Tevii S660 USB Satellite Tuner for PC
I've got one of these little usb satellite tuners that's started to play up. I'm not quite sure where the problem is but I've narrowed it down to a couple of areas, and I'm including possibly more background than necessary on the off-chance it might help someone else in the future. And yes, I know that this is beyond economic repair. A new one is only £50 or so, but I just got curious.
First a bit of technical background (sorry if this is teachin your grandma to suck eggs stuff - I'm trying to keep it to the minimum). A TV aerial is just a bit of wire that sends the signals down to the receiver where all the hard work is done. A satellite dish (well actually the LNB sitting at the focal point) does a bit more work, and as a consequence there are a few different signals going up and down the cable. First of all the signal that the LNB picks up from the satellite is in the range of 10.7 to 12.75GHz . In order to pass this down a cruddy piece of co-axial cable, they do some clever processing with the output from a crystal to drop it to a range of 950 - 2150 MHz. So that's signal one. Of course the LNB needs power to do this so "signal" two is a 13V DC supply coming up the cable from the receiver. Then at some point they realised they needed more bandwidth, so stuck a second crystal in the LNB at a slightly different frequency. A 9.75 GHz crystal is used to process signals in the range 10.7 - 11.7 GHZ, and a 10.6GHZ to process 11.7 - 12.75 GHz. To switch between the two crystals they use a 22KHz tone at 0.5V above the 13V power supply. That's signal three. Finally they realised that they could cram the broadcast frequencies more tightly together if adjacent frequencies were polarised in different directions. That's Horizontal and Vertical in Europe (but circular for anyone in the US). To switch between those two we have "signal" four. The receiver supplies 13V when it wants to receive Vertical frequencies and 19V for horizontal ones.
So at any time the receiver is supplying the LNB with one of four possible supplies:
13V - Low frequency vertical polarisation
13V + 0.5V 22KHZ - High frequency vertical polarisation
19V - Low frequency horizontal polarisation
19V + 0.5V 22KHZ - High frequency horizontal polarisation
My little receiver only outputs 19V, no matter what it is set for, and only picks up horizontal channels. That shows about 0.3 V higher when on a high frequency channel. Clearly the 19V supply and the 22KHz tone work fine, but the 13V supply has a problem. I'm just not sure where.
Here are a couple of pictures of the front and back of the PCB. It's quite a neat little thing. It is all powered from a 7.5V 2A wall wart. That doesn't seem a lot, but the LNB only needs 125 mA or so, and I've got a motor too that claims 350 mA start up (it's got a couple of biggish caps in there that give most of this) and 180 mA once it's going.
In the yellow box in the centre is a Montage M88DS3000 DVB-S2 demodulator. It does all the signal processing, usually in conjunction with a Montage M88TS2020 tuner chip which may well be under the shielding, but I see no reason to start pulling that apart. I can't find a data sheet, but as I can watch some channels I assume it's OK. There may be one bad pin leading to a voltage selector, but I wouldn't know where to start looking. Beside it is a CYT8117T12 1.2V regulator that supplies it.
In the pink box at the top are a LVC 374A octal edge-triggered D-type flip flop and a LVC 74A dual positive-edge triggered D-type flip flop. Thes are way beyond my knowledge. I'm not even sure what all those words mean. They look OK and I don't think you'd need that many pins to switch voltages.
The green box top right is a LM2585 boost converter. It takes the 7.5V supply and jumps it up to 22V. EC10, the electrolytic just above it is on the supply line and shows 8V steady. EC11 just below shows 22V steady, so it looks OK, along with its inductor and the big diode on the board underneath.
Bottom right in the dark green box is an LM317S adjustable regulator. From left to right as normally read (i.e. upside down from the photo) I get 18.7V, 20V and 22V. The 22V is clearly the supply from EC11, EC1 just above and left shows 20V steady so presumably the 18.7V is feedback from somewhere. This was suspect no 1, but I now think the fault is somewhere else. I assume the inductor to the left has something to do with this, but I'm not totally sure. The diode to the left of that, and the inductor left of that are connected to the TD1410C on the other side of the board. I'm not yet sure of the other large inductor and D5 above it.
Moving on to the back, the light blue box top left shows a CY7C680 USB controller and adjacent (presumably related) an EM24CO2A. USB communication is OK, so I reckon this is fine.
Bottom left in the purple box is where I reckon the problem is. The marking on this chip is pretty unreadable, but I think the top line says Techcode. I've had a chance to unscrew an earlier version of this Tevii S660 (as the one shown on LinuxTV.org ) and that's got a Techcode TD1410C in a similar spot but on the front of the board. It certainly seems the right sort of thing. A step down switch mode regulator with a fixed frequency of 380KHZ, which won't interfere with the other frequencies on the board. It's just the pinouts don't match, and I'm not sure what I should be looking for.
Pin 1 should be unused, and I get a square wave from +2V to -2V from the cathode of D2 and inductor L13, through capacitor C64. Pin 2 should be Vin, so I'd expect to see 20V, but instead I.m getting 8V from the wall wart. Pin 3 is the output and I'm getting an output sort of figure of 15.6V. A little high for 13V, but it's possible. Although TD1410C can only output up to Vin, so maybe this isn't a TD1410C, but a separate booster for 13V. Up to now I had assumed that this device would regulate the 19V supply down, rather than having two independent supplies. Because all that is left now to do the switching are a couple of puny SMT transistors (Q4 and Q5 in the orange box). But then Pin 4 is ground, which checks OK. Pin 5 is feedback, the threshold is 1.222V. I read 1.262 which is a couple of digits more than I usually trust my meter. Pins 6 and 8 aren't used, which matches up ot what I read. Pin 7 is the enable, a digital on/off switch. According to the datasheet the absolute maximum voltage here is 12V, and the threshold is 0.7V-1.7V. My reading of 3.6v should be fine.
And at this point I'm sort of lost, and I would be grateful for any pointers of where I might next look. Any ideas gratefully accepted.
I've got one of these little usb satellite tuners that's started to play up. I'm not quite sure where the problem is but I've narrowed it down to a couple of areas, and I'm including possibly more background than necessary on the off-chance it might help someone else in the future. And yes, I know that this is beyond economic repair. A new one is only £50 or so, but I just got curious.
First a bit of technical background (sorry if this is teachin your grandma to suck eggs stuff - I'm trying to keep it to the minimum). A TV aerial is just a bit of wire that sends the signals down to the receiver where all the hard work is done. A satellite dish (well actually the LNB sitting at the focal point) does a bit more work, and as a consequence there are a few different signals going up and down the cable. First of all the signal that the LNB picks up from the satellite is in the range of 10.7 to 12.75GHz . In order to pass this down a cruddy piece of co-axial cable, they do some clever processing with the output from a crystal to drop it to a range of 950 - 2150 MHz. So that's signal one. Of course the LNB needs power to do this so "signal" two is a 13V DC supply coming up the cable from the receiver. Then at some point they realised they needed more bandwidth, so stuck a second crystal in the LNB at a slightly different frequency. A 9.75 GHz crystal is used to process signals in the range 10.7 - 11.7 GHZ, and a 10.6GHZ to process 11.7 - 12.75 GHz. To switch between the two crystals they use a 22KHz tone at 0.5V above the 13V power supply. That's signal three. Finally they realised that they could cram the broadcast frequencies more tightly together if adjacent frequencies were polarised in different directions. That's Horizontal and Vertical in Europe (but circular for anyone in the US). To switch between those two we have "signal" four. The receiver supplies 13V when it wants to receive Vertical frequencies and 19V for horizontal ones.
So at any time the receiver is supplying the LNB with one of four possible supplies:
13V - Low frequency vertical polarisation
13V + 0.5V 22KHZ - High frequency vertical polarisation
19V - Low frequency horizontal polarisation
19V + 0.5V 22KHZ - High frequency horizontal polarisation
My little receiver only outputs 19V, no matter what it is set for, and only picks up horizontal channels. That shows about 0.3 V higher when on a high frequency channel. Clearly the 19V supply and the 22KHz tone work fine, but the 13V supply has a problem. I'm just not sure where.
Here are a couple of pictures of the front and back of the PCB. It's quite a neat little thing. It is all powered from a 7.5V 2A wall wart. That doesn't seem a lot, but the LNB only needs 125 mA or so, and I've got a motor too that claims 350 mA start up (it's got a couple of biggish caps in there that give most of this) and 180 mA once it's going.
In the yellow box in the centre is a Montage M88DS3000 DVB-S2 demodulator. It does all the signal processing, usually in conjunction with a Montage M88TS2020 tuner chip which may well be under the shielding, but I see no reason to start pulling that apart. I can't find a data sheet, but as I can watch some channels I assume it's OK. There may be one bad pin leading to a voltage selector, but I wouldn't know where to start looking. Beside it is a CYT8117T12 1.2V regulator that supplies it.
In the pink box at the top are a LVC 374A octal edge-triggered D-type flip flop and a LVC 74A dual positive-edge triggered D-type flip flop. Thes are way beyond my knowledge. I'm not even sure what all those words mean. They look OK and I don't think you'd need that many pins to switch voltages.
The green box top right is a LM2585 boost converter. It takes the 7.5V supply and jumps it up to 22V. EC10, the electrolytic just above it is on the supply line and shows 8V steady. EC11 just below shows 22V steady, so it looks OK, along with its inductor and the big diode on the board underneath.
Bottom right in the dark green box is an LM317S adjustable regulator. From left to right as normally read (i.e. upside down from the photo) I get 18.7V, 20V and 22V. The 22V is clearly the supply from EC11, EC1 just above and left shows 20V steady so presumably the 18.7V is feedback from somewhere. This was suspect no 1, but I now think the fault is somewhere else. I assume the inductor to the left has something to do with this, but I'm not totally sure. The diode to the left of that, and the inductor left of that are connected to the TD1410C on the other side of the board. I'm not yet sure of the other large inductor and D5 above it.
Moving on to the back, the light blue box top left shows a CY7C680 USB controller and adjacent (presumably related) an EM24CO2A. USB communication is OK, so I reckon this is fine.
Bottom left in the purple box is where I reckon the problem is. The marking on this chip is pretty unreadable, but I think the top line says Techcode. I've had a chance to unscrew an earlier version of this Tevii S660 (as the one shown on LinuxTV.org ) and that's got a Techcode TD1410C in a similar spot but on the front of the board. It certainly seems the right sort of thing. A step down switch mode regulator with a fixed frequency of 380KHZ, which won't interfere with the other frequencies on the board. It's just the pinouts don't match, and I'm not sure what I should be looking for.
Pin 1 should be unused, and I get a square wave from +2V to -2V from the cathode of D2 and inductor L13, through capacitor C64. Pin 2 should be Vin, so I'd expect to see 20V, but instead I.m getting 8V from the wall wart. Pin 3 is the output and I'm getting an output sort of figure of 15.6V. A little high for 13V, but it's possible. Although TD1410C can only output up to Vin, so maybe this isn't a TD1410C, but a separate booster for 13V. Up to now I had assumed that this device would regulate the 19V supply down, rather than having two independent supplies. Because all that is left now to do the switching are a couple of puny SMT transistors (Q4 and Q5 in the orange box). But then Pin 4 is ground, which checks OK. Pin 5 is feedback, the threshold is 1.222V. I read 1.262 which is a couple of digits more than I usually trust my meter. Pins 6 and 8 aren't used, which matches up ot what I read. Pin 7 is the enable, a digital on/off switch. According to the datasheet the absolute maximum voltage here is 12V, and the threshold is 0.7V-1.7V. My reading of 3.6v should be fine.
And at this point I'm sort of lost, and I would be grateful for any pointers of where I might next look. Any ideas gratefully accepted.
Attached Files