The AM Forum

Thought these pix I took today would be of interest. It's a new Nautel 50 KW transmitter. The size is amazingly small, like a large refrigerator. No more taking up half a room. Just incredible.
The internal 700# power transformer takes up 1/3 of the cabinet. It'll do 55 KW and 140% positive modulation at the same time (into a dummy load, of course!)
Eat your heart out, JJ.  ;D

There is an active touch-screen monitor on the front. Not only traditional stuff like modulation % is displayed, but there's also an active Smith chart and spectrum analyzer. The blue line on the spectrum analyzer is the legal mask that the transmitted spectrum must comply with. The little 'boxes' on the left and right of the carrier are to accommodate the IBOC subcarrier. You can see that at 40.8 KHz above the center, the transmitted crud is 76.12 db down. Move your finger across the display to wherever you want to measure. The Smith chart is showing that at 910 Hz above the carrier, the antenna load is 49.525 ohms resistive with only a tiny amount of reactance. Cool stuff! This is what my next homebrew rig should look like- LOL

The tubes on the floor behind me are a pair of 4CX35000s a retired transmitter once used.

And here's the Smith chart:

Awesome Bill!   I had a chance to see one of these on my W7TFO tour of the mountain.  It was not up and running when I was there but I was very impressed with it.  Probably pays for itself in the power it saves.

Nice pics!

C

Bill,
WE need information. I wonder if it is a digital like the Harris DX50?
Schematics, final part numbers, And most of all. How about information on the combiner.

Yes, it is of course solid state digital.
I'll look up how many pills it incorporates and other details once I grab the manual. Give me a few days. It's all new to me. It contains a very complex piece of software, and it's all Linux.

We do have a Harris DX-50 on site, which is now the backup transmitter, and it's been a workhorse but this new rig is insane with what it can do. It's a fraction of the size of the DX-50. I'll bet it could do 100 KW out if the software allowed it. Got a peak power meter in it, I've seen 300 KW PEP out of it while testing into a load, maybe 55 KW and 130% modulation.  Brought a big grin to me. That pretty much maxes out the 3" rigid line it feeds. Damn, it isn't much bigger than an old timey KW-1.  Inside the box, it's about 1/3 power supply, 1/3 RF and 1/4 computer.
The combiner is internal to the cabinet. The final matching net is made of paralleled 3/4" copper pipes, one big sucker of a coil. The 700# power transformer came shipped on casters, uncrate it, roll it into the back of the cabinet with a pry bar and bolt down the cables. IIRC, it's about 90% efficient. With electric bills running ~$4,000/month, it probably does pay for itself over a few years. Hang on, I'll be back.

There's a couple of dozen of these things on the air around the US now.

I wonder if it could be retuned to 1885...? Into the big stick? You can set the carrier frequency in a video screen operating the exciter, LOL.

How technology has evolved.

The capabilities of that transmitter are simply flat, freakin' awesome!

But, you know, I'd still rather have a Collins 20V-2 or a Gates BC-1J with those magnificent 833As glowing through the window in the front.

I admire the myriad functions of the new gear, but I'd rather live with an old hollow-state rig any day of the week.

Just like vehicles. I enjoy driving my pickup even though it has a computer in it. It's a '93 Dodge Dakota -- not overly computerized but still not fully mechanical. All things considered, I wish I was still driving my '66 Chevy 3/4T with a 325 HP 327 CID 'Vette mill in it. Not a computer in sight and I could keep it running with bubble gum and baling wire.

The above opinions may well be a result of the fact that I turn 60 a week after Friday.

ldb
K5WLF

The Smith Chart is an amazing piece of valuable info.

Fred

Bill,
A number of us would be interested in any combiner information you have for the DX50. I had a look at the manual years ago but didn't get a good look at the combiner design.
The DX50 used old FETs, I bet the newer devices work a lot better.
I think the DX50 ran 85% so making the jump to 90% is quite impressive. 

hey Bill
Are you working in an A.M. station playing music?? That would be an outrageous sound.
I hope the processing will make maximum use of the allowable bandwidth for great TX audio.
fred

Bill,

That bandwidth measurement looks pretty good considering you're down at 850. Did you do any work to the stick to have it optimized? When Nautel came out with this box last year, I asked if the Smith Chart was a true representation of what was going on or was it just a reference, and they assured me that it was the real deal. Spectral regrowth looks pretty good too;it looks like it just meets mask, but most are close anyway. The best spectrum shot I have ever seen was WOR's in NYC. TR has his regrowth down -72dB! Looks exactly like the AM output right out of the exciter.
Good luck with that TX, but you won't need it,that box will give you years of service!

Mike

One word....Awesum.... ;D

I'll pull the manuals for the DX-50 and Nautel and detail the combiner arrangements, I just don't have the time to do that right now.

Fred, the station is newstalk format, no music at all, except for that talk show that plays the Pretenders song for an intro- LOL.

Mike, the tower is 5/8 wave, 666' tall, matched with an L network so it's going to be fairly broadband. Never changed a thing when we originally went IBOC with the DX-50. We made the customary bandwidth measurements and it was good. The whole ground system was redone some 5 years ago, 120 new 700 foot radials of #6 copper. The old radials are still in place, a few might be damaged, so there could be actually be 200-some radials in place now.

If it was me, I'd never put up a 666' tower; that's why it gets hit by lightning.  ;D

Ahhh, that explains it. Being 5/8 wave, the selective fading must wreak havoc on the IBOC, and the analog for that matter, about 20 miles out at night.

Nice to see a picture of you man, you're lookin' good.

What's the point in IBOC if it is all talk and no music?

I'm pretty sure the Nautel tx uses a PWM modulator.  The PWM is digitally derived, hence their claim of 'digital' modulation.

The high power Harris transmitters have a series combiner, with each individual PA on a toroid transformer, with the secondary a single turn, a copper pipe down the middle of all the toroids, like beads on a string.

Hey Don
It's just nice to have it. If Rush moves his head on the stereo stage you'll hear it!!!!!LOL
News/Talk network. No personality radio. The new broadcast industry....A lot more players......more diversity.......less music........no giants.

And I guess with the computer screen display there are no meters to see dancing around?? The should be some serious current changes with solid state circuitry.

Fred

Rush and all the other politico-talk windbags could just as well broadcast on slopbucket, and the average listener wouldn't even tell the difference if he didn't have to know how to tune it in.

agree since most of the wind is coming out the 6

I would be surprised that a PDM rig could hit 90% overall.

Yep, same goes for the windbags on NPR. Living off the taxpayers and blowing hot air. SSB would be too good for them.

The transmitter is nice but I'd rather have those 4CX35000's.That is a true work horse.

N8QPC

According to the Lit, it is a high sampling-rate Class D transmitter

http://www.nautel.com/Resources/Docs/Brochures%20&%20Spec%20Sheets/HD%20Radio/NX%20Series/NX50_w_b_1.2.pdf

with 20 RF/modulator modules producing 2.5kW carrier each.

I would think it would be capable of 325kW PEP for 150% modulation.

Phil - AC0OB

I LOVE the bumber music that George Noory plays on Coast to Coast AM.  8)

Even talk radio stations have SOME music to consider.

Wow Bill, that new xmtr is impressive.  Here is a glimse of what is used to take to do 50KW.  Quite a contrast.

http://historyofwowo.com/tech.html

73,  Jack, W9GT

Like it or not, whether you pay taxes or not, you are paying out of your pocket for the mindless garbage on commercial radio and TV whether you watch and listen to it or not.  For example, if you pay medical insurance premiums or purchase prescription meds, you are subsidising the TV broadcast industry.  Who do you think ultimately finances the "ask your doctor" commercials?

Bill, congratulations on getting the Nautel there.  I of course have never even been in the same room with an operating 50 KW rig but I'm told avoiding a 3DX50 is something to be thankful for.  I understand if ur aux is a dx50 that's different.
Fred, I think WSM is the only 1-A still programming music.  They sound great but I don't believe they're doing any digital; analog only.   A few years ago the licensee wanted to take them to some kind of talk format and so many people in Nashville raised hell about it that they backed off and stuck with music.

I think to date the IBOC thing has no listeners (or very few--does anyone have any non-ham average radio listener friends who own digital AM receivers?) because the AM broadcast demographic is pretty much people like us, namely, adults with a common sense view of technology--if it ain't broke, leave it alone.   Analog AM seems to work just FB thanks, but I'll bet the managers running the business now are all folks young enough to be our kids.  To them, anything digital is WONDERFUL and it goes no farther than that. 

For a lot of radio listeners the place where the listening gets done is in the car.  If manufacturers start putting IBOC receivers in new cars, AM digital listening may pick up.

For voice though, which is most of AM now, I think it is a waste.  In fact, domestic AM stations could go analog and start rolling their audio off at around 6 KHz and it would probably be okay.

If Limbaugh would ever talk about studio work, and hosting a talk show, two things he actually knows something about, I'd listen to him.

Not my taxes.  

Funding commercials because I buy a product is hardly the same as funding a network via taxes that is in bed with the government. You're flailing now Don.   ;D

Do you have a choice, if you take out medical insurance?  I'd bet those Ask Your Doctor commercials siphon off as much, percentage-wise, from your premium payments as public broadcasting does from your tax payments.

Yes, I can choose to have health insurance or not. And I can choose which health insurance I buy. Either way, no taxes are involved, so I really don't see any support for your premise. Did you hear such nonsense on NPR?

You obviously have never listened to NPR and you're totally clueless to what is broadcast there.


You are not forced to contribute tax money to NPR or anything else you disapprove of, any more than you are forced to pay for advertising through the purchase of commercial products. You have the right to terminate your employment.  Just as you can choose not to have medical insurance (although you will contribute even more to the Ask Your Doctor commercials by paying for your pills out of your pocket), you can choose not to have a job with a pay check, and therefore avoid any tax liability. Subsist off the land, grow your own food, make your own clothes, and barter for life's other necessities so that no money ever changes hands.  A political candidate made national headlines recently at a public forum when she encouraged citizens to try bartering with their doctor.  "In the olden days, our grandparents, they would bring a chicken to the doctor." Chickens for check-ups (http://crooksandliars.com/john-amato/sue-lowdens-chicken-checkups-health-car).

BTW, I don't think NPR and PBS get a lot of their money from tax dollars any more.  The majority of their financing comes from corporate underwriting.  Local stations also depend on commercial underwriters but about 50% of their income comes from voluntary contributions from listeners. Our local station accumulated over $100,000 during their most  recent pledge drive.  If that many people were willing to put their money where their mouth is, the station must be doing something right.

The fine line of distinction between "underwriting" and commercial advertising is becoming increasingly blurred. Who says there are no commercials on public broadcasting?  But at least their  "underwriter acknowledgements" are far less obnoxious that most of the commercials that appear on the vast wasteland, and they don't (yet) interrupt the programming.  The day they start interrupting the show with those announcements is the day I'll stop watching and listening.

Um, did I miss something?
I thought that the new healthcare law will require us to have health insurance starting in 2014.  If you don't buy the insurance you will receive a tax penalty for the year.
Was that pulled out at some point?

Um... Did I miss the post about a new broadcast transmitter?

     Where did this item become a discussion about taxes and PBS/NPR?

     Bill, nice transmitter.  My brother there in Denver listens to that station...

     Actually his office, till he retired and sold his business, was just across the
     way from the current PBS studio complex.

When the NPR zombies showed up to defend their taxpayer funded hot air shows.

Wow, that is a great transmitter.  Plus you got a lot of cool and usefull features.  Man that Smith chart display is great, I WISH I could have a VNA in the shack. 

Any chance for some internal pics? -- just curious.

Actually, the thread hijack started before NPR/PBS was ever mentioned, when politico-ditto heads showed up to defend talk radio windbags against the suggestion that AM "talk" could just as well use slopbucket as IBOC / "HD".

Yup, will do soon. I worked an overnighter last night and need to chill out today.

I took a pic yesterday of the front inside, the back door is interlocked and I didn't want to shut it down.

The thing only vaguely resembles a traditional radio transmitter when you open the door. 20 years ago, I wouldn't have had a clue what was the box was for, LOL. It occurs to me that there is not a single RF tuning adjustment in the thing
The top 1/3 of the cabinet is all computer, control and display. The exciter is the dark green board in the center. The frequency can be locked to GPS. Unseen behind the top panel is a large inductor which is part of the output pi network. The middle section of the box contains the 20 plug-in RF modules. The 3-1/8" rigid feedline and power connections come out of the top. The combining network is a series transformer arrangement where the output of each RF module is shorted across when it is removed to maintain the series connection on the primary of the combiner. All the modules are hot-swappable and overall the box will still make 50 KW with two removed at the same time.

The CAT-5 cable dangling from the CPU on the left door is connected to the corporate WAN, you can VNC right into the control screen, I do from here at home.

Picture #2 is the ATU at the base of the tower. The buried 4" Heliax feedline from the transmitter enters the shack on the right. That's the big brass fitting. The larger of the inductors is wound with 1-1/2" dia tubing. There is a pair of 1000 pf, 30 KV Jennings vacuum caps on the panel to the right which I will need to replace this coming week. Too many lightning hits over the last 50 years. You can hear the radio station audio FB in the doghouse, the inductors rattle with modulation and make tinny-sounding audio transducers.

And for you solid stare dudes (Steve?), the output beans are vendor P/N APT5010JFLL. Made by Advanced Power Technology. Data file here:

http://www.microsemi.com/datasheets/5010JFLL.PDF

So, that would be 4 per module x 20 modules = 80 fets in the output of the thing, about 700 watts carrier output at W/125% modulation per bean.

Okay Bill, very cool.   Speaking of cool, how much air moves through the box?  Or, more important, what's the fan setup?  I imagine a few big axial fans moving air in and out.  I wonder how loud they are. 

The ATU was interesting to look at.  I didn't see the caps, maybe because of the photo resolution, so I wondered about them.  I saw what looked like two inductors in series and the one 1.5 inch tubing one to ground.  so I was thinking there's gotta be a capacitor in there somewhere  :D.   The RF current sensor and static drain choke both were a bit smaller than I expected--must be new technology hi hi.

Rob

They're axial fans and not loud at all. They suck air in from a filter on the back door and duct it out the top. Remember, the box is 90% efficient, so not much heat.


Rob, good guess, that static drain choke, it's an autotransformer for the mod and air monitors. Maybe 200 turns of #16 and a tap one turn from the bottom end feeds a small coax line into the TX building-

The stick is at DC ground via the big inductors at the upper left and right. Draw it out with a pencil, it's a tapped L network.

Nevertheless, the peak current in lightning happens in around 1 uS, or a megahertz, so it might be a low Z for DC, but not necessarily for lightning.

The meter on the left is a Delta Electronics RF ammeter. They don't make the old-timey bakelite ammeters any more. It's RF sample comes from the octagonal current transformer just above the meter. Also sends an RF current reading to the remote control.  Man, I got one heck of an RF burn just touching that RF ammeter's case..Supposedly grounded.. That is one scary room to be in with the parts buzzing with the audio, I need to get some brass screen underwear..Or chain mail armor..

Here's the two vacuum caps. Their ends are pointing towards the camera.

Thanks Bill, I see why I could not make out the caps, they were end-on to the camera.   It's always interesting to see that the high power tuners are basically like ham tuners just bigger

Yes I was puzzled by the big choke and the only thing I could think of was static drain but the ones I have seen for that at lower power stations were longer. 
So I thought because it looks like a newer ATU someone figured out they could get away with making them smaller.  Well autotransformer explains it.   Ok on the RF burn--wow that must have been a real scorcher.  I have a Delta RF current transformer here that came out of a 1 KW station, it is the model TCT-3, 1 V / Amp and I'm going to do something with it when I get around to it, not sure exactly what, but I'm hanging on to it because it's one of those things I don't see every day.  I got 3 of them originally when a local station with a 4 tower DA shut down and one of the guys who managed it let me have all three for around $15 or $20.  At the time I only vaguely thought they might come in handy but the price was so low I got them.   I guess one idea I had was to put it out at the inverted L and run wire or tubing from the tuner out there to the base of the L then run a coax line back to the shack and look at the current on the antenna in the shack.  That might be useful if I ever put a remote controlled tuner out there.  I also got a static drain choke I put out between the bottom of the 75 m. vertical and ground.
Yeah I've heard about the parts singing in tuning houses before--must be kind of spooky.

Dig the Ciss or the input capacitance! Not necessarily out of range for high current FETS.

They must be using about 250 Volts Vdd?

Are they driving theses modules with a sine-wave or a pulse?

In the Harris system, they would detect the driving waveforms' zero crossover before unclamping the (lower) gate transistors in the classD Full-Bridge amp.

Phil - Ac0OB  

Phil, the PA stages run on 400 VDC. Two FETs are in series across the B+. The gates are driven with a varying duty-cycle square wave. Yes, each PA stage consists of four FETs in an H configuration. The RF load would be the horizontal part of the H. The upper right and lower left FETs turn on simultaneously, then the upper left and lower right FETs are turned on.  So your guess was close, each FET sees 200 volts. Their gates are driven with transformers.

Thanks for the info Bill. Nautel has some nice technology.

It appears to be the modulation method in which the duty cycle ratio of the RF pulse is applied to the gates, sometimes called the "monopolar" PWM system.

Instead of applying a filtered PWM signal to the drains (as in conventional ClassE), the duty-cycle or width of the input switching waveforms is applied to a wide-band gate transformer, which drives the finals (4 pills) in an H-bridge configuration.  

I wouldn't be suprised if those (gate driving) input transformers were transmision line transformers.

Phil- AC0OB

I always run transformer coupled drive in my PDM rigs for isolation. My 160 meter rig runs square wave drive with a BB transformer. PDM at gate drive would be hard work above BC band. 1% modulation is a high frequency but getting it to function flat to 100% is a trip. HMMM I wonder if there is some DSP involved in making it clean. Lot of stuff happening every 1/2 microsecond. Now if we could duplicate it at 125 ns. I'm dreaming of an FPGA with a tracking A/D
Very cool design. I can see how it is more efficient than a DX50. All modules work all the time so there doesn't need to be any extra swithes to remove B+ from the bridges turned off and shorting the combiner primary winding.
 

I am sure there is some feedforward DSP going on as well as some phase and amplitude correction circuits keeping track of things and making high speed adjustments.

Phil - AC0OB

Here's a couple more pix from yesterday. Had to fix a problem in the series combiner and I had the hood up.

First is the PA matching network which is located in the upper 1/3 of the cabinet. The upper 1/4 box has two of the biggest ceramic caps that I have ever seen. Not mica or vacuum, ceramic caps made in England. (The blue things) Maybe 6" in diameter and two inches tall, end to end. Like monster versions of those TV sweep fiilter caps from the 1950s. The coil on the lower right compartment is made of paralleled 3/4" copper water pipes. Some serious I in that inductor.

Next is a closer-up of one of the 20 (ea) PA modules. The four RF PA transistors terminals are on the right, 10 KW PEP and more out of just 4 pills. Wow.
Yes, they manage to run 10KW+ PEP through those PC edge connectors out to the series combiner. Modules are broadband, 530 KHz to 1700 KHz. Who needs PL-259s? I never would have guessed that one could use those common PC card connectors for lots of RF.

The modulator FETs are on the underside of the board so you can't see them (Yes, they still make Orange Drop caps!)- The caps on the board are soldered to the modulator FETs beneath.

Amazing that Nautel used Orange Drop polyester capacitors on a surface mount board. I guess they had a good reason for that approach. Looks like one of us hams did a repair and stuck those caps on the back.....

Thanks for sharing the photos and info, Bill.