PDA

View Full Version : Which new or used amplifier design is representative of these design recommendations



J85K
04-12-2011, 03:11 PM
:confused:

I would appreciate input, on currently available new or used amplifier models, MOST representative of the AG6K's list of desirable inclusions in modern RF amplifier's. Here are his recommendations:-\



Adjustable filament voltage and a simplefilament voltmeter using the multimeter

Step-start relay circuit for the entireamplifier

Switch-adjustable electronic bias switch

High speed RF relays

Low VHF-Q parasitic suppressors

Separate HV and filament transformers -preferably potted; MOF equalizer resistors for electrolytic capacitors; noso-called equalizers on HV rectifiers

Compatible with 100W of drive

Reasonable cost of a replacement amplifiertube

Glitch protection diodes for the metercircuits and the negative HV circuit

A approx. 10-ohm, approx. 10W glass-coatedresistor in series with the positive lead of the HV power supply

Adjustable input capacitors on the tunedinputs


I would additionally ask if a suitable relay interface for modern transceiver keying should be part of the integration......

Thanks and regards, Kumar_J88CF

WB2WIK
04-12-2011, 03:30 PM
My old Henry 4K Ultra had most of those features. Don't know if I'd consider an 8877 to be a "reasonable cost" tube anymore, but 30 years ago it was.

They're hard to find as nobody who has one wants to sell it.

The decision to sell mine was based mostly on running out of space, otherwise I'd still have it.

KD0CAC
04-12-2011, 03:39 PM
Home brew .
If you want it done right do it yourself .
More a comment to keep a link to the thread .

W8JI
04-19-2011, 09:05 PM
Some of Measures ideas are OK, and some are nonsense. This includes his unusual infatuation with so-called low-Q nichrome suppressors. He also is out-of-touch with manufacturing costs compared to the benefits of certain things. For example:



xAdjustable filament voltage and a simplefilament voltmeter using the multimeterTotally unnecessary. This is NOT broadcast where a tube is operated at 20% of capacity and runs steady for 24 hours a day 365 days a year.
Virtually all tube failures in amateur service are related to thermal cycling and poor modern tube quality, not filament emission life. I can't recall seeing any tube failures in the past 30 years in Ham gear related to filament voltage, except one 8877 that had a customer reduce filament voltage to the point he ruined the cathode by stripping it!!!



x Step-start relay circuit for the entireamplifier
Totally unnecessary unless someone measures inrush as a problem. If a filament transformer is properly designed and sized, inrush is never a problem. In the HV circuit, inrush generally only wears on switches and relays. Circuits like that should be used only as necessary, not by demand of someone who has never designed or manufactured a single commercial product.


x Switch-adjustable electronic bias switch
I don't know why. Some have it as a feature, some do not like it.


x High speed RF relays
QUIET high speed relays like vacuum relays suitable for kilowatt power levels cost about $200 to 300 each wholesale. That would add about $500-800 to cost. Good luck on that for new construction.


x Low VHF-Q parasitic suppressors
His low Q suppressors in many cases actually have higher Q at VHF than standard suppressors do.


x Separate HV and filament transformers -preferably potted;
Good luck on that. I cannot remember any time I have seen a power transformer in amateur gear fail from moisture ingress. No one is going to double the cost and increase size 50% to fix a non-problem.


xMOF equalizer resistors for electrolytic capacitors
That makes no sense. While carbons should be avoided, MOF or suitable wirewounds or any metal based resistor is fine.



x; noso-called equalizers on HV rectifiers
I don't know of many people who use equalizers. They harm nothing if properly sized, but they generally help nothing.



x Compatible with 100W of drive
That goes without saying.


x Reasonable cost of a replacement amplifiertube
Good luck on that. Low cost means surplus or limited quality as a general rule.


x Glitch protection diodes for the metercircuits and the negative HV circuit
Most or many amps have that. Every Ameritron amp does.


x A approx. 10-ohm, approx. 10W glass-coatedresistor in series with the positive lead of the HV power supply
Glass coated resistors are not rated for HV use. The proper resistor is a high energy resistor that can take huge overloads and high voltages without internal arcing. The LAST resistor that I would use is a general glass coated resistor that is not rated for HV surge use.


x Adjustable input capacitors on the tunedinputs
Why? It would at least quadruple cost and is totally unnecessary. No one has done that since the 1980's.

Some of the best amplifiers made with the best service history have none of those requirements. I certainly would not pay a dime for *most* of those suggestions in something of my own.

For example I almost never step-start a filament, never include front panel filament monitoring and adjustment, never use a glass resistor as a fault limiting resistor, never use magical nichrome, and never use potted transformers. I use wire wound or metal film equalizing resistors, I clamp the negative to protect the meters. I use fixed mica caps with adjustable inductors in my homebrew amps tuned input. I use vacuum relays because I can buy them surplus for $30 each...but for commercial production used or surplus parts cannot be used.

73 Tom

AC0H
04-21-2011, 06:26 PM
Adjustable input capacitors on the tuned inputs

I don't know of a single modern commercial amp, or any of the home-brew recipe's you see on the net, that use adjustable capacitors instead of slug tuned inductors for the tuned inputs.


I would additionally ask if a suitable relay interface for modern transceiver keying should be part of the integration

Not an issue on modern commercial amps and I can't imagine a home-brewer designing anything else.

Why would you need to adjust the filament voltage unless you've got a really old tube that is starting to go soft and want to avoid replacing it for as long as possible?

CPI/Eimac says running lower filament voltage can extend the life of thoriate tungsten filament tubes like the 3-500Z(G) but they say running lower filament voltage on a tube like the 8877/3CX1500A7 which has an oxide cathode should never happen lest the cathode get stripped.

This is why the data sheet from Eimac says 5V +/- .25V for the 3-500Z and 5.0V (no voltage adjustment) for the 8877.

CPI bulletin #18 and the tube data sheets for the 3-500 and 8877 are the sources.

Running lower filament voltage will extend the life of thoriated tungsten filaments but the max reduction is 5%. Broadcasters will take the time and expense of doing filament management because that 5% reduction can extend the life of a tube or tubes who cost north of $10K to replace.

It's not worth the time an expense for Hams to do it and it's just one more way for folks to screw things up then whine to the manufacturer.

W5THT
05-08-2011, 08:29 PM
Your questions are all about the internal workings of an amp.
If someone has told you that this combination of features is optimum, perhaps the someone was speaking of the ideal amplifier for a technically-oriented ham.
For someone who has the intention of just using the amp, there is no need to have adjustments a grandkid can tweak that will cause a problem.
When you get the phone call that the DX station you need is on the air for a few hours and come into the room at a dead run, hitting the switches and hollering your call with the filament voltage wrong can get expensive.
If you live where line voltage varies, then you may need the adjustment, but also mental discipline to never transmit unless all the meters are right.
If you want to build the ultimate amp, there are many web sites where people show their "ultimate amps".
The ARRL and Editors & Engineers handbooks show well-designed amps which put out clean signals.
You can also buy the amp that matches your rig.
Either way, like the cartoon showing a puppy with computer "they can't tell you are a dog on the internet", they can't tell what amp you are using if you use it right.
Stay safe & Happy,
Pat W5THT

WB2WIK
05-08-2011, 08:36 PM
Why would you need to adjust the filament voltage unless you've got a really old tube that is starting to go soft and want to avoid replacing it for as long as possible?

Actually I can see some sense in this one, to accommodate different line voltages. The Henry 4K Ultra, which I owned for many years, had a panel accessible filament voltage control and meter to read it. Not such a bad idea, when you're using an expensive tube like the 8877.

Some of the other stuff is silly.

I have only one 8877 amp, a homebrew one for six meters, and although I don't have a variac on the filament transformer, I do use a resistance to adjust it, to accommodate varying line voltages. Here my line voltage is 242V almost always, so I adjusted for that. In some places it might be different. Using this amp in the field at a contest site years ago when we ran generators on a mountaintop the line was running over 250V, so I made a small tweak to bring it down.

Same tube in service since 1985, and it was used, then!:)

AC0H
05-09-2011, 06:29 PM
Actually I can see some sense in this one, to accommodate different line voltages. The Henry 4K Ultra, which I owned for many years, had a panel accessible filament voltage control and meter to read it. Not such a bad idea, when you're using an expensive tube like the 8877.

Some of the other stuff is silly.

I have only one 8877 amp, a homebrew one for six meters, and although I don't have a variac on the filament transformer, I do use a resistance to adjust it, to accommodate varying line voltages. Here my line voltage is 242V almost always, so I adjusted for that. In some places it might be different. Using this amp in the field at a contest site years ago when we ran generators on a mountaintop the line was running over 250V, so I made a small tweak to bring it down.

Same tube in service since 1985, and it was used, then!:)

I see what your saying but I was thinking about amp newby's cranking the filament up and roasting a tube.
It would be a support and warranter nightmare.
"Oh look.... if I turn this pot I get more power outta my amp" (for a while).

N0AZZ
05-10-2011, 12:52 PM
This shows a good reason to use a fully transistorized amp or one of the Alpha's Yaesu, Icom, Acom many others have worked this out quite well it seems. I do have a tube amp but it sits in the closet just for an emergency.

WB2WIK
05-10-2011, 03:23 PM
I do have a tube amp but it sits in the closet just for an emergency.

Like if you can't reach something on a high shelf in the closet, you can use it as a stepping stool?:)

W8JI
05-10-2011, 03:31 PM
I see what your saying but I was thinking about amp newby's cranking the filament up and roasting a tube.
It would be a support and warranter nightmare.
"Oh look.... if I turn this pot I get more power outta my amp" (for a while).

How it really works is like this......

You won't see any statistical difference in life of tubes in amateur service by getting all worried about filament voltage. None. Zip. Zero.

If you run the 8877 or 3CX800A7 somewhere between Eimac's minimum up to an extra 1/2 volt, you will never see a life history change.

The only design problem in amateur systems is lack of fast electronic protection of grids in amps like the 8877 or 3CX800A7, lack of arc fault limiting with big filter capacitor banks, and as unbelievable as it sounds some amps do not even have filament warm up timers with indirectly heated tubes!!! Not having a fast grid trip circuit is a sure way to shorten tube life.

It would be penny wise and dollar idiotic to include a variac while not having arc limiting and grid current overload, and a warm-up delay timer with a tube like the 8877 or 3CX800A7 or any heater-cathode power grid tube.

Nearly ALL failures in 8877 or 3CX800A7 or other low-intercept tubes, that are not normal long hours end-of-life failures, are caused by not having grid protection or running grossly excessive or even slightly low filament voltages. Tubes like the 8877 or 3CX800A7 are ruined in seconds by excessive grid current, operation before full warm-up, or operation with low filament voltage. The rest of the failures, other than inadequate cooling, are almost entirely from tube manufacturing defects.

Nearly all failures in thoriated tungsten tubes are caused by thermal cycling of the tube in well-manufactured tubes, although that now has been greatly eclipsed by manufacturing defects in tubes. Operating errors in thoriated tungsten tubes almost exclusively relate to over-dissipation of anodes. Some amps have inadequate cooling, which leads to seal failures, pin unsoldering, or glass failures. Other than amps with inadequate airflow causing glass or pin seal failures, the majority of failures now are manufacturing quality issues. Second to that is thermal cycling. Filament voltage as a general rule is not even on the statistical RADAR.

We kept statistics on this stuff for years. The only 8877 ever returned for low emission had a filament mod to reduce voltage to 4.5 volts. The low voltage ate the tube up in about 3 months.

For Ham use, anything right around the specified range is just perfect. You will never see a life change by getting all anal about filament voltage, but you sure will if you do things like not have a grid trip circuit in a metal oxide cathode tube or run the filament voltage low, or transmit before the cathode is at full temperature.

Filament voltage is important in commercial BC transmitters. This is because the tubes are not cycled off and on, and they are not heated and cooled during operation while on. They are also operated at a fraction of ratings. This means most BC tube failures are emission life failures. This has never been the case in amateur service, except a few Dentron designs that ran grossly excessive voltage on filaments or amps that people have modified.

73 Tom

ad: dxeng