Yes the fist time I used the LP3879 It was powered by the 5v line. As it was a little hot, I changed and used the 3.3v to power it and now no more heat.
Jean-Pierre
Build your own USB tuner- will work with Minitioune
Forum rules
This forum is run by the BATC (British Amateur Television Club), it is service made freely available to all interested parties, please do not abuse this privilege.
Thank you
This forum is run by the BATC (British Amateur Television Club), it is service made freely available to all interested parties, please do not abuse this privilege.
Thank you
Re: Build your own USB tuner- will work with Tutioune
Thanks for the tips, here are my findings...
The 1.8v regulator and diode use was a solution because i did not have any 1v regulators in stock
The tuner works with my Win7 machine so the hardware should be good ...
I've disabled 'USB Selective Suspend' in the laptop power settings but this had no effect, i could not find any other usb power settings to adjust.
I still get the 'pb restart USB deviceB' popup window, the gui is updating while this message is visible with constellation etc.
All psu lines are free of noise ~10mV ripple so i'm happy there, even the 1v rail is good so far.
I've added more decoupling, now getting a bit excessive i think but still the same problem with the laptop only...hi
The I2C lines are over shooting at both high and low levels by approx 1v !! so this may be part of the problem but the overshoot is exactly the same when i use the desktop pc and minitioune is working well there...
I will do some more investigation later.
Rob
edit: Swapped to the third separate usb port on the HP 2.2G dual core laptop and it now works, all be it quite slowly
Still floating point error on the old Compaq USB1.1 laptop so that maybe impossible.
The 1.8v regulator and diode use was a solution because i did not have any 1v regulators in stock
The tuner works with my Win7 machine so the hardware should be good ...
I've disabled 'USB Selective Suspend' in the laptop power settings but this had no effect, i could not find any other usb power settings to adjust.
I still get the 'pb restart USB deviceB' popup window, the gui is updating while this message is visible with constellation etc.
All psu lines are free of noise ~10mV ripple so i'm happy there, even the 1v rail is good so far.
I've added more decoupling, now getting a bit excessive i think but still the same problem with the laptop only...hi
The I2C lines are over shooting at both high and low levels by approx 1v !! so this may be part of the problem but the overshoot is exactly the same when i use the desktop pc and minitioune is working well there...
I will do some more investigation later.
Rob
edit: Swapped to the third separate usb port on the HP 2.2G dual core laptop and it now works, all be it quite slowly
Still floating point error on the old Compaq USB1.1 laptop so that maybe impossible.
Re: Build your own USB tuner- will work with Tutioune
I am building mine on single sided copper clad board, drill holes where components go and back wire.
Have all the bits now apart from pre-programmed USB FT2232 module that I have ordered, hopefully running by end of next week.
Andy G3UEQ
Have all the bits now apart from pre-programmed USB FT2232 module that I have ordered, hopefully running by end of next week.
Andy G3UEQ
Re: Build your own USB tuner- will work with Tutioune
I'm designing a pcb for the tuner at the moment and considering options. I've noted the comments about decoupling on supplies and I2C.
How far should we go with the capabilities? Would it be useful for it to work on broadcast satellites at higher symbol rates? That would increase the current consumption. Talking with Jean Pierre, the 1v supply would need to give 1.8A for the highest symbol rates used. This would mean changing the 1v regulator from an LP3879 to an MP2305, (switched-mode, Ebay from China, not from Farnell / RS) or an NCP5662 (linear but chunky, local sources).
Broadcast use would also require an 18v supply to switch the LNB to horizontal polarisation and a 22kHz generator to switch the LNB to high band. The wideband transponder on the Es'hail2 satellite will be horizontally polarised, requiring 18v. Jean Pierre suggests using the LNBP21D chip to do this. Alternatively you could use a conventional sat rx to do the LNB configuration and use a tee piece to tap off the RF.
On the psu side, I was thinking of using LM2596 modules for 5v and 3.3v to give some leeway to take 1.8A off the 3.3v supply for the 1v regulator if needs be. Considering the low price of the LM2596 modules, it may be easier to use a third LM2596 set to its lowest voltage to supply the 1v regulator, to reduce the overall current consumption.
Brian
How far should we go with the capabilities? Would it be useful for it to work on broadcast satellites at higher symbol rates? That would increase the current consumption. Talking with Jean Pierre, the 1v supply would need to give 1.8A for the highest symbol rates used. This would mean changing the 1v regulator from an LP3879 to an MP2305, (switched-mode, Ebay from China, not from Farnell / RS) or an NCP5662 (linear but chunky, local sources).
Broadcast use would also require an 18v supply to switch the LNB to horizontal polarisation and a 22kHz generator to switch the LNB to high band. The wideband transponder on the Es'hail2 satellite will be horizontally polarised, requiring 18v. Jean Pierre suggests using the LNBP21D chip to do this. Alternatively you could use a conventional sat rx to do the LNB configuration and use a tee piece to tap off the RF.
On the psu side, I was thinking of using LM2596 modules for 5v and 3.3v to give some leeway to take 1.8A off the 3.3v supply for the 1v regulator if needs be. Considering the low price of the LM2596 modules, it may be easier to use a third LM2596 set to its lowest voltage to supply the 1v regulator, to reduce the overall current consumption.
Brian
Re: Build your own USB tuner- will work with Tutioune
Hi Brian,
I am glad you are designing a PCB as hopefully that means I won't have to. Regarding your comments
I worry about the noise from switching regulators in this application. It would need very careful attention to filtering. Could you use 2 linear 1V regulators rather than a switched mode? Instead of running off the 3.3V regulator, a power resistor is another way to reduce the regulator dissipation. Not very elegant or high tech but reliable and cheap.
I can see the use for a 22kHz tone and LNB voltage switching as we may have a 10 GHz satellite transponder and normal LNBs might even work with it. I suppose it could also supply a preamp. However, seems like a good candidate for a plug in board.
Mike
I am glad you are designing a PCB as hopefully that means I won't have to. Regarding your comments
I worry about the noise from switching regulators in this application. It would need very careful attention to filtering. Could you use 2 linear 1V regulators rather than a switched mode? Instead of running off the 3.3V regulator, a power resistor is another way to reduce the regulator dissipation. Not very elegant or high tech but reliable and cheap.
I can see the use for a 22kHz tone and LNB voltage switching as we may have a 10 GHz satellite transponder and normal LNBs might even work with it. I suppose it could also supply a preamp. However, seems like a good candidate for a plug in board.
Mike
Re: Build your own USB tuner- will work with Tutioune
I agree the the tone gen and LNB power could be a plug-in module. I also would steer clear of switchers near the tuner. We achieved such good spectral purity on the DTX1 by using point of load LDO linear regulators where it counts. We found out the hard way after a number of PCB re-spins. £££ Ouch. And as regards how big a linear regulator, I would refer the the late U.S. Army Gen. H. Norman Schwarzkopf who, when criticised about the amount of materiel in theatre, replied "There is no such thing as too much firepower. Next question?"
Re: Build your own USB tuner- will work with Tutioune
Concerning your PC USB connector an the USB cable used:
If you use a USB 1.1 port, max speed is "full speed" = 12Mbps
if you use A USB 2.0 port max speed is "High speed" = 480 Mbps
As we will transfer the TS via the channel B of FTDI to your PC, the "full speed" mode will not be enough for higher SR.
So channel B is set to work at "High Speed" mode so we absolutely need to use a 2.0 USB port.
If you are using a 1.1USB port, the software will be unable to start/restart USBdeviceB in High speed mode, so you will see the message 'pb restart USB deviceB' , Channel A of the FT2232H could work, this is the channel for I2C traffic, so you could have data about RF level and constellations, but you will never get the TS via channelB ( FIFOtransfert at High speed )
We can also encounter some issue with some bad quality USB cable that are not good for "High Speed" mode . 480 mbps in a serial cable gives something like UHF in the cable.
(Last year I had some trouble with my rotors controller for HamTV, all strange errors disappeared when I change the USB cable by another)
Jean Pierre F6DZP
If you use a USB 1.1 port, max speed is "full speed" = 12Mbps
if you use A USB 2.0 port max speed is "High speed" = 480 Mbps
As we will transfer the TS via the channel B of FTDI to your PC, the "full speed" mode will not be enough for higher SR.
So channel B is set to work at "High Speed" mode so we absolutely need to use a 2.0 USB port.
If you are using a 1.1USB port, the software will be unable to start/restart USBdeviceB in High speed mode, so you will see the message 'pb restart USB deviceB' , Channel A of the FT2232H could work, this is the channel for I2C traffic, so you could have data about RF level and constellations, but you will never get the TS via channelB ( FIFOtransfert at High speed )
We can also encounter some issue with some bad quality USB cable that are not good for "High Speed" mode . 480 mbps in a serial cable gives something like UHF in the cable.
(Last year I had some trouble with my rotors controller for HamTV, all strange errors disappeared when I change the USB cable by another)
Jean Pierre F6DZP
Re: Build your own USB tuner- will work with Tutioune
The LM2596 modules do have a lot of ripple on the output 'as is' and will need a lot of extra filtering.
There was ~150mV at 5v on mine before i added some series resistor and more decoupling.
Rob
There was ~150mV at 5v on mine before i added some series resistor and more decoupling.
Rob
Re: Build your own USB tuner- will work with Tutioune
My modules have arrived so just need a PCB and some regulators to complete.
Mike
Mike
Re: Build your own USB tuner- will work with Tutioune
Thanks for the comments on the pcb. I forgot to mention having a connector for the parallel transport stream output.
I would rather not use switching supplies, but the amount of heat to get rid of would be huge otherwise. If the 1v supply can take up to 1.8A, at high symbol rates, from a 13.8v supply, that's 23W to lose for the 1v supply alone. I was thinking of a hybrid approach with LM2596 boards followed by linear regulators. It seems rather belt and braces, but the LM2596 boards are cheap.
I'll make the 18v / Disecq option a plug in.
An Octagon LNB or similar followed by a SUP 2400 would bring the wideband transponder on 10495MHz down to 1655MHz.
Brian
I would rather not use switching supplies, but the amount of heat to get rid of would be huge otherwise. If the 1v supply can take up to 1.8A, at high symbol rates, from a 13.8v supply, that's 23W to lose for the 1v supply alone. I was thinking of a hybrid approach with LM2596 boards followed by linear regulators. It seems rather belt and braces, but the LM2596 boards are cheap.
I'll make the 18v / Disecq option a plug in.
An Octagon LNB or similar followed by a SUP 2400 would bring the wideband transponder on 10495MHz down to 1655MHz.
Brian