diseqc motor driving issues with one of the LNBs

[tom55]

Hi

I'm facing this problem ever since i got inverto wideband LNB, there are times when moving dish via diseqc commands hangs, hitting the button in software, either linux or windows - the dish does not move not even a step.

It is skyrevolt motor bought from germany, using PCTV 461e tuners, no any other equipment inline, if i connect it to IBU, PLL premium or old STi LNB there are no problems.
Or, if i disconnect wideband LNB from motor it immediately responds to commands and moves.

Read elsewhere it might be a cable problem, however i do get all channels, and obviously it works with other or no LNBs. Thought maybe a power issue, but this model consume less than IBU according to specs, and looking at the wall meter i can confirm it consumes less than IBU on 18V, with both of its ports active.
I figured out to try driving it via hw button on motor itself when it doesn't want to move, and surprise, driving either east or west for as long as needed works everytime without issues. This would rule out power problem.

Can't understand why would LNB block diseqc commands to motor. Any ideas?
To return it and ask for replacement?

 

[Channel Hopper]

Since the SkyQ LNB ports are unfiltered at sub IF frequencies, you will probably find the 22kHz signals are masked by the output at the wideband lower end.

You may find inserting a VHF cutoff filter restores operation (one side of a quality VHF/UHF diplexer with DC bypass)

 

[tom55]

so the 22khz passes through right into the LNB and reflects back into motor, this reflected signal is sometimes lost/corrupted and clashes with intended signal? i'll look into that, thanks

 

[ozumo]

How is it all connected up - do both wideband LNB outputs go through the motor? For example if only the horizontal output goes through the motor and you have a vertical channel selected the DiSEqC command may only be sent up the vertical cable, bypassing the motor.

 

[tom55]

only horizontal output goes through the motor (1input/output on the motor) to a pctv card that use one instance of altdvb, for H channels only
vertical output goes directly to another pctv card that is running separate instance of altdvb for V channels only.
thus, the dish moving is always operated from altdvb instance on H port

 

[Channel Hopper]

so the 22khz passes through right into the LNB and reflects back into motor, this reflected signal is sometimes lost/corrupted and clashes with intended signal? i'll look into that, thanks

Almost, the full output from the LNB (from 0 up to around 2.4GHz) is being sent into the gearbox of the motor, thanks to DC continuity. The DiSEqC circuit will never hear what is being sent from the receiver.

 

[deeptho]

Did you try sending the diseqc for the motor over the V-cable?
One possible problem could be that the diseqc commands are sent too soon after
powering up the lnb and motor. If that is the problem then it will be worse on the H polarisation.

 

[tom55]

yes, tried connecting V cable through motor, the problem remained. hard to tell if it was worse as it happens randomly and it is impossible to predict for how long it lasts

what i noticed through H cable, sometimes restarting dvb app or changing transponder can help it to respond, motor accepts one or few commands and then hangs again. when this doesn't help, unplugging card from PC for a few minutes may help, but then there are times when leaving it unplugged for entire day doesn't help

as the wideband LNB does not need to do hi/lo band switching does it even implement diseqc protocol ? if it does it could be 2.x version which is bidirectional and it might be sending some stupid response back down the cable. i see few 22khz blockers on google, probably best idea to try to get one

 

[Channel Hopper]

yes, tried connecting V cable through motor, the problem remained. hard to tell if it was worse as it happens randomly and it is impossible to predict for how long it lasts

what i noticed through H cable, sometimes restarting dvb app or changing transponder can help it to respond, motor accepts one or few commands and then hangs again. when this doesn't help, unplugging card from PC for a few minutes may help, but then there are times when leaving it unplugged for entire day doesn't help

as the wideband LNB does not need to do hi/lo band switching does it even implement diseqc protocol ? if it does it could be 2.x version which is bidirectional and it might be sending some stupid response back down the cable. i see few 22khz blockers on google, probably best idea to try to get one

Nothing within the SkyQ systen uses 22kHz protocol, stopping universal / legacy LNBs from being used. It opened up a can of worms with poor quality products as a workaround, along with callout/labour charges.

 

[a33]

One possible problem could be that the diseqc commands are sent too soon after
powering up the lnb and motor. If that is the problem then it will be worse on the H polarisation.

In that case the menu option for the diseqc "sequence repeat" would help.

I myself have for practical reasons connected my STAB motor not serially, but parallelly: with a power-pass both directions splitter, for just 5-1000MHz by the way. Functions without problem.
So this might also be a (testing) option, against the "sucking in" of the diseqc commands that Channel Hopper suggested.

I'll follow your testing results with interest!

Greetz,
A33

 

[tom55]

appears like some sort of power issue (not due to insufficient current on H line, but lack of power on V port of LNB ), usually when i would want to move dish i'd start only card on H pol. and this sometimes caused problems.

for a few days i was testing this with both cards plugged in and connected to external power supply, V pol card was powered via battery pack and for 2 days it was fine then it hanged again. a day after posting this thread i remembered to check battery it was drained out to 10.6V

then i hooked up both cards to small UPS i recently got and didn't experience any critical hangs yet (there is still bug with windows driver, when both cards are active motor gets stuck after few diseqc commands, but simple restart of altdvb solves it)
with neumodvb there were no hangs even with both cards tuned in to some transponder

also, the weather outside got pretty much colder during past several days. need to check with LNB at room temperature to exclude any possible hw failure

 

[a33]

So, this test result would be a bit contrary to your experience before, as written in your opening post?

.... and obviously it works with other or no LNBs. Thought maybe a power issue, but this model consume less than IBU according to specs, and looking at the wall meter i can confirm it consumes less than IBU on 18V, with both of its ports active.
I figured out to try driving it via hw button on motor itself when it doesn't want to move, and surprise, driving either east or west for as long as needed works everytime without issues. This would rule out power problem.

It still is confusing me, what is the matter exactly...

Greetz,
A33

 

[tom55]

LNB somehow fooling diseqc motor if it's V port is not connected (unpowered). it is lousy and cheap LNB design - if the V and H port circuitry were totally separated inside this wouldn't happen.
IBU consumption is ~2.9W on H, WDB only 1.9W on H and V port consumes 0.6W when both ports active, so ~2.5W in total
the problem seems to be the power from H port is also going to V port when V port DC supply is not connected, but the consumption doesn't increase, yet the V port channels can be tuned

 

[deeptho]

This isolation is often a problem on lnbs. I also saw it on some LNB yesterday when testing neumoDVB on one computer and watching CNN on TV. Each time I start neumoDVB a little bit of pixellation on CNN. It is one of those narrow neck lnbs.

About the WDB: I would argue that power consumption is not the only criterion, but rather
1) how does rotor react to the the transients at power up. In other words how fast does current increase and how well can equipment cope with that. I have see rotors go bonkers and stop working until reset. That is why neumoDVB never powers up from zero volt to 18Volt in one go. Instead it first goes to 12 V, waits a bit and then goes to 18V

2) what kind of transients does the lnb generate itself?

If possible, I would start by putting a scope on the system. I use a picoscope. I have cut a small hole in the mantle of some small extension cable, which allows me to grab the inner
conductor with a probe. This can disrupt RF of course, but that is immaterial for switching testing and I did not even notice an impact on the signals.

A workaround (alternative to the splitter solution) could to insert a diseqc switch between motor and lnb. Connect LNB to port 2, 3 or 4. FIrst send diseqc to control rotor. Afterwards send diseqc switch to connect it to port B. This may need software adjustment.

Or install the diseqc swicth before the rotor and connect the lnb to port 1 and the rotor
to port B. First select port B and tune the rotor. Then switch to A.

The annoying bit about the last workaround is that you need to know when the rotor has
stopped before cutting its power. Its advantage is that lnb influence is eliminated.

 

[tom55]

about question 1)

with neumoDVB (same with altdvb) the transponder in positioner console is already tuned before issuing diseqc command. so the power is on and at 18V. additionally, there was driver fix for this hardware ([v2] media: dvb-frontends: a8293: fix LNB powerup failure in PCTV 461e - Patchwork) to deal with exact same problem you described here, i have this applied and had no problems with that since then

about workaround

here's what i was able to find; there are 0/22 khz switches and 22khz blockers available (some of them don't have required frequency range)

a switch could probably do the job, but everytime you move a dish, signal from LNB will be cut for very short period of time - it can be annoying if you want to fine tune by moving only for a single step at a time

a blocker should filter out certain range within center of 22khz and shouldn't cause any signal disruption.

first thing you find is axing, highlights say 5-2400, detailed specs say 5-2200 MHz, and the supplier you contact has no idea which is true. so you avoid axing because it is average quality. there is also hama but not available unless you can find supplier