Re: Front End (TUNER)

Thank you so much for the information. Now I managed to find an application note for a DVB-T Tuner from infineon. I intended to expand my scope of knowledge. (not in depth in RF) What I am interested there is how to configure the settings for respective blocks in the tuner. I am quite eager to know how the I2C serial bit stream is recognized by different control blocks in the front module. And I wish to know the scheme that the processing unit sends commands to the front end module, and how it understands the commands from the programming point of view. With RF receiving system, I understand the basic block diagram but wish to know more about it. I am not going to design a front end module but I am interested in the power supply (usually 5 to 33V boost converter) of it. I would like to have the relationship between the power supply specifications and the tuner performance. However, now I have one applications notes which encourages me to find more information on it. Thans for your initiation. Regards

Easy to buy, yes. But don't always expect a whole lot of dynamic range. With some tuners that I wanted to use I got the impression that their designers didn't know what a dynamic range is.

"Joerg"

The TV and FM tuner modules I have used seemed well optimised for their jobs.

Maybe you expected something more.

...... Phil

I can see where Joerg is coming from ... I recently completed a gig where I needed to work with TV tuners and TV manufacturers. TV tuner dynamic range is being compromised lately by some tuner manufacturers. The best manufacturers of "can" analog tuners I found were LG and Samsumg.

TV tuner on-a-chip companies like Xceive have even poorer IP3 ratings. Here is an overview of a linearity test:

The prevailing architecture in most modern TV's seems to favor a "can" down converter, e.g. Sancy ($2.25), followed by a digital bandwidth SAW, e.g. Epcos. The rest is done in the demod chip. This architecture is compatible with both analog and digital broadcasts.

There are still large manufacturers designing new analog-only TV's, some of which incorporate the can tuner + 2nd SIF output; however, pricing pressure is favoring Xceive to win the tuner market if they can improve dynamic range performance.

I still have a 197? Sony Trinitron with a clunk-clunk VHF tuner and a vernier dial for the UHF. Nice picture, good tuner - no varactors.

Frank

My best estimation: the i2c stream specifies well published channel frequency, the uC running the tuner tunes the Local Oscillator (LO), the IF appears on the output, some automatic frequency control (AFC) trims it to exact tuning, done. This is not the way "it used to be". I remember my dad manually re-tuning the "slugs" on an TV back in the day of mechanically switched tuners. There were may iterations of "typical" implementations in the intervening decades.

It's a little more complicated. The OP's tuner has a CVBS and 2nd SIF output, which means there is an integrated PIF/SIF processor as well as the PLL for the L.O. Parameters that get set for the PIF/SIF include: Wideband AGC takeover point NEG/POS video modulation

Frank Raffaeli

It's been too long ago since I gave up on TV tuners (mid 90's) because of dirt poor 3rd order IP. After hearing from other engineers that is has gotten even worse I never looked back.

If you find a really good one again please let us all know. One of them I investigated further and I found a wimpy dual-gate running at 5mA or so in front. That won't hold any water with strong signals.

Yeah, probably. But what really miffed me was this: Placed the antenna on a mast after re-roofing the house. So, I equipped it with a nice preamp, splitter, postamps with eight outputs and a home-run style wiring closet. It cost a bit but I wanted to do it right. I was very diligent with the gain distribution plan as well.

Hooked up a professional grade receiver to see how FM and stuff comes in. Sweet, could now get Bay Area stations etc. Listened to the audio on a few TV channels and found that I had now lots more than ever. Then I turned on the TV and it was pandemonium. Utter disappointment. All channels pumping with noise like crazy, not even the local stations showed clearly, audio either not there or completely garbled. Tried the downstairs TV, same thing. Then through the VCR's tuner, same thing. I had to attenuate by more than 10dB plus place a notch on a strong local channel 29 station, plus another channel I don't remember, plus notch out the complete FM band. It was pathetic. The result is that we don't have a single additional station from when the antenna was on the roof (with a shot preamp). Oh well, most TV programming ain't worth it anyhow.

( snip drivel)

( snip drivel)

Pathetic really.

....... Phil

I presume you meant to say that the performance of most preamps would limit linearity, and would dominate IP3, etc. over the tuner. That's true, although I've seen some that don't ... I've designed some that don't; maybe Joerg used a commercial model.

Frank

"Frank Raffaeli" "Phil Allison"

Do not alter context.

...... Phil

Well, yeah. If a lab grade receiver works just fine and the TV on the same antenna is turning into a hopeless pile of intermodulation it's pretty clear that the TV isn't up to snuff. Which didn't surprise me. I just didn't expect it to be this bad.

"Joerg" "Phil Allison"

You did a bunch of stupid things that fouled up the whole band for a TV receiver.

Find out the difference between AM and FM - sometime.

Find out the difference between a TUNED pre-amp and a whole TV band one.

Idiot.

....... Phil

Agreed. Did you do any tests on the TV tuner? IP2, 2-tone IM, etc.? Did it use a solid state tuner or a "can"?

Frank

TV tuners used to be built to accommodate the wide range of signal levels from over the air TV reception. Now, they "Assume" that every TV will be connected to Cable, DVD, VCR or SAT, all of which have a higher than 0 DBM signal level, and never over +10 DBM, so they don't give a damn about the dynamic range. Also, some TVs had an extra IF stage that would help in far fringe areas. That was why a cheap portable would give a better picture than a console. Portables were expected to have a wider range of signal levels to deal with, and the design reflected this.

Ignore the troll hiding below the Equator.

I did back in the 90's. Shook my heard and discarded the tuners afterwards, then cracked out a Mini-Circuits DBM to build the receiver. In the 80's they were better but there I used tuners scrapped out of older TVs. Probably the older engineers still knew how to do it. One had Ge-Transistors (AF239?) and was, well, somewhat ok. The other one had two tubes and was quite excellent.

And sure enough the TV where that tube tuner came out of was the best (until it's flyback XFMR went up in a plume of smoke one fine day). When a powerful pager system was installed near our home almost all "modern" transistorized TVs fell off the rocker. This signal was even out-of-band, BTW. Telecom came out with lots of expensive Rhode&Schwarz gear, determined that poor tuner design was to blame so they were not legally obliged to do anything about it yada, yada, yada.... Lots of negative PR mounted and then they agreed to supply notch filters for free. I mounted a lot of these for neighbors because installation was not included.

Besides the transistor versus tube thing there was another major difference: The tube tuner had a ganged variable capacitor while the transistor version used varicap diodes. I did not investigate the tuners in our current sets. Not really worth it. Not just because of lousy RF performance but except for an occasional old movie and the evening news there ain't much to watch anyhow.

Even the old tuners were "in a can", except that those cans were the size of a small paperback book. Very nicely done. For example, they had thin copper foil on a rubber cushion to make sure that the lid was RF-tight. They don't make'em like that anymore :-(

"Myauk" wrote in news:1174627870.460950.97910 @e65g2000hsc.googlegroups.com:

Reverse encgineer the tuner. Of find a datasheet on it.

It isn't. I2C only controls the PLL IC, which may have GPIO to control band selects.

In a nutshell, it is a couple voltage tuned filters, feeding a mixer supplied by a voltage tuned oscillator. The PLL IC, via I2C, tunes the voltage tuned oscillator, and by proxy, the tuned filters. The output of the mixer goes into an IF strip, and a video detector. There may be a number of bands using different coils in the tuned filters, selected by PIN diodes, which are controlled by GPIOs on the PLL chip, or discrete inputs on the module.

As said, the 33V is just to tune the varicaps. It requires a higher voltage to tune higher frequencies. It only needs to be a few ma for that. The 5V can be a standard 5V supply, although if your project has sigificant digital, it should be a separate analog 5V.