Print

[HAMHOCK75]

The manual is pretty insistent on ALC usage.  My K3 is an older model and does not have ALC.

It sounds like you are new to amplifiers. You do not need to use ALC at all. The only connection needed is the "send" line from the rig to turn on the amplifier.

Without ALC, the output from the rig needs to be adjusted every time you change bands because the amplifier gain is not constant as shown below.



This creates a potential problem. For example, if you are operating on 10M where it takes 50 watts in for 500 watts out, then change to 20M where it takes 25 watts in for 500 watts out, the amplifier will be over driven by the 50 watts unless you remember to reduce it to 25. The 2KL has built-in protection for this. It will shut down if the output power goes over 600 watts.

I appreciate that this may be too technical for you but for those who are interested, the analysis of the circuit that limits output power is at this link.

https://www.eham.net/community/smf/index.php/topic,133714.msg1243506.html#msg1243506

Even though the amplifier is protected from over drive, over drive cannot happen with ALC which is why Icom wants you to use it. The ALC voltage adjusts the rigs output for you when you change bands so the output power is maintained even though the gain of the amplifier is changing.

It all comes down to how comfortable you feel about remembering to reset the drive power to the amplifier when changing bands. I recall that VK6HP mentioned that some rigs allow you to program the output power from the rig depending on band. If the Elecraft can do that, then that is another reason ALC is not needed.

P.S. The 40 dB attenuator was between the Icom 7300 and digital scope to act as a 50 ohm load and lower the output power to the scope. It was not used with the amplifier.

[K6BRN]

I've used a variety of radios and amps over the years including tube/analog, SS/analog, analog/digtal hybrid and SDR, including mixing tube and SS amp/transceivers.  I've never damaged an amplifier by initial key down overshoot, though I've seen TX power overshoot to one degree or another in just about every case.

And I have usually used ALC, but adjusted it by various means (sometime HW changes) to avoid kicking in unless the amp was well into overdrive.

Earlier SS amps were more prone to damage from many causes, so perhaps some were vulnerable to this.  My Yaesu Quadra was not, but did have a pretty poor, slow-acting (CPU driven) output protection circuit.  My SB-200 tube amp didn't care, either - it was most vulnerable to duty cycle and was really best used for SSB.

Some ceramic tube amps were also suseptible to overdrive damage - but this usually had to be sustained, not a brief blip.

Currently I've used my own or friends Yaesu FTDX-1200, -3000, -991 and Icom IC-7300 with KPA-500 amps.  They ALL seem to have initial key-down overshoot, per my attached LP-700 wattmeter's TX 'scope trace.  The KPA-500 does not seem to care at all, either.  But I rarely push it beyond its rated 500W.

In my experience, momentary overdrive peaks on TX are not usually a problem for a well designed HF amplifier, as previously pointed out.  I suppose its always possible to make it a problem by using no ALC at all and overdriving an older SS amp or ceramic tube amp without fast input/output protection.  But there is always a way to damage equipment if one is careless or deliberately pushing it too hard.  No stopping that.

Brian - K6BRN

[AC2RY]

This creates a potential problem. For example, if you are operating on 10M where it takes 50 watts in for 500 watts out, then change to 20M where it takes 25 watts in for 500 watts out, the amplifier will be over driven by the 50 watts unless you remember to reduce it to 25. The 2KL has built-in protection for this. It will shut down if the output power goes over 600 watts.

The gain difference is usually not that severe. You like need to adjust output within few percents. So if output is set to get 600W on bands with highest gain, on the band with lowest gain you get 500W. If that difference is tolerated, then no adjustment is needed. Some modern transceivers have output power setting separate for each band.

[HAMHOCK75]

The gain difference is usually not that severe.

Unfortunately, the 2KL is a 40 year old design. The gain variation is a bit greater than 2 to 1 over all bands. The devices are very well protected however. Mine still has all the original transistors. The 2KL suffers mostly from aging electrolytic and dirty relay issues.

My SB-200 tube amp didn't care, either - it was most vulnerable to duty cycle and was really best used for SSB.

I still have mine. It is relegated to SSB use with my vintage Yaesu FLDX400/FRDX400 twins station.

What caught my eye about the 2KL was even though it is old, its specifications allow for a 10 minute key down time at full output in CW, RTTY modes. Similar to the KPA500 specification except Elecraft specifies 10 minutes key down followed by 5 minutes in standby.

[KH6AQ]

One solution to overshoot would be to insert an attenuator between the transceiver and the amp so that the transceiver must be set close to 100W to drive the amp to its specified output power. A 100W attenuator having settings of 2, 3, 4 and 5 dB should take care of many situations and this could be something for MFJ to produce. A 5 dB hit on receiver noise figure I'd think would be acceptable on HF.

[VE7RF]

One solution to overshoot would be to insert an attenuator between the transceiver and the amp so that the transceiver must be set close to 100W to drive the amp to its specified output power. A 100W attenuator having settings of 2, 3, 4 and 5 dB should take care of many situations and this could be something for MFJ to produce. A 5 dB hit on receiver noise figure I'd think would be acceptable on HF.

You can already buy 1-2-3-4-5-6-10-20 db attenuators from henry radio, in both ( I think) 35-150 watt range.  Ideally, the padding should be installed inside the amp, then no loss on RX.   Or perhaps the padding could be external to the amp, but cabled to the inside of the amp. A few db loss on RX is usually a non issue on lower bands, but could be an issue on 20-10m.

[AC2RY]

One solution to overshoot would be to insert an attenuator between the transceiver and the amp so that the transceiver must be set close to 100W to drive the amp to its specified output power. A 100W attenuator having settings of 2, 3, 4 and 5 dB should take care of many situations and this could be something for MFJ to produce. A 5 dB hit on receiver noise figure I'd think would be acceptable on HF.

This is not good at all. Best distortion values in transceiver with AB1 output stage is usually between 25 and 33 % of rated power. Though some show constant improvement when turning power output all way down. So the problem should be addressed at source rather than by crutches. All LDMOS amplifiers have attenuator at the input to reduce influence of high input capacitance on exciter. This is sufficient to protect from accidental overshoot to full 100W power along with automated protection circuit which must be there by design.

If your old analogue transceiver causes excessive overshoot - it is time to retire it and move to modern SDR type, which likely won't have significant problem in that regard.
 

[K1KIM]

One solution to overshoot would be to insert an attenuator between the transceiver and the amp so that the transceiver must be set close to 100W to drive the amp to its specified output power. A 100W attenuator having settings of 2, 3, 4 and 5 dB should take care of many situations and this could be something for MFJ to produce. A 5 dB hit on receiver noise figure I'd think would be acceptable on HF.

This is not good at all. Best distortion values in transceiver with AB1 output stage is usually between 25 and 33 % of rated power. Though some show constant improvement when turning power output all way down. So the problem should be addressed at source rather than by crutches. All LDMOS amplifiers have attenuator at the input to reduce influence of high input capacitance on exciter. This is sufficient to protect from accidental overshoot to full 100W power along with automated protection circuit which must be there by design.

If your old analogue transceiver causes excessive overshoot - it is time to retire it and move to modern SDR type, which likely won't have significant problem in that regard.
 

Not so. Many of the newer SDR have overshoot. There are volumes of discussions on the IC-7300 alone.
I only pair my old boat anchors with a valve amp that needs 100W drive to be happy. No overshoot in them then.

[VR2AX]

Can't one just turn down the DC voltage on the DC supply, most transceivers are specified as "13.7" or something volts, goodness knows what the designer used. Using +/- 10% or 15%, you will experience some reduction. More TX IMD, who knows, but that is not the point of this thread...

[WB2VVV]

I believe this whole "worry" propaganda about transceiver overshoot has been put forth at least partially for ulterior motives (Icom haters for one, and blame-placers for another). Meanwhile, I can't even imagine who would even desire a 1kW output amplifier driven by a Watt or two from a 100W transceiver, as that would just be a recipe for disaster for multiple reasons.

I built my 1kW output LDMOS RF deck using a single LDMOS soldered to a 1/2 inch thick oversized Copper heat spreader that is in turn bolted to an oversized Aluminum heat sink with carefully designed forced air cooling of the heat sink and Low PassFilters. I also bolted two -6dB microstrip flanged attenuators to the heat sink away from the LDMOS and the Copper heat spreader. The first one in the drive path is rated for 250W and the second one is rated for 100W, so that's -12dB of well designed and executed attenuation placed between the 100W capable transceiver and the LDMOS. I am using modern Icoms that have been blasphemed with this overshoot propaganda with my LDMOS RF deck and have no troubles at all. I do generally have to drive the amplifier with 50 - 80 Watts from the transceiver because of the attenuation but the transceiver(s) aren't complaining and the S11 is excellent due to the pads.

My understanding is that many that have been blaming transceiver overshoot when their problem was really poor design implementation, little on-board attenuation, and the root cause of the LDMOS failure was more typically overheating the junction due to poor thermal design implementation, and not transceiver overshoot at all. Again I am seeing no trouble with the very same blasphemed modern Icom transceivers...

[G4AON]

This whole ALC and overshoot issue is colored by guesswork and speculation. Let me explain...

Regarding ALC, manufacturer A makes a transceiver, manufacturer B makes a linear. Along comes the user and connects the ALC (feedback loop) between the transceiver A and the linear B. The user and everyone else expects that feedback loop to work seamlessly, even though the transceiver manufacturer had no idea of the level of feedback when designing the transceiver. You can see the potential problem. ALC dates from the last century when you had 100 Watt tube transmitters feeding tube amplifiers. Generally there was no easy way to adjust the drive level on those old radios, except with the ALC.

Modern transceivers should be able to control power output without resorting to feedback from an amplifier.

Regarding power overshoot. It is easy to check with an oscilloscope, why so few amateurs have one is bizarre. My own measurements with an IC-7300 show no overshoot on 160 to 6 metres at any power level. However, on our 4m band, and only on SSB, mine and some others do suffer from full power peaks when set for around 10 Watts, this happens at random times during transmissions and will trip an LDMOS amp. I made a plug-in negative Voltage generator to provide a fixed ALC input to cure the problem. It is only used on 4m with my Linear Amp Gemini 4 amplifier.

The IC-7300 and IC-9700 both have an RF "tail" of around 3 mS after PTT release. I have measured this and so have the ARRL. The linear key line drops at the same instant that PTT is released, which may cause hot switching of an amp. That is a design flaw by Icom. Some amplifiers have a PTT hold on menu item, or at least the Elecraft KPA500 has one.

I used to have an early K3 and did modify it to take negative ALC, but never used ALC. Incidentally, the K3 did not have overshoot.

A PTT hold on circuit (not described as for an amplifier, but it should work with one) and the IC-7300 ALC fix are described on my web site at:
https://www.qsl.net/g4aon/radio_mods/

73 Dave

[VE7RF]

This whole ALC and overshoot issue is colored by guesswork and speculation. Let me explain...

Regarding ALC, manufacturer A makes a transceiver, manufacturer B makes a linear. Along comes the user and connects the ALC (feedback loop) between the transceiver A and the linear B. The user and everyone else expects that feedback loop to work seamlessly, even though the transceiver manufacturer had no idea of the level of feedback when designing the transceiver. You can see the potential problem. ALC dates from the last century when you had 100 Watt tube transmitters feeding tube amplifiers. Generally there was no easy way to adjust the drive level on those old radios, except with the ALC.

Modern transceivers should be able to control power output without resorting to feedback from an amplifier.

Regarding power overshoot. It is easy to check with an oscilloscope, why so few amateurs have one is bizarre. My own measurements with an IC-7300 show no overshoot on 160 to 6 metres at any power level. However, on our 4m band, and only on SSB, mine and some others do suffer from full power peaks when set for around 10 Watts, this happens at random times during transmissions and will trip an LDMOS amp. I made a plug-in negative Voltage generator to provide a fixed ALC input to cure the problem. It is only used on 4m with my Linear Amp Gemini 4 amplifier.

The IC-7300 and IC-9700 both have an RF "tail" of around 3 mS after PTT release. I have measured this and so have the ARRL. The linear key line drops at the same instant that PTT is released, which may cause hot switching of an amp. That is a design flaw by Icom. Some amplifiers have a PTT hold on menu item, or at least the Elecraft KPA500 has one.

I used to have an early K3 and did modify it to take negative ALC, but never used ALC. Incidentally, the K3 did not have overshoot.

A PTT hold on circuit (not described as for an amplifier, but it should work with one) and the IC-7300 ALC fix are described on my web site at:
https://www.qsl.net/g4aon/radio_mods/

73 Dave

  That 3 msec  RF  'tail' is a NON issue if using  VOX  SSB / CW...with their typ 300-500 msec VOX delay.  It's only when using QSK  CW, that it will become an issue.

And if using QSK CW,  a simple RF detector on the input of the linear amp will  inhibit releasing the TR relays in the amp.  That RF detector used on the W7RY input board on his QSK kit, costs a whopping 82 cents...and consist of just 4 x components.   Alpha does the same.  IE: if say running a 1 to 1.5 KW  CXR.... reach over and pull out the key line at either the xcvr, or the amplifier, and the amp will  NOT unkey the TR relays.  It will continue to put out 1 kw.  No rocket science here folks.

[VR2AX]

However, on our 4m band, and only on SSB, mine and some others do suffer from full power peaks when set for around 10 Watts, this happens at random times during transmissions and will trip an LDMOS amp. I made a plug-in negative Voltage generator to provide a fixed ALC input to cure the problem. It is only used on 4m with my Linear Amp Gemini 4 amplifier.

Good info and website. Wonder why this happens in the first place.

[W1VT]

Oscilloscopes used to be big heavy beasts that cost as much as an HF radio.
A few years ago they were down $400 for one with good reviews.  Now there are some on Amazon for just $200!

Zak W1VT

[KH6AQ]

One solution to overshoot would be to insert an attenuator between the transceiver and the amp so that the transceiver must be set close to 100W to drive the amp to its specified output power. A 100W attenuator having settings of 2, 3, 4 and 5 dB should take care of many situations and this could be something for MFJ to produce. A 5 dB hit on receiver noise figure I'd think would be acceptable on HF.

This is not good at all. Best distortion values in transceiver with AB1 output stage is usually between 25 and 33 % of rated power. Though some show constant improvement when turning power output all way down. So the problem should be addressed at source rather than by crutches. All LDMOS amplifiers have attenuator at the input to reduce influence of high input capacitance on exciter. This is sufficient to protect from accidental overshoot to full 100W power along with automated protection circuit which must be there by design.

If your old analogue transceiver causes excessive overshoot - it is time to retire it and move to modern SDR type, which likely won't have significant problem in that regard.
 

The attenuator fixes the overshoot issue. Please provide links to amateur transceiver TX IMD measurements at different output power settings.