Device malfunctions when 22khz tone is switched on

[bohan]

I purchase 6 x QBOX3 (TBS 5921) a month ago after having very stable performance and good build quality from TBS 8920 for over a year. Since then, I have been unable to get them to work properly when going through any type of distribution multiswitch. My problem appears to be similar to this:

https://shop.tbsdtv.com/forum/viewtopic.php?f=21&t=196

I believe there has there has been a design error in the TBS 5921, just like the TBS 6920 (notice that the rep never reported back at the end whether the returned card was indeed defective be design, and just quietly replaced the unit with an improved , or should I say corrected, TBS 6921 card). When using 22Khz=on, the device sends some disruptive signal that shuts down the multiswitch, causing signal loss on all ports until the device is off, reset or unplugged.

Before we continue I wish to remind everyone the 22khz switching is a VERY BASIC and universally used distribution switching method found in SMATV systems, universal LNBs, DirectTV multiswitches, etc. If a DVB-S tuner product does not work with 22khz tone, switching, it might as well be an empty box.

I was told by support to test one card at a time. In the first week I was reluctant to do this, as, if it does world that way, what does it mean? That I can only use one QBOX3 in the SMATV network???

After spending a week trying Windows XP and Windows 7 on two machines (one Atom, one i3, clean installs - nothing on them) with DVBLink, DVBViewer, TBSViewer and SageTV, I was unable to get a working setup for the 22khz=on port (the 22khz off ports always workd fine). Sometimes is will scan about halve the channels, sometimes none. With DVBLink, I once manage to scan all the channels, but actual use is not reliable - just switch the channel a few times between 0/22Khz channels and boom - stops working. Restarting does not help. 2 days wasted.

In the second week, I thought, maybe I will follow the advice of support, so I try just one device - same results - another 2 days wasted.

In the third week, I thought maybe it is my hardware. So I got out by regulated linear lab PSU, set it to 7.5V, and plugged it into the Qbox3. I then got out my Promax TV Explorer (Prodig 5), and painfully verified the signal quality (50dB<power<85dB, VBER < 1E-7) of each transponder on all my 4 LNB signal lines - both before the inputs and after the outputs of my DirectTV approved, top quality Zinwell MS6x8WB multiswitch (which by the, way, has been running an Openbox S9, Mediastar HD1100s and a Mediastar HDX9100 with zero pixelation of dropout, throughout these weeks). I when and got an Intel D525MW mainboard with integrated Atom CPU, 2GB a Kingston RAM, and a Kingston S100 SSD. After installing windows and the Intel official drivers, plus all updates, I rebooted, and downloaded the latest TBSViewer and 1.0.0.2 drivers off tbsdtv.com. After plugging in Qbox3 and installing the drivers, I rebooted and ran TBSViewer to do a scan of Chinasat 6B, I set LOF1=5150, LOF2=5150, LOF SW=3400 (min C-band freq) as I have the V/H independent outputs from my commercial LNB connected to the 22khz=on ports of the Zinwell MS6X8WB. After doing a scan - same problem - no lock on about half the transponders. Exited TBSViewer, rebooted and tried again - this time, the same problem as the other guy with his TBS 6920 - signal loss on all ports of multiswitch until Qbox3 is shut down. Unplug the same cable and plug straight into the Mediastar HD1100s - perfect picture - 99% signal quality.

One of the many transponders that it fails to lock on 22khz-on port is Shanghai (3808Mhz, H, 8800Ms/s), but the bigger problem is the emitted signal that causes multiswitch to fail. I will try to measure this with my scope - by why waste so much time on the faulty hardware - where is you test Engineers TBS? Your hardware is great quality - with solid caps, well moulded cases and overengineered 265V PSU, why waste all the hard work on these little problems. They are probably just a little bug in you driver - but look, you did admit to any problem with you product, and now I have wasted 3 weeks, which I could have used elsewhere. We could have done a lot of business together, but you ruined it badly.

Just a note: I have just tried a 2 year old TT-budget S2400 unit, which worked straightaway after plugging in to the same USB port on the same PC and installing drivers. but Qbox3 have much better chipset and S2 support, plus better build quality - but does not work - such a waste, such a pity!

[cody]

your post has been brought to my attention and while i'm not sure for your problem with 22 kHz i want to comment on few things:

I believe there has there has been a design error in the TBS 5921, just like the TBS 6920 (notice that the rep never reported back at the end whether the returned card was indeed defective be design, and just quietly replaced the unit with an improved , or should I say corrected, TBS 6921 card).

there is no any design error in TBS 6920 - please, don't make statements, which are simply not true - we were selling TBS 6920 for 3 years (since 2008) and it was recently replaced by newer model TBS 6921. however, do you think product with "design error" can be successfully sold for period of 3 years and further more that we shouldn't have fixed them for such a lot time. so, you're making some wrong generalizations based on wrong assumptions. also, TBS 6921 is not supposed to be "corrected" TBS 6920, but it's entirely new model.

verified the signal quality (50dB<power<85dB,

something is definitely wrong with your report: you have over 50 dB of signal?! either it's 5.0 dB to 8.5 dB, which is very weak signal and in fact it will be outside C/N requirements by the DVB-S/S2 specification for most transponders depending on their other signal parameters like FEC, etc:

http://www.satellite-calculations.com/Satellite/Eb_No.htm

or measurement unit is not "dB" but something else.

the same problem as the other guy with his TBS 6920 - signal loss on all ports of multiswitch

again, you're talking for a single trouble case with TBS 6920 - no any other customer using multiswitch except that particular customer was having such problem and it is clearly some isolated case - for 3 years product live span of TBS 6920 there are no any such cases.

so, as i mentioned while i'm not sure about your problems with 22 kHz all parameters of 22 kHz tone can be controlled by the driver - i will re-check the settings there.

[pjm]

I believe there has there has been a design error in the TBS 5921, just like the TBS 6920 (notice that the rep never reported back at the end whether the returned card was indeed defective be design, and just quietly replaced the unit with an improved , or should I say corrected, TBS 6921 card). When using 22Khz=on, the device sends some disruptive signal that shuts down the multiswitch, causing signal loss on all ports until the device is off, reset or unplugged.

Before we continue I wish to remind everyone the 22khz switching is a VERY BASIC and universally used distribution switching method found in SMATV systems, universal LNBs, DirectTV multiswitches, etc. If a DVB-S tuner product does not work with 22khz tone, switching, it might as well be an empty box.

fwiw I have the TBS 6921 and the 22khz switching has never worked on my multiswitch either. Not sure if is is a hardware or a driver fault.

https://shop.tbsdtv.com/forum/viewtopic.php?f=23&t=203&start=0

[bohan]

To cody:

Firstly, I want to apologise for being rude regarding your product design. I wasted a whole week testing the TBS 5921m so was very angry that I could not get is to work.
Secondly, when I stated that 50db<power<85dB, I rely meant 50dBuV<RF level<85dBuV. However, if you install satellite TV systems often and use a professional analyser such as Promax, Ikusi or Onaohm, you would know that the instruments all (by convention) call the dBuV figure "power". Please read this instructive article by Maxpeak, which will hopefully clarify things for you:

http://www.maxpeak.tv/articles/When_is_a_signal_good_enough.pdf

You will find that 50dBuV<power<85dBuV is the ideal signal level. In either case, my point was that the issue was not the quality of my input signal. I run two TBS 6921 in one of my servers I once tested a TBS 6920 in that same PC, and the 22Khz switching did not work. At that time, I had some single satelite fixed dish customers, and just used the TBS 6920 for them. However, for many clients, I installed dual LNB, dual satellite C-band systems for them, and 22khz switching is the most reliable method in this case, as no DiSEqC version or commited/uncommited issues need to be dealt with.

To date, the most stable TBS products for me are still the TBS 8920 and TBS 6921 - I have a couple dozen of these out in the field at clients premises and they never crash - interesting, I never had any driver issues with them either - interesting, wouldn't you say.

Turbosight has shipped a TBS 5922 for me to test, sighting that is fixes the low symbol rate locking issues on the TBS 5921. I will test it tomorrow and hope it works well, because I am starting to provide energy efficient Atom based servers to clients and they do not have many slots, so I must find a reliable USB solution. As a temporary measure, I am shipping TT S2-3600 units, which work well, but the circuit hardware quality is not as good the TBS units, hence why I took the time to raise the issue.

Cody - you must understand that I spend my time to post here because I care about Turbosight. If I did not, I would have just got a refund and just moved on to TT, TechniSat, Prof or whatever. But no, I persist in trying the get the TBS solution working as I like the way you run you business (active windows and linux driver development, real technical support provided by engineers).

I develop electronics myself, so I know what is like - ICs are full of bugs all the time (e.g. PICs). The chip companies are always in a race and often realise ships with immature firmware which have to be patched. Errata notes for hardware design flaws occur all over the place - most of the time they have workarounds but it is always a huge pain to have to go through all the Errata pages and sometimes, when time is running short, you do not get a chance to investigate all the problems with you chosen parts - you just test the finished product in you most important market environment - which may not be the actual field use for some of your customers.

[steven]

Hi bohan
Thanks for your trust,cody and all turbosight colleague will help you to solve the problem you meet,and would you mind to tell us what about the tbs5922?
Thanks