SPE/Flex 6xxx drive control

  • 1
  • Problem
  • Updated 3 months ago
The LDMOS failed in my SPE1.5K. There were numerous unexplained overdrive errors in the SPE log. The Flex drive power is set to 10W on all bands. I notified Bob H of the errors, prior to the total failure, but he did not have an explanation. My question...could this be a fault with the Flex6600M, there is no ALC connection the drive is controlled by the SPE firmware.
Photo of Rick Ciotti W3DIY

Rick Ciotti W3DIY

  • 39 Posts
  • 5 Reply Likes

Posted 4 months ago

  • 1
Photo of Pat N6PAT

Pat N6PAT

  • 1002 Posts
  • 301 Reply Likes
I recently posted an issue with the power levels on my 6700 jumping higher than the level I had set it to.

This was not an issue with the SSDR power slider but rather the actual output power as indicated by the slider as well as 2 LDG watt meters. I had to immediately shut down my amp as it only requires 2 watts for full power.

I will be interested in knowing how your issue turns out. Good luck
Photo of Jon - KF2E

Jon - KF2E

  • 707 Posts
  • 237 Reply Likes
Given the way profiles currently work, I would recommend setting the max transmit power on the transmit tab to however much your amp can take. Then you should never exceed that power.

Jon...kf2e
Photo of Rick Ciotti W3DIY

Rick Ciotti W3DIY

  • 39 Posts
  • 5 Reply Likes
Thanks! The 6600M profiles have always been set to 10W on all bands. At this drive level the SPE output is about 1KW.  BUT the SPE overrides the programmed values. As I change the SPE drive control the 6600M Tx profile changes.
Photo of Jon - KF2E

Jon - KF2E

  • 707 Posts
  • 237 Reply Likes
Rick, I wasn't referring to the setting in Profiles. There is a setting on the TX Tab of Radio setup that has Max Power. It will(supposedly) not allow the radio to transmit above that setting. It will also restrict third party apps from trying to transmit at a higher setting.I set mine to 35 watts to protect my Yaesu Quadra.

73, Jon...kf2e
Photo of Clay N9IO

Clay N9IO

  • 704 Posts
  • 198 Reply Likes
Jon KF2E,
Thats how I have my 6600 set is at 30 watts max to drive my 2KFA. No issues quite flawless. In the TX of the menu.
Photo of Rick Ciotti W3DIY

Rick Ciotti W3DIY

  • 39 Posts
  • 5 Reply Likes
Occasionally run barefoot so the max power was 100W. Until I get to the root cause I will set the max power to 10W.
Photo of Clay N9IO

Clay N9IO

  • 704 Posts
  • 198 Reply Likes
Same here like the NAQPs I leave the amp off and set max TX power back to 100. That setting you can rely on the profiles I have found you cannot. No idea why I only know I cannot trust the profiles.
Photo of Rick Ciotti W3DIY

Rick Ciotti W3DIY

  • 39 Posts
  • 5 Reply Likes
Set the Flex to 12 watts max power on the Transmit General screen. Several more overdrive events were recorded, while Txing on 40M.
Photo of Pat N6PAT

Pat N6PAT

  • 1002 Posts
  • 301 Reply Likes
When that happens do you notice any quivering or motion on the SSDR power output meter?
(Edited)
Photo of Rick Ciotti W3DIY

Rick Ciotti W3DIY

  • 39 Posts
  • 5 Reply Likes
Operating with the 6600M as a standalone, no SSDR.
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 4649 Posts
  • 1098 Reply Likes
The SPE has no working drive protection?
Photo of Rick Ciotti W3DIY

Rick Ciotti W3DIY

  • 39 Posts
  • 5 Reply Likes
It goes into standby mode but can still take out a LDMOS.
(Edited)
Photo of Rich

Rich

  • 11 Posts
  • 1 Reply Like
I had the same thing happen with my KPA1500 back in Nov.
Elecraft support had told me there were issues with Flex 6k series with power spikes (similar to Icoms problems) that they referenced. They repaired the KPA under warranty,

This is running under 2.49. There were also various posts about the spikes too.
So far, it has not happened since.
Photo of Michael Walker

Michael Walker, Employee

  • 1362 Posts
  • 401 Reply Likes
Just about all current amps have overdrive protection and then shutdown the amp.  I would find it hard to think an overdrive issue caused this.

Since I am involved in another amp (non-FRS) project that has had its share of failures with no explanation and I am starting to believe that the BLF189's (in this case) just blow up on their own.  They are not impervious to failure even though we wish they are.  We have seen the LMDOS supplier actually replace the LMDSO's at no cost.

In the case of SPE, they should reach out to their LMDOS supplier and talk to them about it.

Mike 
Photo of Mike N9NTC

Mike N9NTC

  • 4 Posts
  • 2 Reply Likes
Are you using the Interfacing cable from Dam Tassell KC5PCB. If you are set the power for each band from the SPE Amp. the Flex will not exceed the input power usually 5 to 10 watts. I use the SPE 1.3 and works well and can't accidently overdrive the amp.
Photo of Rick Ciotti W3DIY

Rick Ciotti W3DIY

  • 39 Posts
  • 5 Reply Likes
Yes I am using KC5PCB's cables.
 The SPE's drive power has always been set to 10W on all bands. No need for maximum output with the present band conditions. I assumed it would be impossible to overdrive the amp with these settings yet there were 9 overdrive events in the log. 
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 4649 Posts
  • 1098 Reply Likes
It seems that the over drive is not from the Flex then as you have everything set correctly. I have also read in other forums people having the same problems using other types of radios as well. It seems to be SPE 1.5K issue.
Photo of Jay / NO5J

Jay / NO5J

  • 1685 Posts
  • 266 Reply Likes
The numerical scale of the RF power slider isn't using watts for it's units, it reads in percentage of Max output power. and an SWR mismatch could falsely raise your Max Power output.

100% of 100 watts produces 100 watts.
100% of 180 watts produces 180 watts.
If your driving your rig into the red on the mic level scale, it could be even more.

     SDRgadgets

#FlexRadio IRC chat

   73, Jay - NO5J

Photo of Pat N6PAT

Pat N6PAT

  • 1002 Posts
  • 301 Reply Likes
The max power out for Flex radios is 100 so 1% = 1 watt, 2% = 2 watts ,etc.
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 4649 Posts
  • 1098 Reply Likes
I believe Pat is right. 50% on the slider should be 50w.
Photo of Dan Quigley N7HQ, 4O7HQ

Dan Quigley N7HQ, 4O7HQ, Service/Support Manager

  • 105 Posts
  • 76 Reply Likes
Not all the time.  We calibrate the PAs of the 6400/6600 series radios to produce 100W at 45C.  If the PA is cold, the PA can (and usually does) emit more than 100W when the power slider is at 100%.  You can see this in action easily by using an external wattmeter or observing the RF power output meter in SmartSDR.  The correct way to think of the power slider is relative to that calibration. It is not a precision power setting. 
Photo of Pat N6PAT

Pat N6PAT

  • 1002 Posts
  • 301 Reply Likes
My amp provides maximum rated power on 160 with just 2 watts while 3 watts generates more output than the amp is rated for. That 1 watt makes a big difference.

How can I be certain that my 6700 will consistently provide 2 watts to my amp? I don't care what the slider says. I'm talking about 2 actual stable watts coming out of the radio.
Photo of Dan Quigley N7HQ, 4O7HQ

Dan Quigley N7HQ, 4O7HQ, Service/Support Manager

  • 105 Posts
  • 76 Reply Likes
Pat,  the low input power requirement of your amplifier complicates the issue you describe. There is no SmartSDR setting that assures two "stable" watts directly from your 6700.

The least complicated solution is to employ a 50-ohm attenuator between your radio and the amp.  The attenuator would need to be able to handle 100W and be PTT aware (switch in for transmitting and out for receive). Ideally, you would calculate the attenuation value to provide your amplifier the drive it needs for full output power when the SmartSDR level control is at 100.  This would allow small excursions of output power from the 6700 to be far less impactful to the output of your amp.  Bird sells such attenuators, but you would still need to add an outboard TR switch. 

The second (and more complicated) solution is adding small gain stage to amplify transverter level outputs of your 6700 to a level that matches the input power requirement of the amplifier.  It too would need to be PTT aware and you would need to select the transverter output for TX in a SmartSDR slice. 

Best,
Dan
Photo of Pat N6PAT

Pat N6PAT

  • 1002 Posts
  • 301 Reply Likes
Let me see if I understand you correctly.

Are you saying that there is no way for a user to specify and achieve an accurate and consistent power output from the Flex 6700? Is that correct?
Photo of Dan Quigley N7HQ, 4O7HQ

Dan Quigley N7HQ, 4O7HQ, Service/Support Manager

  • 105 Posts
  • 76 Reply Likes
That is correct. 
Photo of Pat N6PAT

Pat N6PAT

  • 1002 Posts
  • 301 Reply Likes
Well then, that might be the cause of my amp's blown finals. I had to pay over $200 for repairs + shipping and wait 3 months to get that amp fixed at DX Engineering.

Now I can no longer depend on any power settings in SSDR to be accurate. That's just great! I see more blown finals in my future.
Photo of Geoff - AB6BT

Geoff - AB6BT

  • 237 Posts
  • 55 Reply Likes
Add an attenuator as Dan suggests and you'll be good to go. There will be much less variation, percentagewise,  at the output of the attenuator. For example, a 25:1 divider will give you about 4 watts when the slider is set to full scale and the 2 watts you need will be easy to achieve at about mid-scale.
Photo of Chris DL5NAM

Chris DL5NAM

  • 737 Posts
  • 163 Reply Likes
That's right. And some amp's use this solution; using a ATT with 13dB, can change to 16dB or if a QRP TRX is used to any lower value like 3 or 10dB
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 4649 Posts
  • 1098 Reply Likes
That's the down side of owning an amp that can't handle even a 1 watt change before blowing up. Bad design.
Photo of Geoff - AB6BT

Geoff - AB6BT

  • 237 Posts
  • 55 Reply Likes
Agreed.

Perhaps it's a good match for a QRP radio but I can't imagine it being safe to use with a 100 watt output transmitter. Maintaining an output at 1% to 2% of maximum output seems an engineering challenge.
(Edited)
Photo of Rick Ciotti W3DIY

Rick Ciotti W3DIY

  • 39 Posts
  • 5 Reply Likes
Update...I was informed by the US SPE dealer, Bob Hardie W5UQ, that this is a known glitch with the Flex software. Unfortunately I was unaware of this when I made the SPE purchase. The workaround is resetting the 6600M Tx delay. Anyone know what that value should be?
(Edited)
Photo of Michael Walker

Michael Walker, Employee

  • 1362 Posts
  • 401 Reply Likes
Rick, can you explain more on what the Glitch is that is a Flex issue?  Have we acknowledged that?  Any details you have will be appreciated so we can make sure the details are correct.  
Photo of Rick Ciotti W3DIY

Rick Ciotti W3DIY

  • 39 Posts
  • 5 Reply Likes
Michael, it was explained that the Flex drive power can spike unexpectedly, causing an overdrive condition, and was informed that the SPE will go into standby when this occurs. I realize the sequence of events that trigger the spike would be helpful, but I'm reluctant to risk taking out another LDMOS to reproduce this condition.
(Edited)
Photo of Alan

Alan

  • 207 Posts
  • 28 Reply Likes
I am not an expert, or even a novice when it comes to RF engineering.  However, I have 40 years of experience in converting volts and amps to thermal power and dissipating the thermal power to the environment.

So if there is an overshoot from the Flex that is harming the amplifier's transistors, is it thermal overload or something else, like high voltage that breaks down the transistor junctions?

If it is thermal, it takes amps, volts AND time to produce and dissipate thermal energy, resulting in an increased transfer of heat to the environment, thus, a higher delta T to drive the watts to the heat sink. (temperature).  A "spike" doesn't seem like a large unit of time to drive such a thermal outcome?  Do any transistor experts out there care to comment?

However, there could be a harmful amount of voltage (not time-dependent), but other factors seem to be at play for that to occur, such as feedline and antenna characteristics.  (outside my wheelhouse of knowledge, but know it is out there).


Just asking.

Alan
WA9WUD
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 4649 Posts
  • 1098 Reply Likes
Alan, I think in theory your correct. In the case of a Flex, they have been well known for rock solid TX power control. But there have been rare cases when people have had over power problems, but for most, it has not been a problem.

In the case of over shooting people claim to see on their SPE 1.5 amps it does not make any sense at all that it is software related. If it is then all Flex radios would be overshooting but they are not. Mine does not.

If it were my amp, I would go after the company and ask why the amp allowed damage from over power if their is any.
Photo of Alan

Alan

  • 207 Posts
  • 28 Reply Likes
Agree.  I have followed the issue of the Flex power setting exceeding the recommended excitation level for the amp. (due to operator error or not paying attention to changes the Flex makes to your TX profile).

Under the power setting scenario, it seems reasonable that the thermal limits of the transistors of the amplifier could be exceeded with typical transmission lengths (especially FT8), but then again, with these modern, microcomputer-controlled amps, it seems like the amp would protect the power transistors before failure.

Not buying the "spike" scenario when it comes to a thermal overload of the amps power transistor unless it is a high voltage induced breakdown of the transistor junctions.  I would not blame the Flex for high voltages on the feedline.

Alan
Photo of Alan

Alan

  • 207 Posts
  • 28 Reply Likes
I put some math behind my thoughts to bracket in the idea:

The manual for my Palstar amp says it can detect the incoming RF, determine the frequency, then close the output relay, then the input relay: all in 20.5 milliseconds.  Let's assume this is typical of the modern amp's time to detect an overexcitation and shut itself down.  I will use 25 milliseconds (25E-3) in my math.   Even if the amp does not shut down automatically, 25 milliseconds seems a reasonable period for a "spike" given that the middle of the audio spectrum is 2KHZ (presumably the source of the SSB overexcitation), the period of the over-excitation (2KHZ) is 0.5 milliseconds, much less then the conservative value used in my math.

Next, I will select a high overpower condition that we can all agree on as being the highest possible.  I will use 5,000 watts (5E+3).

Now, convert the thermal energy equivalent of  5E+3 watts for 25E-3 seconds to continuous watts for one minute; gives 2-watts thermal equivalent.

Alan
WA9WUD

PS. I have too much time on my hands today.
Photo of Rick Ciotti W3DIY

Rick Ciotti W3DIY

  • 39 Posts
  • 5 Reply Likes
The SPE just flagged an overdrive alarm on 40M. The 6600M max Tx power was set to 20W, the SPE drive power set to 15W. With the SPE ATU active the SWR was reading 1:1,the Flex SWR is hardly noticeable. BUT on this freq. the antenna SWR is about 3:1. I have changed the 6600M to 12W max and the SPE to 10W. More tests ahead.
(Edited)
Photo of Alan

Alan

  • 207 Posts
  • 28 Reply Likes
I did not purchase the SPE, but read the manual as part of my decision prior to making a decision.

As I recall, in the manual, it stated that operation with feedline SWR greater than 3.0, before tuning, required a de-rate.....not for thermal reasons, but on reasons for the potential of damaging high voltages within the internal tuner.

Or....am I mixing this up with the Elecraft Amp....my memory is not so good.  It's in one or the other, but the concept is the same.

I do not have a resonate antenna, so this was a non-starter for me.

Alan
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 4649 Posts
  • 1098 Reply Likes
Glad I didn't buy an SPE amp. looking back at all the comments over several weeks most problems people are having are with the 1.5 amp. Very few if any are having this problem with other amps.

Because of this discussion I'm sure Flex is running test as we write here.
Photo of Rick Ciotti W3DIY

Rick Ciotti W3DIY

  • 39 Posts
  • 5 Reply Likes
From the SPE manual...
With the internal ATU, the amplifier is able to overcome mismatches up to 5:1
Photo of Rick Ciotti W3DIY

Rick Ciotti W3DIY

  • 39 Posts
  • 5 Reply Likes
According to the SPE dealer there are many Flex owners with this amp who have no issues. The reason I posted here was to locate a few of those happy owners.
Photo of Chris DL5NAM

Chris DL5NAM

  • 737 Posts
  • 163 Reply Likes
upssss:


Photo of Rick Ciotti W3DIY

Rick Ciotti W3DIY

  • 39 Posts
  • 5 Reply Likes
On the 40M - 160M bands I operate with non resonant long wire antennas. On the primary operating frequencies most times the SWR is below 3:1. Will creep above that with wind and ice.
Photo of Rick Ciotti W3DIY

Rick Ciotti W3DIY

  • 39 Posts
  • 5 Reply Likes
"With the internal ATU, the amplifier is able to overcome mismatches up to 5:1"
 is from the Rev. 2.0 Second series manual.
Photo of Alan

Alan

  • 207 Posts
  • 28 Reply Likes
OK.  No mention of de-rates with the higher feedline SWR?  Then it must have been the Elecraft that had the de-rate requirement if subjecting the internal tuner to a high feedline SWR.

Imagine putting your SWR meter between your internal tuner and the antenna feedline, NOT the usual placement, between the internal tuner and the amp (SWR shown on your SPE Panel).   The restriction I spoke of is for the SWR the tuner sees, not what the amp sees.

In my case and antennas, the tuner can see an SWR greater than 5, so I selected a Palstar HF external tuner, instead of the internal tuners available with the SPE and Elecraft amps.  I was concerned about derates and high voltages with my particular antenna situation.  Others will have a different use case.

The Palstar Tuner is built like a "tank".....but slow...!  If you need to change bands quickly, the Palstar Tuner is not a good choice.

My amp never sees an SWR greater than 1.25:1.

Alan
WA9WUD
Photo of Clay N9IO

Clay N9IO

  • 704 Posts
  • 198 Reply Likes
The 2KFA built-in tuner will clean up any reasonable resonant antenna be it dipole , triband yagi or vertical. All of my antennas tune across the band.
One exception is of course 80m. 80 is separated between two, HF2V 3.5 - 3.65990 and a double bazooka I made about 26 years ago covers from 3.600 - 4.0mhz.
Once set up it is smooth and instantly tuned to the preset. I am feeding freq data via USB from the Flex with a FTDI chipset converter cable to a transition cable I made up (DB9 to DB15).

I might add that I have not experienced any of the powe spike anomalies being mentioned. Got to be something else going on I am thinking.

My Acom 1000 is the quickest and smoothest tuning manual tune amp I have ever owned but the 2KFA is REALLY nice. Instant on is great.
Photo of Rick Ciotti W3DIY

Rick Ciotti W3DIY

  • 39 Posts
  • 5 Reply Likes
No further info from SPE about overdrive errors with SWR at the high end of their spec. I have acquired a Palstar HF Auto from a fellow ham for test purposes. Disabling the SPE AT the Palstar matches my 40 thru 160 antennas FB. The time to search for a tuning solution is about the same. With this configuration so far no overdrive errors.
Photo of Andy - KU7T

Andy - KU7T

  • 277 Posts
  • 26 Reply Likes
Anyone know what value can protect from overshoots if this really is the issue? 50 ms enough?

I have blown 4 LDMOS and never found what caused it, so want to be very conservative...

I am driving an 6600M.

Andy
KU7T
Photo of Doug

Doug

  • 212 Posts
  • 70 Reply Likes
You will find that a Flex radio in the 6xxx series will run more PEP power than you set. Set it for a 100w and press tune and it will run right at 100w. I guess it must be the modulation setup that Flex talks about I can't recall the formal name for it. But it does provide big time output in SSB. I have a new LP700 meter (cost $1200) with two sensors and both will read up to 135w barefoot output PEP and drive the pants off my 3-500z amp.  Hook up a meter such as this and I think you will be surprised how much you are actually getting PEP even when the radio is set to 10w. I also believe the Flex does have an over shoot issue.  BTW my radio is the 6400M
Photo of Lasse Moell

Lasse Moell

  • 185 Posts
  • 32 Reply Likes
I had a look at the output waveform of my 6500, and there were absolutely no evidence of any overshoot. Very very nice envelope. Still I lost a BLF188 for no reason. Now it can be a lot of thins causing this, but one have to remember the gates are very fagile and the protection has to act in nanoseconds rather than milliseconds. One half period of RF may be enough to kill the transistor.
Photo of Mark Smith - KB5KYX

Mark Smith - KB5KYX

  • 18 Posts
  • 3 Reply Likes
My expert 1.5k has blown 2 ldmos chips. Both replaced under warranty. I have 2 other amps that I run with my flex 6600 besides the expert Amp. No problems with those amps blowing. I have spoken with other hams that have blown their expert amps and they aren't running flex radio's, so not a flex radio problem. I put my expert amplifier on the scope and it was not linear, so that's a major issue. If you own an expert amplifier, I would suggest checking your amplifier and not running it until it is aligned properly. On air reports with the expert amplifier were terrible. Switched to the ACOM and unsolicited reports were great. I too have numerous overdrive warnings on the expert amp, but its never been over 1500 watts. According to my lp700 it won't do 1500w, but that's another story. My expert amplifier is back in the shop for the 3rd time.....probably in Italy this time. Not happy at all with expert amplifier. Not willing to sell it either because I wouldn't feel right passing on this amplifier to another ham.....Not my style. Expert needs to make this right and refrain from putting the blame on other radio manufacturers and operators! Even if my expert 1.5k amplifier could run over 1500 watts on a real power meter (which it can't), the software is suppose to catch this and shut off without damaging the ldmos chip. Additionally the chip is rated to 1.8k, so barring an 1800w plus key down the chips should not be blowing. Like I stated earlier my lp700 station monitor has never seen 1500 watts with this amplifier. Waiting for my PGXL to be built.

As a side note, I have my flex timing set to 30ms but I believe the expert manual says 10ms is good enough.....but check. I've never seen my flex radios flash higher like the icom problem on key down, but you can over drive the flex power higher than your original drive setting by over driving with your speech. Setting the delay won't fix this. You have to learn which words or syllables you are saying louder and learn to regulate those or just set the drive back so it doesn't occur. I have my flex drive set so my amplifiers are at least 100w to 200w back from the peak on low, mid, and high settings while on SSB, but you'll be even lower when you go to CW or Digital......c'est la vie.
(Edited)
Photo of Rick Ciotti W3DIY

Rick Ciotti W3DIY

  • 39 Posts
  • 5 Reply Likes
A nearby ham confirmed the nonlinear condition after numerous reports of splatter.
He does not run the legal limit because his amp is unable to output that much.
The exciter is a late model Apache Anan. Pure signal is not operational at this time.
(Edited)
Photo of Erik Carling

Erik Carling

  • 12 Posts
  • 1 Reply Like
My Expert 1.5k has blown its LDMOS chip after 6 weeks of ownership. I also own and have owned a fault free 1.3k for 3 years so I know my way around the SPE. Not at any time did I exceed the maximum stated input power but I did get 3 'phantom' excess input warnings in those 6 weeks. At the failure there were no warnings and the amp was at less than 1KW in the MID position. It is now back in Italy and a depressing fight to assert my rights and get it repaired under warranty has started. I shall be standing my ground. I do not believe it correct, as Mark says, to blame the Flex. 
Photo of Pat - WH6HI

Pat - WH6HI

  • 626 Posts
  • 114 Reply Likes
I have been reading the post regarding the drive issue.  I own a SPE 2K-FA so the trend seems to be that the 1.5 has a problem.  My comments tend to be more towards the philosophy of how the SPE amps are driven.  And it is simple, setting drive power from the driver end is not a good idea.  Using the SPE supplied SW is the best methodology to use.  As it allows the amps power sensing to measure the power including PeP.  The app allows setting the limit that is fed back to the Flex (as the driver) allowing real time limiting of the drive.  When I set this value (in real time) from the SPE SW , I do it int he mode that I will be using.  I set the drive as I modulate to the point that gives me the out put I want, and never exceeding max value. In fact I run conservatively.  This is not another form of ALC, but will limit at the set point as seen from the amp point of view.  This is important in that setting the drive from the driver side will and can cause overdriven condition, that will may or may not be captured by the amps protection.  In the case of the 1.5,  there may be a problem with this and more the reason to use the recommended procedure and SW. 
Photo of Rick Ciotti W3DIY

Rick Ciotti W3DIY

  • 39 Posts
  • 5 Reply Likes
After corresponding with the US SPE dealer and with input from Italy as well I learned the overdrive faults are not caused by the 6600M after all. 

From SPE Italy...
 "when the amplifier sees the SWR jump (greater than 1.7:1), immediately switches to MID without any alarm in the log.  But the driving power previously set for MAX may be too much for the MID position, then the OD alarm".

At my QTH a 1.7:1+ SWR spike is very possible since my 80M/160M inverted Vs are 260' long and supported by 110' poplar trees, which move around a lot in the wind. This along with wet conditions does not help matters. 

From SPE Italy...

With SWR 3.0, the maximum power handled by the antennas is greatly reduced respect to the factory declared value. 
The customer can try to lower the driving power when use the amplifier on the edges of the
band. 

With the 40M monoband antenna the bottom of the phone band is where the SWR is near the limits of the SPE 1.5K. Caused by coupling with the 80M array, it is what it is.

In summary the amp is FB if connected to a dummy load, but to my antennas not so much.

(Edited)
Photo of Pat - WH6HI

Pat - WH6HI

  • 626 Posts
  • 114 Reply Likes
When my 2k faults and falls back to mid, because the Amp is controlling the flex out put and that output was previously set to a proper level, there is no overdrive problem.  Getting the amp set up properly using it’s available tools is the safest way to go.  Not trying to fault you or Flex or even the amp. But proper operational insight is necessary for safe and reliable operation.  
Photo of Rick Ciotti W3DIY

Rick Ciotti W3DIY

  • 39 Posts
  • 5 Reply Likes
The low, mid and high power settings are set exactly according to the SPE manual.

Apparently when it trips on high SWR the drive remains high momentarily, when in the mid setting. SPEs position is if the SWR never went high this would not be an issue.

More on this from SPE...

The protection "attack time" was greatly increased to help improve the protection of the amplifier. 
(Edited)
Photo of Pat - WH6HI

Pat - WH6HI

  • 626 Posts
  • 114 Reply Likes
In the case of the 1.5 it could be not enough.... no protection system is perfect.  For a lot of reasons.  
Photo of Rick Ciotti W3DIY

Rick Ciotti W3DIY

  • 39 Posts
  • 5 Reply Likes
Since the LDMOS was replaced I have been operating with the SPE AT disabled on 40 80 and 160M, using a Palstar HF Auto instead. No further overdrive errors BUT today on 160M, while hitting the Flex 6600M tune button, there was a flash from the SPE, an arcing sound and then the Italian smoke leaked out. Heading back to SPE US for round 2.
(Edited)
Photo of Ignacy Misztal

Ignacy Misztal

  • 8 Posts
  • 0 Reply Likes
I have 2k-fa and 1.3k-fa.

I use 2k-fa with K3. 2k-fa went 4 time for repairs, with at most one due to overdrive. 
I use 1.3k-fa with 6600. Low power drive mod. Numerous bad antennas and high SWR. Never a problem. Automatic power control for LOW/MID/HI works well. 

Ignacy, NO9E
Photo of Tom Worthington

Tom Worthington

  • 45 Posts
  • 5 Reply Likes
I just received my SPE 1.5K amp.  I have it hooked to my 6600M with a 3 wire cable as described and with an ALC cable and Hardware ALC is turned on (Version 3.1.8).  All that seems to work correctly but I am just getting started.  The max drive for the LOW setting is 7-10 W depending on the band.  For MID I need about 15 W and about 25 W for the HIGH setting.  I set those in CW mode.  I have been monitoring the ALC voltage with an oscilloscope and it seems to be working as expected; if I raise the drive power or sit close to the output limit in CW mode and then switch to SSB (DX processing) I occasionally see some ALC action.  So far so good. 
73 and Aloha, Tom, NH6Y
Photo of Erik Carling

Erik Carling

  • 12 Posts
  • 1 Reply Like
Just my opinion but to me having to change the input power depending on whether the amp is in HIGH, MID or LOW is not good. My 1.5 behaves like that (actually 'behaved' as it is back in Italy after a failure). But my SPE 1.3 takes exactly the same input power to drive it to just under maximum output for all of the power settings, 500W, 1000W and 1300W. There is no need to adjust and no risk of forgetting to reduce input when you change from HIGH to LOW. 
Photo of Ignacy Misztal

Ignacy Misztal

  • 8 Posts
  • 0 Reply Likes
This is not my experience. With low power drive, about 4W for LOW, 6W for MID, and 8 W for HIGH. More on 10M. 

One way to determine the optimal driving power is to increase drive watching output power, and stop when teh increase is much less than linear.

For instance if 1 W produces 100W, 3 W produce 290W, 5 W produce 450W, and 6 W produce 470W, 5 W seems OK and 6W is too much. 

One can drive with 8 W in Low and the amp will work but will be greatly distorting on SSB. Probably no effect on CW.

Setting different power levels on many bands is trivial with Flex setting in SPE.  
Photo of Eric Gruff

Eric Gruff

  • 67 Posts
  • 27 Reply Likes
FWIW, I had my 1.3K-FA go up in smoke three times with my 6700 driving it. The last time was on the Low power PA setting on 80 M FT8 (probably 600-700 W RMS power) into a tuned dipole with SWR of less than 1.1:1. Half of the RF board was immolated by the heat, but there was no alarm or other warning before I smelled the magic smoke coming out of the amp. 

I suspect it was more of a heat dissipation issue than a spike causing the LD-MOSFET to pop. I've seen that happen with another SS PA. 

My opinion is that the SPE PAs should not be used for digital modes, period. The heat dissipation leaves a lot to be desired. My PGXL arrives today, and I intend to baby it quite a bit. I will use my Alpha 9500 and Acom 2100 for the digital modes. They can handle all sorts of insults and just keep on humming along.

Photo of Pat N6PAT

Pat N6PAT

  • 1002 Posts
  • 301 Reply Likes
I have a Tokyo Hy Power 450B that's built like a tank. Works great in any mode.  THP amps are always in high demand because they're so rock solid and reliable. In 7 years the only issue I've had with it is a blown fuse.  I sure wish they hadn't gone out of business.

The only reason I purchased an RM Italy HLA 305V is because the THP amp doesn't do 160. The 305V needs only 2 watts for full power on 160 and is very touchy if you overdrive it.  Not nearly as reliable as the THP.

Some of these new very expensive amps act like they're made out of glass. One false move and it's back to the factory.
Photo of Erik Carling

Erik Carling

  • 12 Posts
  • 1 Reply Like
If it were true that "..SPE PAs should not be used for digital modes, period" and were effectively not fit for digital they would not be fit for much, given the high percentage of digital operating that now takes place. But their respective user manuals state they are "able to mange all (digital modes) currently in use". Further, Gianfranco I0ZY posted to the SPE Forum that there is no restriction on use for digital modes. False moves are one thing but when they fail through no fault of the user, that is a different matter. 
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 4649 Posts
  • 1098 Reply Likes
And considering that the amps protection circuit is way to slow to protect it from anything. I would see about getting that fixed.
Photo of Michael Walker

Michael Walker, Employee

  • 1362 Posts
  • 401 Reply Likes
In comparison, the PGXL is pretty fast about shutting down.  It does it in micro seconds.   

There is a video on YouTube of Ranko from 4O3A having the amp transmit into a dummy load and then, with the center conductor exposed with a T connector, he uses a screwdriver to totally short the antenna.  

The amp immediately goes into standby.  No damage.

Mike
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 4649 Posts
  • 1098 Reply Likes
I read were Gerald said the PGXL is so fast that a spark can not even be seen with this test. He said before buying any amp do that test first, if it sparks,,don't buy it.
Photo of Ignacy Misztal

Ignacy Misztal

  • 8 Posts
  • 0 Reply Likes
Same was apparently done with SPE 1k-fa. After a series of problems the amp was redesigned. Testing included random short and open at full power.
Ignacy, NO9E 
Photo of Lasse Moell

Lasse Moell

  • 185 Posts
  • 32 Reply Likes
Using a SSPA and reducing power to baby it may be counter productive, UNLESS the power supply do reduce the transistor voltage. Running less that designed o/p will increase the output mismatch and dissipated power. SPE do reduce the voltage with it's three settings, allowing for a better output match vs power.
Photo of Rick Ciotti W3DIY

Rick Ciotti W3DIY

  • 39 Posts
  • 5 Reply Likes
Thanks for this feedback. Reminded by SPE US numerous times about many Flex owners with SPE amps who have no issues. Started to wonder about my operating conditions. The ACOM1000, which is my backup amp, is running FB into the same antennas as the SPE and does not require a tuner.
Photo of Chris DL5NAM

Chris DL5NAM

  • 737 Posts
  • 163 Reply Likes
your Acom using a PI filter at output !!
Photo of Rick Ciotti W3DIY

Rick Ciotti W3DIY

  • 39 Posts
  • 5 Reply Likes
Yes it is, I have not located a SPE schematic to know how its designed.