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

donate to eham
   Home   Help Search  
Pages: Prev 1 [2] 3 4 5 6 7 ... 16 Next   Go Down
  Print  
Author Topic: Better IMD please  (Read 121566 times)
VK4DD
Member

Posts: 79




Ignore
« Reply #15 on: November 21, 2011, 12:20:22 PM »

QST sounds like a good magazine to me.
A pity we dont see it down under.
Logged
W1BR
Member

Posts: 4189




Ignore
« Reply #16 on: November 21, 2011, 12:20:53 PM »

   
If you tell them that their rig which they bought for 1200 bugs stinks... than you got a real problem. So I wont. Lets face it we are a group of idealists, who know that rigs were better 30/40 years ago.  73 Ron VK4DD


The only way things will improve is if magazines publish honest reviews,
and if hams are educated enough, and willing enough, to walk away from
a product that isn't quite what it should be.

Pete

Mhh thats like pointing the gun at their own head.
They publish about their advertisers.

Ron

The ARRL lab does a good job. They  (QST) have the highest advertising ratio of any ham
magazine these days.

Almost everyone else is doing "fluff" journalism with no concrete lab based data.

Are you saying the ARRL is biased in favor of their advertising base?

Pete
Logged
VK4DD
Member

Posts: 79




Ignore
« Reply #17 on: November 21, 2011, 12:37:08 PM »

I dont know..... I. dont see these magazines down under.
Down under we say... dont bite the hand which feed you.
Logged
M0HCN
Member

Posts: 566




Ignore
« Reply #18 on: November 21, 2011, 12:45:55 PM »

If I recall Radcom have on occasion commented in a review on a rigs poor transmit performance.

The ARRL reviews I do not really get, I mean they do all the lab work (For all that some of the methods are debatable) then don't publish it unless you buy the extended report, and some of the reviews read like the reviewer has not bothered to read the lab work.

Some of us are not just moaning about IMD, we are attempting to solve the technical issues behind building a much better transmitter without horrifically increasing the cost.
Prototypes are **expensive**, that is the nature of the game, but if I can cost optimize the loop enough and get the power efficiency up, there might be a net cost saving overall on a rig in the few hundred watt class with the loop and envelope modulation of the power supply over a conventional one (Due to the reduced cooling requirements).

There will always be a few AKTTR operators and a few 'screwdriver experts' who can stuff anything up, not much to be done there apart from shouting at them.

Regards, Dan.
Logged
KM1H
Member

Posts: 5281




Ignore
« Reply #19 on: November 21, 2011, 05:21:39 PM »

Quote
The SR-2000 Hurricane, basically an SR-400 with a pair of 8122s in the final. It cost $1095 in 1965, which inflates to about $7500 today. That's without the power supply. My guess is the power supply would bring the price to at least $10,000 in today's money.

There was also the NCX-1000...

I have a pair of the latter here and they are no IMD gems, about -28 3rd is about all you will get out of an 8122 without extreme measures such as a G3SEK type of mod.

There are devices on the market now that work at 1.8-600MHz, shrug off extreme VSWR at rated power, high efficiency, run serious power and arent expensive per watt. IMD meets critical multichannel data requirements.

http://www.freescale.com/files/rf_if/doc/white_paper/50VRFLDMOSWP.pdf

Here is another example with lower VSWR and frequency specs but at $205 in single quantity from Richardson (the most expensive) its competition for the 3-500Z even on CW when properly cooled.
 
http://www.freescale.com/files/rf_if/doc/data_sheet/MRF6VP11KH.pdf

While tube prices soar LDMOS will be coming down.

Carl
KM1H
Logged
W8JI
Member

Posts: 9748


WWW

Ignore
« Reply #20 on: November 21, 2011, 05:31:38 PM »

There are devices on the market now that work at 1.8-600MHz, shrug off extreme VSWR at rated power, high efficiency, run serious power and arent expensive per watt. IMD meets critical multichannel data requirements.

http://www.freescale.com/files/rf_if/doc/white_paper/50VRFLDMOSWP.pdf

Here is another example with lower VSWR and frequency specs but at $205 in single quantity from Richardson (the most expensive) its competition for the 3-500Z even on CW when properly cooled.
 
http://www.freescale.com/files/rf_if/doc/data_sheet/MRF6VP11KH.pdf

While tube prices soar LDMOS will be coming down.

Carl
KM1H


Carl,

Please tell us what page gives the "IMD meets critical multichannel data requirements" data.

The links you posted appear to be for very short on time, 10% duty cycle, pulsed operation. Did you post the wrong links, perhaps?

73 Tom


Logged
M0HCN
Member

Posts: 566




Ignore
« Reply #21 on: November 21, 2011, 05:35:23 PM »

Be careful of the claims for a lot of that LDMOS, the elephants in the room are:

1) Some of the specs apply to pulse service in applications like MRI machines, not essentially continuous carrier modes.
2) Cooling, a right nightmare at those power levels, IXYS in particular are guilty of rating things at 25 degree case temperature with a really nasty derating curve (You try maintaining 25C case at 1000W output without LN2).
3) Linearity, some of them are not very up near rated power.
4) Some of these parts are really intended for VHF service, which makes stability if you try to use RF feedback to tame the nonlinearity problematic. Not insurmountable, but a serious ball ache.

Read the datasheets CAREFULLY, particularly the input/output power graphs and the thermal derating curves.

Personally I would probably rather use two 300W devices instead of a single 1KW one (that will only actually manage 600W due to heat management issues), it is just easier all round.

LDMOS is probably the future, but do not get too hung up the the headline numbers for those freescale parts.

Regards, Dan.
Logged
VK4DD
Member

Posts: 79




Ignore
« Reply #22 on: November 21, 2011, 06:34:53 PM »

Hi there,

Yes Dan is pretty much spot on. The 300W device the MRFE6VP6300H from freescale is pretty much bullet proof. This part takes a lot of punish ment, I have played with them and even give away PCB for hams who want to make a 2m amplifier with them. They pay postage thats all.
All details are in my forum. www.vk4dd.com/forum . People need to register to see this section.
As always the lazy ham gets nothing.

NXP has also nice devices and the BLF578 is more linear in my opinion than the equivalent freescale. The parts are not as tough as the above mentioned 300W device.

The problem is that these devices beep all the way to 600 Mhz or so. Which may be a bit tricky if you want to make an HF amplifier with them. Even trickier with a feedback circuit, because before you know it these things could take off and start to oscillate. I am not saying it isnt possible... but I havent seen any one doing it, thats all.

Two tone is pretty good also provided you stay under 250W on 2m, maybe on HF more.
Even 300W produced really good numbers. I am not from NXP or Freescale, I have just experimented with these things and designed some VHF stuff for them.

73 Ron
VK4DD
Logged
W6RMK
Member

Posts: 680




Ignore
« Reply #23 on: November 21, 2011, 08:24:04 PM »

Be careful of the claims for a lot of that LDMOS, the elephants in the room are:

2) Cooling, a right nightmare at those power levels, IXYS in particular are guilty of rating things at 25 degree case temperature with a really nasty derating curve (You try maintaining 25C case at 1000W output without LN2).

Read the datasheets CAREFULLY, particularly the input/output power graphs and the thermal derating curves.

Personally I would probably rather use two 300W devices instead of a single 1KW one (that will only actually manage 600W due to heat management issues), it is just easier all round.

Oh man... did we encounter that.... nice small kW+ amp, big honkin rack sized chiller for the liquid cooling.  And a moment's inattention when breadboarding..oops, you just bought another set of transistors...

However, there's no question that some of the parts becoming more available now.. and more importantly, switchmode power supplies and clever DSP to do predistortion/operating point setting... we're on the verge of a revolution in high power amps...
Logged
ZS5WC
Member

Posts: 713


WWW

Ignore
« Reply #24 on: November 21, 2011, 11:08:35 PM »

 :)I do like some comments posted here--I have been looking at new LDMOS HV transistors as well--it is only a matter of time before we break the Tube vs. Transistor Dollar / Watt barrier.
Problem still is the large LPF networks required. I also don't see why a PI network cannot be used with HV transistors, apart from the 'Instant QSY' problem.

Regarding the current crop of 100w rigs--I do think they could benefit from a feed forward dynamic bias loop--BUT, I suspect the issue at hand is undersized cooling, as eveything HAS to fit inside a matchbox today.. Angry
Also the CHOICE of RF mosfet used in these PA's are all about bean counting and not quality.

73 de William
ZS4L / ZS5WC
Logged
VK4DD
Member

Posts: 79




Ignore
« Reply #25 on: November 21, 2011, 11:49:24 PM »


Regarding the current crop of 100w rigs--I do think they could benefit from a feed forward dynamic bias loop--BUT, I suspect the issue at hand is undersized cooling, as eveything HAS to fit inside a matchbox today.. Angry
Also the CHOICE of RF mosfet used in these PA's are all about bean counting and not quality.

73 de William
ZS4L / ZS5WC

William,

Yes cost is always a factor... hams are very concious buyers, usually because they need to have permission from the minister of finance Wink

What is your preffered mosfet for a HF PA, and why.
What is the preffered solid state PA for HF / which brand and why if known.

73 Ron VK4DD / PA4K
« Last Edit: November 21, 2011, 11:52:04 PM by VK4DD » Logged
ZS5WC
Member

Posts: 713


WWW

Ignore
« Reply #26 on: November 22, 2011, 01:34:11 AM »

 Smiley Hi Ron,

Thinking about it logically I guess we don't have much other choice than the Mitsubishi mosfets, gain, cost and IMD wise.
(These are used in..IC7600, FT-950 , TS-590....etc!..)
Emotionally, I am still a bit old school and would have used a bipolar instead--but costs dictate the use of newer devices. RD100hhf1 mosfets will typically deliver a 100w @ -31dB 3rd order IMD, whilst old bipolars, like MRF454 (ancient as they are) will typically run 100w in push pull at -37dB.
Over 4 times better than the fets.
I also don't see why there could not be a didicated base version with HV Mosfets--some can run happily at 250v, the PSU requirement could be very simple too, and IMD figures good with an active bias system.
Some local military radio manufacturer used this active bias system to great effect, and quoted figures better than -45dB!.

Regarding QST--If you subscribe , you'll get it in OZ no problem.

73 de William
ZS4L / ZS5WC
« Last Edit: November 22, 2011, 03:04:14 AM by ZS5WC » Logged
TANAKASAN
Member

Posts: 933




Ignore
« Reply #27 on: November 22, 2011, 02:37:21 AM »

The technical discussion here is fascinating however I would like to propose a far simpler way to reduce splatter and IMD on ham radio transmitters.

Teach people how to use the microphone gain control. Then teach them not to overdrive their amplifiers

A certain amateur that I know (who shall remain nameless) has a loud booming voice which increases by at least 3dB whenever DX is involved. In addition to this the microphone gain control on his rig is turned up all the way because he falsely believes that this is how he should get 100W out on SSB. I know that this amateur is not the only person on air using this technique.

So, why not have a simple LED by the microphone gain control triggered when the modulation level to the transmit mixer exceeds 100%? Add a second LED to the front panel of the linear amp that shows when the box is being overdriven.

We can tackle ALC, IMD and gain distribution later once we have eliminated operator error.

Tanakasan
Logged
W8JI
Member

Posts: 9748


WWW

Ignore
« Reply #28 on: November 22, 2011, 03:03:44 AM »

Tananasan,

If mic level changes bandwidth, the radio has a problem in design.

I suggest we find out what radio the fellow has and learn why he can turn the mic up to make his signal wide, and get manufacturers to correct that design error.

73 Tom
Logged
M0HCN
Member

Posts: 566




Ignore
« Reply #29 on: November 22, 2011, 06:18:56 AM »

There is also this thing called an audio limiter (possibly with a speech clipper as well), leave it hard wired in circuit all the time set to limit at whatever level corresponds to 100% mod, and turning the input up will just make it sound nasty (but no wider).

You can implement in either DSP or a single cheap analogue chip.
An RF speech clipper is optional (But is about the most benign way to make transmitted audio louder).

ANTTR operators are a pain, but are actually a minority, ALC is a problem (and I never connect it, if you know the amp gain and output power you can just set the exciter power per band and never need ALC, which is good because that loop is NEVER designed correctly).

Some of the LDMOS parts are actually competitive with the bipolars in IMD terms, and can usually get away with a simpler bias network (No need to have a bias network that can supply an amp on peaks), but you need parts that are actually designed for linear operation, which adds cost and are more specialist.

Granted, education is always a good thing (And is part of what this hobby is supposed to be about), but the rigs could be more idiot proof as well.

Regards, Dan.
Logged
Pages: Prev 1 [2] 3 4 5 6 7 ... 16 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!