Help me debug my synthesizer design

[liberal]

detect my pll

I design a synthesizer with ADF4212 .when I set the MUXOUT RF R COUNTER OUTPUT,it is right.But when I set it RF N COUNTER OUTPUT ,it is wrong,about 75MHz or 50MHz with different N.and when I set it RF ANALOG LOCK DETECT ,MUXOUT outputs RF/N ???and when I open the loop,and only add the RFin signal ,the Vcontrol is 5v,why not 0V?I have debuged it for two weeks,but no progress.HELP me !SOS!

 

[RFDave]

how do i find my pll setting

How are you programming this? Are you using the software that Analog provided to control this? If you are using software that you wrote, can you use the software that Analog Devices provides to isolate whether it's software or hardware?
Do you have the charge pump polarity correct?
How did you design the loop filter?
Do you have enought power at the Prescaler input port?

Dave

 

[liberal]

Re: Help!MY PLL!

I program it with AT89C2051.
I do it as following:
open the loop,remove the loop filter .
The R divider is right ,and the polarity is right too(in my application it is positive,but even though i set it negative ,the voltage is near 5V).and if i set RF POWER-DOWN ENABLE ,then the chip doesn't work,so it prove the RF N COUNTER is right,because RF POWER-DOWN is a bit of RF N COUNTER LATCH(and whatever the N divider is ,the voltage doesn't change).and the RF input
is enough.and the DC voltage at RFin is 2.04V.
I am mad mad mad

Added after 2 hours 20 minutes:

My Vdd1=vdd2=Vp1=Vp2=5V

 

[biff44]

Re: Help!MY PLL!

Check obvious things, like are all dc voltages and grounds proper, is board layed out correctly, etc. If you put a DC voltage into the VCO tune port (always use a series 1 K resistor to avoid accidentaly blowing out the tuning varactor), does the VCO tune across your desired band, or does it stop oscillating at some input voltages.

Odds are good you are programming the PLL chip wrong. That chip does have a Mux Out pin. You can put in a precise synthesized RF signal for Ref, IF, or RF input, and look at the divided down output on the Mux Out with a precise counter. I like to run the test equipment synthesizer and counter off of the same 10 MHz source. You look at the Mux Out and figure out if the frequency on the counter is EXACTLY what you expect it to be. It is easy to be off by 1 bit, or to enter a not-allowed divisor ratio.

If all the dividers are being programmed properly, then I would suspect that my control loop filter is unstable. I would play around with some control loop simulation software to get more gain or phase margin, and see if it locks up. If you are not familiar with control loop filter, try using just a passive R-C loop filter which while not optimum should be more stable.

If it locks with a passive loop filter but not the active loop filter, either your control loop design is flakey, or there is something unstable in the op amp itself. Read the op amp data sheet carefully. Make sure the op amp layout is good (especially if it is a wide bandwidth op amp).

 

[liberal]

Re: Help!MY PLL!

The VCO works well,and in the band.
The problem is that no RF N divided signal .The program is right .
when i close the loop,i set the polarity negative ,the Pll outputs the lowest frequency.and when i set the polarity positive,the frequency is the highest.This show that RFin doesn't work..........
Another question,if i don't open the circuit,and i want to measure some point's power with spectrum analyzer,i don't have microwave probe but only a coxial line,how should i do ,if i want to konw the power with it?

 

[biff44]

Re: Help!MY PLL!

RFDave and I have given you around 10 things to go check. I suggest you go check them and don't putz around with meaningless measurements.

 

[RFDave]

Re: Help!MY PLL!

At this point in troubleshooting things, I like to knock off work early, go out for a beer or four, and get a good night's sleep. Once you have a clear head, stop hacking at things, and do the troubleshooting that has been reccomended. Once you finish all those things, you should be able to figure it out. If you can't, come on back with the results and ask for more help.

As far as measuring power, without a high impedance probe, the easiest thing to do is to build your owm passive high impedance probe. Here is a web page with detail's of building a DC-1 GHz probe

h**p://emcesd.com/1ghzprob.htm

Dave

 

[biff44]

Re: Help!MY PLL!

A couple of Mai Tai's and some chow at the China Buffet usually get me thru!

 

[liberal]

Re: Help!MY PLL!

haha,I have a rest on the weekend ,drink and dance.But no chow mein.But when you come to China,we can drink and have chow mein together.

 

[biff44]

Re: Help!MY PLL!

"chow" is a generic term for food, and does not mean just chow mein. China Buffet is a restaraunt in Lowell MA USA that draws the local engineers like a magnet for lunch/dinner. It serves all different types of Chinese food in an all-you-can eat buffet style for a fixed price. After working all day long, and possibly skipping lunch, you can "chow down" (eat way too much!).

The Mai Tai is an alchoholic drink from Hawaii, that for some reason is served often in US Chinese restaurants. I once tried to order a Mai Tai in London at a Chinese restaurant and they looked at me like I had 2 heads--never hear of one.

 

[liberal]

Re: Help!MY PLL!

HI
I add a c=1.5nF between the MUXOUT and GND,Then the Vcontrol begains to change,I don't know why?The detect output can influence the PFD output??
I attached the Pic from my SA.
[/img][/list]

 

[biff44]

Re: Help!MY PLL!

It looks phase locked to me. Notice that the center frequency shifts from step 1 to exactly 1.4 GHz in step 2.

What do you see if you put an AC coupled oscilloscope probe onto the VCO tune line in case Step 2? Is there a sinewave tone there, maybe 100 mV or so? That may give you a clue where it is coming from. If there is a tone, what frequency is it, and what control loop bandwidth did you design the loop filter for?

Also check the DC voltage on the Vtune line and make sure it is ok. For example, if you have a 5 volt power supply, and the Vtune line is at 4.2 V, the op amp is probably cutting out as the output is too close to the supply rail.

 

[liberal]

Re: Help!MY PLL!

Hi
Thank you very much every guys!I have done it!!
it is a resistor seriesed between the loop filter and the VCO's varator,in general,it maybe 10K,But I am so negligent that i soldered a 10M,you konw how terrible it is......Take me many days to check the circuit...........I hate it.But i am happy now ,so excited ,every guys here definitly had this spirit before and now...
I HAVE DONE IT!