Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: Prev 1 2 [3] 4 Next   Go Down
  Print  
Author Topic: How to find someone to design an 80m transceiver?  (Read 18960 times)
VE3UAL
Member

Posts: 18




Ignore
« Reply #30 on: August 30, 2013, 09:34:06 AM »

Superb! Dave,  Smiley
And anyone else willing to join in and participate in this effort - PLEASE DO!

As far as our giving up on what we worked on for over two years -
and spent all that money, time and effort on -
yes that is a tough bullet to bite -
but we really did give it a thorough try.

I will see what we can do about putting out our updated schematics
but I think it is a dead horse.
You can see the board at: www.webpal.org/aNewsletter/radio/board version two.jpg

Peace and love,
Bruce
www.webpal.org
www.webpal.org/SAFEems/
Logged
KE3WD
Member

Posts: 5689




Ignore
« Reply #31 on: August 30, 2013, 09:57:29 AM »

You should put that nonworking circuit board and any other pertinent info such as schematics, datasheets, etc. into the hands of someone skilled in both the design and troubleshooting aspects and find out what they can come up with. 
Logged
AF6WL
Member

Posts: 142




Ignore
« Reply #32 on: August 30, 2013, 10:08:48 AM »



I will see what we can do about putting out our updated schematics
but I think it is a dead horse.
You can see the board at: www.webpal.org/aNewsletter/radio/board version two.jpg


From this layout, three problems are immediately apparent:
1.. You have no real ground. Flooding the PCB with ground on both sides of the board and pinning the planes with lots of vias is needed.
2.. Folding the driver, PA and filter stages with no screening between them is a recipe for oscillation.
3.. That stub trace running along from the antenna connector all the way to the LHS is a very bad

« Last Edit: August 30, 2013, 10:12:12 AM by AF6WL » Logged
WB6BYU
Member

Posts: 13243




Ignore
« Reply #33 on: August 30, 2013, 01:02:17 PM »

Quote from: AF6WL

1.. You have no real ground. Flooding the PCB with ground on both sides of the board and pinning the planes with lots of vias is needed.



A very important point.  I'd use the whole top layer as a continuous ground, except where you
need to clear it away from the holes for the leads through to the other side.  Then the traces
are on the other side.  You might need to use the ground side occasionally when traces would
otherwise cross, but try to maintain a continuous ground surface as much as possible.

TX stages should generally be laid out in line, to keep the output as physically separate from
the input as possible.  Common DC feeds and ground returns between different stages
should be avoided where possible - we had a lot of problems with a 2.4 GHz amplifier due
to a common ground path shared between input and output (which caused positive feedback
in a 3-stage amplifier.)  If you can't move the input further from the output, metal shields
can be added (cut from a tin can if needed for initial testing.)

You may find that mounting the PA on a separate board provides better separation, as well
as allowing the board to be positioned where the transistors can be attached to an adequate
heat sink.
Logged
VE3UAL
Member

Posts: 18




Ignore
« Reply #34 on: August 30, 2013, 10:04:40 PM »

Thanks for the critique on the Printed Circuit board.
All very valid -
and yes we had caught each of those points.

Everyone picked up on the grounding problem -
and we had determined to put the PA on a separate board.

I think it is the instability and variability of the NTSC color crystal
itself
which is accentuated by variations from manufacturing batches 
but there appears to be something else
inherent in the design itself
that has caused Benson with the Warbler -
others with Softrock design -
and the two RF engineers who worked on ours for two years -
to abandon the design.

With all those corrections planned for the next board layout -
it was still determined/decided that the signal from the exciter
was never going to be stable enough
and that the amplifier was a poor design also
that others had difficulty with.

The world changes - and nothing is impossible -
to say that we couldn't come back to it -
but the consensus at this moment is that we should move on -
and I am very much interested in examining the approach
that Dave has proposed:

 
A rather simple way to obtain a crystal controlled channelized 80 meter SSB transceiver is to mate a CB SSB transceiver ($130 in single quantities) to a transverter.

Example

Receive
LP filter
A Mini-circuits ADE-1L mixer and a 23.410 MHz LO convert 3.555 MHz to 27.405 MHz (CB channel 1), CB channel 1 to 40 give frequencies of 3.555 to 3.995 MHz

Transmit
ADE-1L mixer and 23.410 MHz LO
ERA-5 amp
PA driver
PA using two inexpensive MOSFETs
LP filter


Once again - anyone interested in helping us to pursue this -
you can help by emailing me directly at:
bruce at webpal.org

Peace and love,
Bruce
Logged
NI4L
Member

Posts: 12


WWW

Ignore
« Reply #35 on: September 01, 2013, 10:35:57 AM »

I came across this QRP rig, it is new and cheap. http://www.wouxun.us/item.php?item_id=302&category_id=65

It is worth looking at, and close to your budget.


73

Chris NI4L
http://www.ni4l.com/
Logged

73
Chris Fox NI4L
Ni4L ANTENNAS & ELECTRONICS
http://www.ni4l.com/
AD6KA
Member

Posts: 2237




Ignore
« Reply #36 on: September 01, 2013, 03:01:26 PM »

Any reason not to copy this and slap a 20w amp on the back end?
http://www.njqrp.org/warbler/

X 2
I have two Warblers and they are excellent little rigs.

Crystal control is the way to go for a dedicated simple PSK-31 rig.
I have built the BITX20, and while it is a nice rig, IMHO it is really
too drifty for digital work, even after a warm up period. And a crystal
rig would be far cheaper and use far fewer parts.

A 15-20w amp using inexpensive IRF510 MOSFETS would
be cheap and easy to build. In fact, you could easily copy,
(but improve!) the Ramsey "Q-Amp" monoband 20w amp
that uses IRF510's.  I have a PSK-20 from SWL built into
an enclosure with a Ramsey Q-Amp. I only get 15w out of
it, but many improvements to that amp could be made.
GL ES 73, Ken AD6KA
Logged
VE3UAL
Member

Posts: 18




Ignore
« Reply #37 on: September 01, 2013, 04:22:13 PM »

Thanks guys:
Yep the Wouxun looks REALLY nice -
but it is too pricey for our members.

Yes, we spent two years
and thousands of dollars
trying to tie the Warbler to the Ramsey.

The Warbler wasn't stable enough.
Benson quit selling them and tried to modify it also -
but apparently gave up.

There are seem to be troubles with the Ramsey -
so we have had to bite the bullet and go off in a new direction
suggested by Dave Cuthbert.

I have attached a diagram -
and I may have gone too far in considering the
BaoFeng UV-5R




I am told that we must have SSB
and I don't know that it is that.

Hopefully someone will give me some input on this -
and if we can figure it out
I may start a new thread.

Peace and love,
Bruce
Logged
WX7G
Member

Posts: 6041




Ignore
« Reply #38 on: September 02, 2013, 07:13:58 PM »

Bruce, the tranverter block diagram and the main parts of the schematic are ready. I can email the block diagram as soon as I have an email address.

  Dave
Logged
VE3UAL
Member

Posts: 18




Ignore
« Reply #39 on: September 02, 2013, 07:20:20 PM »

Superb Dave!

language at www.webpal.org
Logged
AF6WL
Member

Posts: 142




Ignore
« Reply #40 on: September 02, 2013, 07:33:56 PM »

Any reason you have to stick with PSK31 for the net ?

DominoEX seems to less succeptible to frequency drift and you can run the amplifier class C.

http://www.w7ay.net/site/Applications/cocoaModem/UsersManual/mfskManual/mfskManual/dominoex.html

Fldigi :

http://www.w1hkj.com/
« Last Edit: September 02, 2013, 07:39:22 PM by AF6WL » Logged
VE3UAL
Member

Posts: 18




Ignore
« Reply #41 on: September 02, 2013, 07:48:23 PM »

There are a number of other modes/methods but in examining them - while they all had their unique benefits - PSK31 is the most widely accepted and the easiest for us to enthuse others about. It does appear to do the job sufficiently reliably and we have invested a LOT of time in developing a new user interface. You may download it at www.webpal.org/SAFEems/

Peace and love,
Bruce
Logged
AF6WL
Member

Posts: 142




Ignore
« Reply #42 on: September 02, 2013, 09:25:40 PM »

My final comment:

Any inherent frequency drift of the low cost CB will be added to by the drift in your 24MHz transverter oscillator.

The MFJ 94XX series schematic looks like you could just replace the main VFO coil with a XTAL, swap a couple of caps and you would be done.

KD1JV's aptly named  'Survivor' 75m SSB/CW TCVR falls into the same camp and is only $100 : that must be worth a try !
http://www.qrpkits.com/survivor.html

Adding a switched in XTAL oscillator in place of the VFO would be even better as it would retain free tuning for voice comms when you find EMP has fried all the PCs within 50 miles and you discover that it really was not the best time to go looting your neighbours or Best Buy.

http://www.webpal.org/SAFEems/G.EMPProtection.htm
"The possibility is very great that one's computer is going to be destroyed by EMP. But, it is not likely that ALL computers will be destroyed - so that even in an emergency situation one will probably be able to find a computer that has not been destroyed. There are a lot of computers out there. Some in their original shipping in stores and metal framed warehouses."
Logged
VE3UAL
Member

Posts: 18




Ignore
« Reply #43 on: September 02, 2013, 10:18:51 PM »

 Cheesy You write with good humor - although the scavenging solution will not be humorous in that situation.

 Talked with the designer at MFJ two years ago and they wouldn't consider helping us make the necessary modifications.

Ditto for Doug Hendricks the distributor of the survivor and  Steve Hendricks its current designer. Just didn't think it possible.

Presently waiting some schematics from Dave Cuthbert for this other path. So far nothing is really opening up.

Peace and love,
Bruce
Logged
WX7G
Member

Posts: 6041




Ignore
« Reply #44 on: September 03, 2013, 08:36:37 AM »

I'll email the preliminary schematics this evening. If this looks good the next step is a design review perhaps with the schematics posted here or at webpal. The design is then refined for PCB layout or rejected. I would go for a 4-layer PCB on this simple design. I'm sure a 2-layer PCB would work but there is higher risk. My goal is always to go from a paper design to a finished product with no prototyping. 

I'm not saying the CB transverter is the ideal solution but it is very simple and something that can be built and tested in a couple of days. It provides 40 channels within the 80 meter band. There is the drop in system efficiency by having the CB radio transmitting at full power into an attenuator. The upside to running the CB radio at full power is that no ALC loop between the transverter and CB radio is needed. The components are sourced by Digikey and Mini-Circuits. The design philosophy is to go for low parts count with the tradeoff being increased parts cost. The reduced parts count reduces assembly cost, PCB layout time and size, and increases first pass assembly yield.

Being that the CB transceiver synthesizer is referenced to a crystal oscillator it should be as stable as a typical amateur transceiver tuned to the 10 meter band.

I think the ideal solution is a box having no controls and only three I/O's: RF, DC power, and USB.

« Last Edit: September 03, 2013, 08:44:56 AM by WX7G » Logged
Pages: Prev 1 2 [3] 4 Next   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!