The Designer's Guide Community
Forum
Welcome, Guest. Please Login or Register. Please follow the Forum guidelines.
Aug 17th, 2024, 9:18pm
Pages: 1
Send Topic Print
Laser transmitter design (Read 4029 times)
neopl
New Member
*
Offline



Posts: 3

Laser transmitter design
Apr 11th, 2007, 8:25am
 
Hello everyone, I am totally new to RF design and would like to seek some help in designing a simple laser transmitter. When I am trying to evaluate the performance of my laser transmitter, I noticed that there is significant jittering in the eye diagram when I input a PRBS signal to it. (Please refer to the attached photo.)

Is there any way I can reduce the jitter?

I have kept my design rather simple as I am new to RF design.
DC + RF signal (using a bias tee) - SMA connector - SMA connector on PCB - PCB track - damping resistor (for impedance matching) - PCB track - laser

My RF signals is generated using a high frequency function generator.

I have a few questions regarding RF design:
1. When do we use transmission line theory (TLT) to design the width of the track? If the PCB track is about 30mm, do we need to apply TLT?
2. Because the PCB track has to be broken up into two, due to the damping resistor, is there any PCB layout rule I must apply?

I have tried everything I can to remove the jittering, but it does not seem to help. I hope that one of the experts out there is able to spot some obvious omissions on my part to help me with the design. If you need for information, please let me know! Thanks.

Poh Ling
Back to top
 

CIMG7177.JPG
View Profile   IP Logged
mg777
Senior Member
****
Offline



Posts: 131

Re: Laser transmitter design
Reply #1 - Apr 11th, 2007, 6:55pm
 

For starters it looks like you're using the hp 54120 series of rep scopes (also explains why you needed a camera for the screenshot, because the printer is, ahem, GP-IB only). Make sure you're not on line trigger or something. Check the trigger source - if you're using the old hp 3.0 Gbps BERT you should start with TRIG=CLK/32 and feed that to the trig channel of your scope. If I rememeber right the 54120 trig can't handle wirespeed clocks.

If the scope is operating OK, try increasing the bias current on your laser. If you have a reference transmitter board (at any wavelength) try that with your set-up. Looks like your O/E converter is OK.

M.G.Rajan
www.eecalc.com



Back to top
 
 
View Profile   IP Logged
neopl
New Member
*
Offline



Posts: 3

Re: Laser transmitter design
Reply #2 - Jun 5th, 2007, 10:59am
 
Hi,
I am still working on the laser transmitter design, using MAX3735 laser driver. I am using a Anritsu MP1650A as the differential data inputs to my laser driver, through 2 SMA connectors.

I realised that my function generator has ECL, 50 ohm data outputs and my MAX3735 laser driver needs either PECL or CML data inputs. I tried searching around to see if there are some circuits for me to convert ECL data outputs to PECL/CML data outputs. However, there aren't any hits.

I have to admit that I am not familiar with these logic families. I am wondering if the lack of hits is an indication that it is not necessary for me to do the conversion? Are ECL compatible with PECL/CML logic families?

Thanks.
Poh Ling
Back to top
 
 
View Profile   IP Logged
mg777
Senior Member
****
Offline



Posts: 131

Re: Laser transmitter design
Reply #3 - Jun 5th, 2007, 11:36am
 
I suppose the Anritsu PG o/p can support 50 Ohm to ground? If your laser driver can take CML inputs, these are very likely pulled up to +Vcc - check the Maxim datasheet. If so, simply a.c couple the Anritsu PG's output via good blocking caps & watch out for the differential skew.

What are good blocking caps for data transmission? The best are from Picosecond Pulse Labs. However you've got some ground to cover before requiring such quality, so I recommend starting with a X7R 10nF 0805 cap that you can solder between two SMA launchers - tab to cap to tab, GND to GND.

If your PG can output only ECL then you can always use a translator IC to get PECL out of the NECL:
http://www.onsemi.com/pub/Collateral/MC10EP90-D.PDF
is a 3.0 GHz 3.3V / 5V Triple ECL Input to LVPECL/PECL Output Translator.

You seem to have good equipment, so enough of the sermon - let's see a good eye now!

M.G.Rajan
www.eecalc.com

Back to top
 
 
View Profile   IP Logged
neopl
New Member
*
Offline



Posts: 3

Re: Laser transmitter design
Reply #4 - Jun 5th, 2007, 12:20pm
 
Thanks M.G.Rajan. You have been replying my postings. I hope you wont mind me bothering you around for a while, so that I can get this laser transmitter design sorted out. I have been working on it for a while.

Regarding your comments about CML inputs being pulled to Vcc, I have checked the MAXIM datasheet, pg 13. In the simplified input structure, it is noted that IN+ and IN- are connected to Vcc. http://datasheets.maxim-ic.com/en/ds/MAX3735-MAX3735A.pdf  

I am placing my sample order for the 3.0 GHz 3.3V / 5V Triple ECL Input to LVPECL/PECL Output Translator now.

I hope I will see a good eye soon. Thanks for replying and hear from me again.

Poh Ling

Back to top
 
 
View Profile   IP Logged
Pages: 1
Send Topic Print
Copyright 2002-2024 Designer’s Guide Consulting, Inc. Designer’s Guide® is a registered trademark of Designer’s Guide Consulting, Inc. All rights reserved. Send comments or questions to editor@designers-guide.org. Consider submitting a paper or model.