Hi Dennis

<<<that, others do not. In one of my radios I had to build a box with a 9 volt battery and a precision pot to apply a steady voltage in to the ALC input on the radio. I call it my Dial-a-Watt. It absolutely limits the radio at the prescribed power down to a few hundred milliwatts with NO overshoot.

I doubt it. The only thing external negative ALC does is limit gain. It does not add clamping. It actually decreases dynamic regulation, although the reduced gain might reduce overshoot.

Think about what an ALC system is.
 

>Tom wrote:
"ALC systems in radios are commonly notoriously slow on attack, and virtually ALL have overshot on leading edges."

Tom, how many times do I HAVE to say it??? I DO NOT and never have relied on amplifier-generated ALC for ANYTHING when it comes to using an amplifier. You keep admonishing me on this point and I have said 5 times now that I don't recomend it for the very reason you>>

Dennis, the reason I keep repeating it is YOU don't seem to understand it is the internal ALC that limits power in a radio. It has NOTHING to do with EXTERNAL ALC.

The ALC system INSIDE radios virtually always contains some overshoot. It has nothing to do with anything EXTERNAL.

When you turn down a pewer control, you change bias on a comparitor for internal ALC threshold. Turning the power down INCREASES the percentage of overshoot.
 
It is a BAD idea to do that in most radios.

Tom wrote:
>To adjust the two, you have to deduct 6dB from specs >of radios and from ARRL measurements.

<<You're correct here, and I hadn't done that. But it doesn't change the fact that the IMD numbers on the amplifier deteriorate as you continue to drive it harder.>>

Yes, but my point is the radios we have are the limiting device unless you have something seriously wrong in a PA, or are using a tetrode amp with poor regulation and no negative feedback.

Inflating the IM performance makes your arguement sound valid when it isn't.

>>Many radios actually get worse at lower power, but the -38dB you quote above is below PEP and for two steady tones.

<<Yes, some do get worse at lower power, but most get better. Working the averages here.. especially since the amplifier's numbers get BETTER as drive power is reduced.>>

No, that isn't correct either. IM goes through peaks and valleys as drive is changed. IM is more related to negative feedback than power in a GG amp, because negative feedback is so high. Tuning the way I suggest actually increase negative feedback. Tuning the way YOU suggest decreases it, because it raises the anode impedance.  

>My comments were coming from wanting to enjoy the purity of emission that Eimac advertises for the tube.>>>
>Then you'd better go purchase a KWM2. ;-)
Actually Tom, I come pretty close to that.. my FT-102 is rated at better than -40db. With power reduced to 65 watts the 3 6146s are almost in to a Class-A condition.>>

When is the last time you measured IM in any rig? -34 single tone puts it about on par with an 80A being run hard IF it does what you think.

<<It's not just a "bandscope"..
It's a Tectronics 564 storage oscilliscope, with a 3B3 timebase and a 3L5 Spectrum Analyzer (with a mixer) connected to the output of the two receivers. When I hear buckshot on a signal I can tune it in, set the scope to "store" and in a few seconds have a pretty good picture whats going on. The only limitation is that the signal has to be fairly strong.>>


So it shows you what the guy is actually doing in his shack?? Must be a web cam then!!

Seriously, I have new test gear like a current Agilent analyzer with good digitized storage, and making an off the air measurement is almost impossible. Especially when I try to do it off the IF of a receiver!!!!! Measuring an IF port is next to useless.

Tom Wrote:
>>I really don't think either you or some "unknown" WB4 can trump Eimac's own engineers.

Tom, Did you look at the link I sent?? It isn't just some "unknown WB4" saying something... the link takes you to the ACTUAL EIMAC DATA SHEET published BY Eimac.

I have data sheets. I don't need to go web crawling to see them.

The data sheets are NOT final operating sheets. They are typical operation of what generally was a sample of one unit in a large fixture at Varian. What YOU keep ignoring is the IVS ratings were approved by Salt Lake, as was the manual. Eimac had no problem with it, neither have customers.

Eimac and other manufacturers very often approve modifications of operation outside typical guidelines.

How many amps have you designed? How often have you talked to Eimac's engineers about IVS ratings?

73 Tom

Tom:

Just to get things straight in my aging mind ...

<<<<You are saying to tune up my AL-80A with the full 100w drive from my TS-570 and THEN drop the drive back to give me either:

(1) the desired output
(2) whatever a max Ip/Ig of 400ma/150ma will give me.>>>

Yes. Absolutely. 100 watts or even more of drive, despite what some will try to tell you, will not cause a problem at all as long as the amplifier loading is heavy enough.

There is a technical reason for this, and it has to do with the very high negative feedback inherent in a GG amplfier. Most of the power is "fed through" to the output, there isn't excessive grid dissipation if you follow the manual.  

<<Am I correct? That's the way I have been tuning the amp to date. We are talking SSB here, not key-down modes or CW.>>

Even on CW, you should tune for full drive (assuming you don't have a huge rig, like a 150 watter).

Then back the drive down to rated (400mA) cathode current. This reduces voltage in the tank, increases negative feedback, and allows the amp to be much more more linear than tuning at 60 watts!!
 
<<Can I or can I not run with 100w drive for a full KW+ (PEP) output?>>

That amp is perfectly fine in IVS service if you load it correctly with 100w PEP drive. This all hinges on the fact you MUST load it heavily enough. The manual even says to go just slightly beyond the point of maximum output.

The key is watching grid current. Grid current reliably indicates proper loading. More than 150mA at full drive on carrier indicates excessive drive or insufficnent loading.  

The worse possible thing you could do is load it at low drive, and depend on ALC circuits to clamp the power. Virtually all radios overshoot, and even a few milliseconds of a transient can cause a damaging arc.

It's definately the lesser of two evils to tune at higher power and back off from that slightly.

73 Tom

Hi Dennis,

>>>Taken by itself, Tom's comment is correct. However, it is ONLY correct if subsequent operation after tuneup allows the full 100 watts from the exciter to the amp. Tuning up with 60-70 watts then driving the amp with 100 watts virtually guarantees splatter, excessive grid current, angry neighbors, and all kinds of bad things. I stated repeatedly that the power to the amp MUST be limited to the value to which it was tuned.>>

The problem with that statement is you are asking for an impossible operating condition, and presenting grossly exaggerated results for what really amounts to a non-problem.

ALC systems in radios are commonly notoriously slow on attack, and virtually ALL have overshhot on leading edges. Very few people have fast responding storage or long-hold meters to watch peaks, and very few people would understand what to look for.

Tuning instructions MUST be tailored for the likely operating condition and NOT something that is almost certain to cause problems.

When you start paying for warranty parts and become responsible for a manufacturer's reputation, you can re-write the operating manual! Until then, please let the factory do it.      

>output control. But Wendy, your FT-990 has a good one that very effectively limits the power output to the value you select without overshoot.>>>

...and we know that because???


TOM WROTE:
"At the rated IVS specs from Ameritron, the 3-500Z is substantially cleaner than most radios on the market. The IM limit is in the radio, not in the amp, if you load properly."

>>It is true that the 3-500Z is cleaner than most radios WHEN RUN WITHIN the specifications published by the manufacturer of the tube.(EIMAC) Whatever garbage goes into the tube also comes out... stronger.>>>>

First, Eimac specs that tube at continuous service, not IVS. I am using an IVS rating approved by Eimac was back in the early 80's. Let's not put that important point aside. Some manufacturers are not as dumb or careless as some people would have others believe. Eimac fully warranteed the tubes sold by Ameritron, and even reviewed the manual. There was never a single case of failure due to following the manual. Life is a non-issue. Eimac had no issues with it, lab tests showed no problems, and field failures or complaints did not reflect any problems.  

Second, you are mixing dB below PEP with dB below one tone. Eimac and I use the commercial standard method of dB below one tone. Radio manufacturers and the ARRL use the "non-standard" dB below PEP.

To adjust the two, you have to deduct 6dB from specs of radios and from ARRL measurements.

>>These days there are several very clean transmitters available. Wendy, your FT-990 is a very good -38db when running at 100 watts. Its even better at 70 watts.>>

Many radios actually get worse at lower power, but the -38dB you quote above is below PEP and for two steady tones. That translates to -32dB PEP, which is NOT particularly clean when you consider an old KWM-2 from 1960 is typically in the -40 to -50dB range.

The AL80 series maintains IM performance even with voice signals, the FT-990 will not. The reason is radios have notoriously poor biasing and power supply systems, and they waddle around with various frequencies loading the supplies. They just aren't as good as you think, and certainly are the major limit in system IM. The exception is some commercial tetrode amps with poor regulation and low negative feedback, they qactually are worse than radios!  

>>Some of these are "fine points" that won't quickly kill a tube when operated either way. The SSB duty cycle on an AL-80 is such that it won't burn the tube out of the socket if you hit it with 100 watts of drive. My comments were coming from wanting to enjoy the purity of emission that Eimac advertises for the tube.>>>

Then you'd better go purchase a KWM2. ;-)

Otherwise, you are radio limited.

>> As I tune around the band these days and look at the scope attached to the I.F. ouput from the receiver I see far too many "dirty" signals.>>

I'm not aware of any bandscope than can accurately measure signal BW. The measurement device would have to have significanly less BW than the signal it was measuring. Bandscopes do not.

Amplitude measurements from a receiver are worse yet. With AGC, they display the effects of AGC and ALC systems. Looking at audio peaks on a scope actually gives no indication at all if the signal is wide or not!!! The only accurate way to estimate linearity is to sample the RF in front of and after the amplifiying device.

The only accurate way to measure BW is to measure bandwidth. The only accurate way to tell the cause is to be at the station when it is measured, not 1000 miles away!!!

The 3-500Z saturates at over 2500 watts of RF output on peaks. It is VERY unlikely you are seeing tube emission saturation, if you are actually seeing anything at all that is meaningful.

Upper emission current in a thoriated tungsten filament is roughly approximated by filament current times voltage (power) times .1. In larger tubes it approaches .15 times power. If you can actually "see" flat topping on the scope from an amplifer, it is probably becuase the tube has VERY low emission or they have the loading control too far shut (probably from tuning at low power).        

>> It boils down to this.. Eimac BUILT the tube, not Ameritron... From the horse's mouth, the link below (courtesy of WB4HFN) is the Eimac data sheet on the 3-500Z. It shows the maximum ratings and "typical operation" specs for several types of service... Plate modulated class-C transmitter, oscillator/amplifier, linear.. grid-driven and cathode driven. NO PLACE on the sheet is there any permission or rating that suggests operation higher than 400 mils.>>

I really don't think either you or some "unknown" WB4 can trump Eimac's own engineers, as well as characteristic curves published by Eimac. Data sheets list "typical" measured specs for one form of operation based on 24hr a day 7 day a week operation.

IVS service is a well known form of operating.

Perhaps we should look at transformers, and see how thay are abused in amateur service?? A typical 35 pound hypersil has a "core rating" of about 1200VA. Why not use a 35-pound transformer in a single 3-500Z amplifier? Because of the service!!! This isn't CCS.

It never fails to amaze me how operating two or three of something a few hours a week makes people more experienced than people who have spent 30 years of 40-50 hour weeks looking at problems and talking with the design engineers of components!!

Had you quoted WA4GPM, who was a designer of high power tubes and engineering R&D manager at Eimac-Varian's power grid tube plant where 3-500Z's were made, I would have been impressed. But some unknown WA4 with a webpage showing a tube data sheet really doesn't trump the product engineer who looked at that design and said it was fine.

73 Tom

That's right, the EH antenna is a hoax.

Besides Steve's article, I have an article at

http://www.w8ji.com/e-h_antenna.htm

I agree with the other posts here. The only people who love em are the people selling them, and the people who have nothing else to compare them to.

73 Tom

The system is notoriously unstable. The transistors break into oscillation, and draw high current.

That's partially because matching system Q is high, partially because layout is very poor, and partically becuase the combiners are really not combiners so the parallel output transistors have additive feedback paths.

The bias system also is horrible, and the amps are very non-linear.

I sat in Morgan City for a few weeks in the early 90's and watched Mirage amps come off the line and virtually 100% of them oscillated. The techs used the next thing to "chicken claws and eye of newt" to make them stable.

The later 160w amps sold by MFJ are actually much better, being true push-pull designs with active bias systems. (As long as factory QC is good.)
They have a low-Q matching system, and are very stable unless you run them into a load like a cavity.

73 Tom

Good luck on fixing the bugs in the original Mirage amplifier! You can get a schematic from Mirage.

The original amp had a poor circuit design, and poor layout. It was subject to instability and other problems, and had a poor bias system.

What "bugs" are you going to work out? Are they related to stability? I spent some time looking at those amps several years ago, maybe I can suggest something.

73 Tom

I have some technical info about tubes and tube failures on my web site www.w8ji.com/amplifiers

All low and medium power tubes are going obsolete. Eimac dumped the glass tube line several years ago, and Svetlana was actually selling new-old-stock glass tubes like the lousy 572B's they sold.

Your only choice now is to buy glass tubes from limited vendors. Eimac makes smaller ceramic tubes still, and there are many surplus Russian tubes, but the real source where tubes are being manufactured now is mostly China.

Richardsons Electronics, who almost single handedly ruined the tube market for Hams, owns Amperex and they still manufacture 3-500Z's in France.

It looks like your choice is to use Russian Tetrodes or surplus triodes, which may or may not be around for a while, Eimac ceramic triodes, or glass tubes from offshore.

Myself, I'd stay away from tetrodes. They not only have higher splatter, they are touchier to tune. You also can not easily substitute another tube in a tetrode amp, like you can in a trode GG amplifier.

All of my amplifiers (I have about 15 of em) are triode tube type amps, except for solid state ones. You just have to shop around and see what you like.

Stay away from Swans and old Heathkits. Stick with something newer if you can afford it. Simple amps like grounded grid amps are more reliable, and easier to repair.

73 Tom

Dennis is trying to be helpful and I'm sure his intentions are good, but his detailed replys really contain some serious misinformation.

There is nothing wrong with Ameritrons tuning or operating instructions. You should ALWAYS tune the amplifier for maximum output with FULL drive from the radio, NOT with 60 watts as he suggests. Tuning with 60 watts will virtually guarantee excessive tank voltages (arcs, which by the way do NOT very often if ever come from parasitics)and flat topping (splatter).

What he forgets is that rigs commonly have ALC overshoot, and most have no real drive control. They depend on notoriously slow ALC to limit power. If you tune at 60w and the rig hits 100w or more for a millisecond while ALC kicks in, you can damage a bandswitch or tuning cap. Once an arc triggers it ionizes the arc path and even when power cuts back the arc will continue!!!    

At the rated IVS specs from Ameritron, the 3-500Z is substantially cleaner than most radios on the market. The IM limit is in the radio, not in the amp, if you load properly. People get dB below PEP and Db belone one tone confused. A typical radio is -33dB PEP, which is -27dB one tone. The 3-500Z run at 600mA peak in IVS service is -32dB to -35dB one tone (third order), which is better than most radios. The real splatter that annoys people is higher order IM caused by non-linearity in semiconductors in the radio, and extends out for several kHz. It isn't the close in stuff caused by a slight slope in waveform at the tube.

Of course if you underload the amp, all sorts of bad things happen. That's why you ALWAYS want to tune for maximum available drive, and NOT use a radio that runs 150-200w out with the AL80. You don't want to use a rig with significantly MORE than 100w to drive the AL80.

Most of the above is explained at and documented at:

http://www.w8ji.com/Amplifiers.htm

Also, the Ameritron factory specs were approved by Eimac, as IVS operating specs. IM distortion was tested, and the tube is absolutely fine running at Ameritron specs. The tube does not, as some seem to believe, operate fine for years at 750W and go to heck at 20% or so more power.

As for filament voltage, it makes a big difference when the tube is on 24 hrs 7 days but you'll never notice the life change in Amateur service unless you do something drastic like run 15% or more excessive voltage. Most 3-500Z failures are arcing related due to material or manufacturing defects such as outgassing of elements, seal leakage, or failure of manufactures to pump down the tube enough. Filament emission life failures, which are the only failure related to filament voltage, are VERY very rare.

The 3-500Z getters (removes gas) by a catalytic action in the Zirchronium coating on the anode. That's the gray looking coating. In order to activate, the tube has to be red hot on the anode. If you have a tube sit for months with being operated it can leak via seals or release trapped gas in the materials inside the tube and arc when you power it up! So it is NOT good to store 3-500Z's for a long time without using them.

People will worry you to death about the 3-500Z and other things, mostly because people love to talk about amplifiers, but the truth is most tube failures are because high power tubes are VERY difficult to manufacture corectly. Historically tubes are becoming less and less reliable as older people retire and manufacturers cut costs. Tubes have ALWAYS been the shortest life component in electronic equipment. That's not some secret, it's been known for many many years. Even low power low voltage easily operated tubes failed so often virtually every drug store in North America had a tube tester and sold tubes!!

All this chatter about doing this and doing that to save filament life and all the worries, and virtually ALL 3-500 failures are almost always due to problems entirely unrelated to HOW you operate the tube!!!!

The biggest sin you can commit with the tube is running the tube anode white hot to the point where it melts, or running it stone cold!! Neither is good.

The biggest problem for components like band switches and tank capacitors is tuning the amplifier with LOW drive at the last tuning step, like 60 watts drive!!!

You can see why on my website. It is ALL documented.

73 Tom

The 3-400 and 3-500Z are virtually identical tubes, both are "zero bias" triodes. A 3-400 will NOT be in cutoff when substituted for a 3-500, as a matter of fact they are virtually identical other than anode dissipation.

4-400A's (although tetrodes) make a good test substitute, and can sometimes be found cheap. Many older AM broadcast transmitters use them.

I have some stuff about tubes on my web site, www.W8JI.com

The problem with any of these tubes is very large glass tubes tend to be "leakers". Air gradually migrates in through imperfect metal to glass seals. Unless the tube is heated to at least a dull red on the anode regularly, it will not degass. The gettering (degassing) agent is the grey powder coating you see on the anode. It activates at over 1000 degrees C.

A second source of gas is outgassing from materials inside the tube.

If a tube has sat for a while, or is a "leaker" it will quite often pop and arc when you power up the amp. This may damage some components. Becuase of that, you should try to get a tube that has recently been used. A tube that is "brand new" in a box and has been sitting for years would actually be less likely to be good than one that has streadily been used.

73 Tom  

 

Almost everyone I know likes a pitch between 300-800 Hz. Very weak signals generally are easier to resolve at lower pitch, so weak signal ops tend to use low pitch.

At very high speeds (40WPM and up), people tend to lean towards higher pitches because it is easier to resolve dots and dashes.

I like 350-500Hz myself for most of my work which is DX and 20-30 WPM ragchews.

73 Tom

Many thanks
Let's say I have an AL-84 (which I do, but haven't
gotten it on the air yet.) Planning to use it on SSB,
driven by an HW-5400, and perhaps by a Valiant on 160
[driving the Valiant on 160 with an FT-817].
Does this make sense? What problems should I watch
out for?
73
Steve
WA2DTW


You might want to reconsider.

First, the Valiant is a class C power amplifier design. It has NO screen regulation, no bias regulation. It also uses the driver tube, which is NOt very linear, as an amplifier. It was fine when the bands were full of splattering AM signals, and nothing else for SSB was around. But the SSB input was at best a makeshift addition to a class C design.

The FT817 is bad enough as it is for transmitter IM products, it doesn't need help from class C stages.

The result of all that is you will almost certainly have IM products in the -20dB range, and occupy much more bandwidth than state-of-the-art. If you do that, at least keep the radio away from weak signal areas of the band.

Worse yet, you plan on adding another amplifier to the chain....after the 3 6146's in the Valiant. To gain what? 2 or 3 dB at the most? The IM of the amplifiers can all contribute to overall IM, and the result won't be pretty.

73 Tom

Actually the replies indicating GAIN is important on receiving are NOT correct, unless you have a receiver that is not into external noise floor.

See:
http://www.w8ji.com/receiving.htm

The correct factor for receiving is directivity, not gain. A negative gain antenna can have high directivity and be a better receiving antenna than a higher gain antenna, as long as the directivity of the lower gain antenna is higher and the response is aligned with the desired signal.

This means a 2dB gain yagi, with low gain because of loss, can be better (or worse) for receiving than a 10dB gain antenna.

This is why Beverages, even though they have negative gain, receive much better than yagi's on the lower bands.

You should NEVER consider gain directly for receiving, but it does apply to transmitting as long as the main lobe is in the correct direction. Directivity is the key for receiving, and loss does not factor into directity.

73 Tom

I have several C lines, and I even bought two of them new in the 70's or early 80's. The C line was OK, but by no means really good. By today's standards it is really a poor system.

You can see some receiver info inclusing measurements at:
http://www.w8ji.com/receivers.htm

Honestly, even though I like the Drake and have several, it really isn't that good compared to things out today. It is like the difference between a 56 Ford Crown Victoria, and a modern Crown Vic. You gotta love old stuff, even if it isn't as good in many areas.

Dial calibration is nothing like modern rigs, the PTO's drift a bit during cold starts. The audio systems are somewhat noisy, as is the third mixer.

The early SN receiver (under 18500 or so, with only the CW1 CW2 SSB and AM mode sitch positions)is exceptionally poor, with very very poor close-spaced dynamic range. Drake actually upgraded mine for me, because the MOSFET mixer and filter blow-by made the receiver useless for working close to strong signals.
Avoid the early receiver (identified by the LACK of bandwidth designations and a three bandwidth positions in CW MODE no matter what others might tell you!!!!

The AGC popping and overshooting is virtually unfixable, with filter group delay in narrow filter positions several mS) ruining the AGC leading edge response time. There is an AGC patch that slows the AGC to the RF amp, but it doesn't help fast AGC on CW with narrow filters.

The transmitter has -40dB AM tone at 1kHz on the CW waveform,because the sidetone amplifier and PA share the same cathode returns (the sidetone amp is also the AM modulator!!).

On the plus side, the filters are very good. The PTO has low noise. It can be reworked into an acceptable system with mods from Sherwood Engineering, but without major work (that amounts almost to rebuilding) it generally will not be as good as the off-the-shelf better modern radios available.

It is fun to drive, but not comfortable and definately not high performance!

73 Tom

I agree with Bill, but the problem is with ALL major SSB contests on 160 meters. What they need is restricted "no transmit" areas so people can hear DX, and a rule that forces people to move to new frequencies once in a while.