The following discussion about preamplifier design software took place in the Moon-Net reflector. I found the subject so interesting that I decided to put together all the messages in this page.
On 18-Jan-2004 WB9UWA wrote:
Designer SV is available for download from Ansoft. It is the free student version of Design. This software is very capable and can be used to optimize your favorite RF preamp on your favorite band. Circuits can be input to a scematic in much the same way as in PCExpress. It is designed to run on Windows 2000 or Windows XP. Here is the web site to download it from: (I hope you have a high speed connection! )
It has included example circuits and tutorials on optimizing designs.
I have only Windows 98SE, so I had to find a copy of Serenade SV. It is older software that does the same thing (basically). I have gone through some of the examples for "tuning" a component and "optimizing" a circuit. It can plot gain, Zin, Zout, K+b1 (stability), p1, IP3, and whole lot of other parameters that are over my head. Apcad is pretty, neat, but Serenade SV and Designer SV is the "real deal" for designing preamps. These programs are not limited to RF preamp design, but that is all I need it for.
My problem is to get started, I need the parameters for the Agilent ATF33143. I do have the .pdf spec. file for the ATF33143, but the problem is to match the parameters to the needs of the program. If someone has a scematic of an ATF33143 preamp saved in Serenade SV, that could pass it on to me, it would be a great help. No doubt a few of you in the group are already doing this.
When the design goal is to have a high power, low noise, STABLE, 2M preamp, the answer is currently illusive. The MGF1801 and other high power GaAsfets do not have parameters that allow modeling in this software. Current designs tend to be empirical designs that turn out to be conditionally stable. The ATF33143 DOES have parameters that allow modeling, but the device was not designed to operate as low as 2M. It generally becomes unstable. The usual fix is to swap the output with resistance, so IP3 suffers. I wish to investigate possible solutions without burning my fingers first. There are others that have designed, built and marketed stable, low noise, +20dbm p1 designs, so I know it can be done. The problem is that the designs are a closely guarded secret (not in the true ham spirit! ).
Again, to the users of Serenade SV; Do you have an ATF33143 circuit (project file) you can send me? A Serenade SV scematic with even only the ATF33143 on it (with parameters defined) would be sufficient.
On 18-Jan-2004 F5VHX wrote:
Can you share with us how to get our favourite device characterisations into Designer SV ?
On 18-Jan-2004 WB9UWA wrote:
First, I am brand new to Serenade, so I can speak only in general terms. Others that have designed using Serenade (or Designer) SV, may be able to provide a better answer.
When I am on the Desktop of Serenade, I can left click on an FET on the toolbar to add it to the scematic, as I would any other component. A right click will allow you to select exactly which type of FET you want. You can then move it around and place it on the scematic. When it is placed, a box comes up. If your FET is already placed, you can right click on it and select properties. The box allows direct input of parameters. The problem I was seeing was that the parameter names did not match. I might have selected the wrong FET.
I now have a scematic for Serenade SV from another Serenade user that uses an ATF33143. When I right click on that FET and observe the properties box, I see that a file is referred to. I don't know how he accomplished that. The tutoring and help files are extensive. I probably will not further investigate the problem of entering FET parameters soon, since I have the one I am currently working on. I suspect a J310 will be easy enough to enter, later.
What would be really tremendous, would be if someone could measure the parameters of power gaAsFETs like the MGF1801, etc., so they can be computer simulated. A good computer simulation can lead to the development of a stable design.
On 18-Jan-2004 SM5BSZ wrote:
> Where do you get the ATF33143? I am interested in building one
> for 6m. I hear they are pretty sweet ;-) TNX and 73, Lance
This is one possible source: http://antennspecialisten.se/en/ham/components.html
On 18-Jan-2004 EA3AXV wrote:
Other possible source: http://www.lnatechnology.com/lnatech_020.htm
On 19-Jan-2004 WA6PY wrote:
Hi, Few years ago I calculated noise parameters for MGF1801 from M. Pospieszalski MesFET model (used for years as a basic noise model especially for millimeters wavelength ) and added to the Mitsubishi factory given S-parameters in Touchstone format. S-parameters starts from 500 MHz, but any software should be able to extrapolate down to 100 MHz. Noise parameters are calculated fro 100 MHz to 2.1 GHz.
On 19-Jan-2004 F5VHX wrote:
Perhaps a full Ansoft Designer user is here ?? with real knowledge and understanding ?, and can let me know if there is a full S-para editor/input tool included in that, and which has been 'unplugged' in the SV Designer ??
>First, I am brand new to Serenade, so I can speak only in general
>Others that have designed using Serenade (or Designer) SV, may be able
>to provide a better answer.
yes but I want to get s para's into *Designer* (to get a full s2p file into SV8.5 is very simple)
As you were/are trumpeting the use of Designer for people to model their pre amps I thought you had solved the problem/found the input method, but apparently not then.
So PLEASE if anyone else does know how to do that (and let me make it clear again, s/noise para into DESIGNER _not_ SV8.5) I'm really keen to find out, thus far several of us using this software have not found the right solution, only 'a' solution(s) which is very hard work.
(Note: I'm not showing any E-Mail
address here in order to avoid them from being collected by SpamBots. You can
possibly find the E-Mail addresses of the above OM at QRZ.COM.)
This discussion is still not closed. If you have any opinions or additional related information you would like to be published here, just send me an E-Mail
Read other Moon-Net discussions