0YDH
Posts: 264
Joined: Sat Dec 31, 2016 5:59 pm

Re: Success Stories

Post by M0YDH » Sun Jun 04, 2017 8:45 am

Mark
I quietly had 2 printed at work as a training exercise. Please don't ask for more.
Fit is good to the LCD. Mounting to your Farnell case is easier ie 2 tapped M3 holes and an aperture nibbled in for the connector. My son spray painted it with 2 coats of black primer. Ergonomically it gives an easier reading angle plus Dave G7MEG and I found out the pushing angle helps the case stay still. I made a wooden packing piece to fill the space behind and miss the connections and IC's.

73
David M0YDH
Attachments
20170605_221955_M0YDH Portsdown LCD in frame.jpg
20170605_221955_M0YDH Portsdown LCD in frame.jpg (352.99 KiB) Viewed 6344 times
Last edited by M0YDH on Mon Jun 05, 2017 10:44 pm, edited 1 time in total.

g0mjw
Posts: 2327
Joined: Sat Sep 20, 2014 9:15 am

Re: Success Stories

Post by g0mjw » Sun Jun 04, 2017 12:29 pm

G8GKQ wrote:Mike

Are you using a new Serit tuner? You might want to check that your phase noise problems are not in the receiver first. Put a 10dB pad on the tuner input and see if it improves matters at 70 cm and 146.

Yes, I'm about to need a bezel. Flat to an Aluminium panel. Has anyone on the Forum got a successful 3D print design for this?

Dave
I suppose pictures don't lie - Serit does not work so well on 2m natively. The three images are all exactly the same set up and power levels and in degrading levels of performance. This is a direct cable connection, with attenuators. The first is the sharp with an up-converter. The second is the Serit with the same up-converter, the third is the Serit on 2m directly.
Sharp NIM - Portsdown.jpg
Sharp NIM - Portsdown.jpg (233.1 KiB) Viewed 6444 times
Serit NIM - Conveter - Portsdown.jpg
Serit NIM - Conveter - Portsdown.jpg (228.61 KiB) Viewed 6444 times
Serit NIM - Portsdown.jpg
Serit NIM - Portsdown.jpg (209.49 KiB) Viewed 6444 times
I could not get the Serit NIM to lock to the portsdown. However, it will lock to my digithin and my DATV Express, so there is something about the signal it doesn't like. It must be the spurious signals. I then put in a low pass filter. This recovered the situation considerably. The forum will not allow more than 3 screenshots, so I will put this in another message.

Moral - you can't expect the Serit to work on 2m without good filtering.

g0mjw
Posts: 2327
Joined: Sat Sep 20, 2014 9:15 am

Re: Success Stories

Post by g0mjw » Sun Jun 04, 2017 12:31 pm

The Serit after I fitted a low pass filter to the portsdown source.
Serit NIM - LPF - Portsdown.jpg
Serit NIM - LPF - Portsdown.jpg (229.19 KiB) Viewed 6445 times
I suppose the other thing to add is the Sharp Nim, even after an up-converter, has a much better MER.

Mike

G8GKQ
Site Admin
Posts: 2798
Joined: Sun Mar 22, 2009 1:21 pm

Re: Success Stories

Post by G8GKQ » Sun Jun 04, 2017 7:37 pm

For information, Noel and I have done some tests of the relative performance of the Sharp and Serit tuners, and come to the conclusion that whilst the indicated MERs might be different, the weak signal performance is indistinguishable: See here: http://batc.org.uk/forum/viewtopic.php? ... 725#p12725

Dave, G8GKQ

vk5bd
Posts: 35
Joined: Thu Dec 08, 2016 10:17 am

Re: Success Stories

Post by vk5bd » Mon Jun 05, 2017 2:09 am

Hello Mike,

While it is not the cause of your issues, I do see that in your received images that there is the lightning bolt showing up, this indicates low power supply to the PI, I had reported what I had believed to be an issue with the modulated signal but it turned out the Pi was struggling due to the power supply not being stable, just thought I would mention it as something to keep in mind if you have troubles.

Regards,
Bevan

g0mjw
Posts: 2327
Joined: Sat Sep 20, 2014 9:15 am

Re: Success Stories

Post by g0mjw » Mon Jun 05, 2017 5:14 am

vk5bd wrote:Hello Mike,

While it is not the cause of your issues, I do see that in your received images that there is the lightning bolt showing up, this indicates low power supply to the PI, I had reported what I had believed to be an issue with the modulated signal but it turned out the Pi was struggling due to the power supply not being stable, just thought I would mention it as something to keep in mind if you have troubles.

Regards,
Bevan
I was aware of this and you are right its not relevant. The Portsdown is working just fine with a very good MER as can be seen from the top image. The problem, if there is one, is the Serit is more sensitive to harmonics and other signals in the band than the Sharp. Especially when tuned to low frequencies outside its nominal range. The solution is filtering. At lower SNRs the MER is just as good. The Serit doesn't seem to get as good a constellation at high MER but this is largely irrelevant because its a digital signal with forward error correction. As you will see with phase noise, the presence places a minimum on the ultimate symbol error rate, but as long as this is well below what the FEC can deal with, it doesn't really matter. I have not bothered to do the maths but you can probably make an AWGN like assumption so as long as the MER with phase noise its well above the threshold, it's not going to matter and channel noise will almost always dominate.

However, I have ordered a new higher current USB cable, the voltage drop on the one I cut up as a temporary measure, which is only 4" long, is nearly 0.25V! at high load, so even with the 3A PSU set to 5.1V it occasionally triggers. Hopefully the new cable will fix this but 2A seems an awful lot for such a small connector. I am tempted to wire directly to the PI PCB.

Mike

G4FRE
Posts: 197
Joined: Fri May 26, 2017 6:57 am

Re: Success Stories

Post by G4FRE » Mon Jun 05, 2017 3:01 pm

g0mjw wrote:
Hopefully the new cable will fix this but 2A seems an awful lot for such a small connector. I am tempted to wire directly to the PI PCB.

Mike
That is what I had to do to resolve a similar issue with Volts drop down the USB cable on the RPI3 on my PiHPSDR

Dave
G4FRE

Post Reply

Return to “The Portsdown Digital ATV System”