|
Chris,
Long ago, on QDX, we learned that L14 was capable of producing very large Ldi/dt spikes when the current through it changed abruptly at the end of a transmission. A commutating diode across L14 was proposed to absorb the spikes.
In QMX the presence of an envelope modulator was thought to moot that problem by gracefully winding down the current in L502.
However, rapid changes in load can also cause Ldi/dt reaction at L502. Rapid relay switching of an ATU almost certainly can do that.
I believe that may be the source of the high voltage spikes you observe when allowing an ATU to seek a match.
JZ KJ4A?
I'm doing a deep dive into cause of BS-170 failures.??
Currently I have finished evaluating peak BS-170 Drain to Source voltage and am in the middle of power dissipation for various SWRs.
- On 160m and 80m Peak voltage exceeds the BS-170 60v spec.? Probably should not use at 12V operation, particularly at 160m
- Using an Elecraft TI ATU there are spikes during tuning of over 100v!
I will am putting together a detail presentation.? Here are a couple of charts:
?
?
|
On Tue, Sep 17, 2024 at 03:37 PM, ajparent1/kb1gmx wrote:
What a small minimum, usually 1.5:1 or less.? depends on how your pushing
the finals.
?
Thanks Allison, that was what I needed.? up to 1.5 should be safe, may get away with 2:1 but your pushing your luck.
?
Chris, your first chart was the the one I understood the most, but you are 12V and I am 9.? But it looks like 40M should be okay on both sides of 2:1.? Transient spikes are not an issue, I don't use an auto tuner for the QDX, have seen too many posts about this, although some have had no issues, I don't want to find out I am one who will.? And not sure I understand the last graph.
?
either way, I think I have the answer I was looking for... was hoping 2:1 was still safe but looks like that is a maybe, but probably not.? But I should be okay at 1.5:1 and this gives me something to work with.
?
Also I do acknowledge that the two tuners I use may not be functioning at 100% and will look into those issues as well.? As I use the 4SQRP 4S? tuner for QRP and as Daniel said that tuner is not great on a resonant antenna, I am probably going to install a bypass switch on it, after I test it and or build the spare I have sitting in my shack.
?
73
Bob
KG5GTE
|
Allison,
Tuning the antenna is all well and good, but when you use an antenna portable (as I and many other POTA/SOTA) ops do), that same antenna can have quite different properties depending on the deployment. Tuning the antenna in the field takes time and probably extra parts and equipment. Using an antenna tuner (with appropriate protection, like an absorptive bridge or much lower power output) often makes much more sense than tuning the antenna itself.?
--
73, Dan? NM3A
|
Rather than use a ZM2 (a good unit) I tune the actual antenna for less than 1.5:1
then I can omit the tuner.
?
Of course depending on the design of the EFHW some do not tune uniformly
on all bands.
?
--
Allison
------------------
Post online only,?
direct email will go to a bit bucket.
|
The problem can be there is no acceptable SWR above a small minimum...
?
For example when you tak about 2:1 swr the impedance can look like 25 ohms?
and that would cause high current in the finals. or it can be 100 ohms where
higher voltages might be developed.
?
That is one point and if its really 50 ohm btu has 25 ohms of capacitive
reactance the impedance is not 50 ohms and that also stresses the transmitter.
?
What a small minimum, usually 1.5:1 or less.? depends on how your pushing
the finals.
?
--
Allison
------------------
Post online only,?
direct email will go to a bit bucket.
|
Tony,
"Your zener fix was one of a few that were discussed way back when. Hans
added diode pads to the schematic so he was OK with this workaround."
Hans had a lot of uncertainties about Zeners and never included any. When I got my last QMX, I noticed that those pads were missing on the newer Rev.?
?
I, too, have had no issues with PA failures over the last 6 years and 9 kits. No good answer as to reasons for some having multiple failures.?
--
73, Dan? NM3A
|
You may use a 3dB or 6dB attenuator between TRX and Tuner to protect the PA. This way the PA will always have a reasonable SWR no matter what the tuner does.
BR
toggle quoted message
Show quoted text
On 17. Sep 2024, at 21:37, Tony Scaminaci via groups.io <tonyscam@...> wrote:
I second Joe¡¯s selection. I used the ZM-2 to tune my EFHW antenna so all the 40-10 meter WSPR frequencies showed less than 1.5 VSWR. Now I don¡¯t even need to use the tuner for WSPR. The absorptive bridge will keep the QMX safe during tuning. The best tuner that I have found for the QMX is the ZM-2 ATU. It has a tune option that has a resistor that keeps the QMX from seeing a high SWR. Just adjust the two knobs until the light goes out and then switch to Operation mode.?
|
One note¡ the ZM-2 is not an ATU, it¡¯s a manual tuner.
On Tue, Sep 17, 2024 at 2:38?PM Tony Scaminaci via <tonyscam= [email protected]> wrote: I second Joe¡¯s selection. I used the ZM-2 to tune my EFHW antenna so all the 40-10 meter WSPR frequencies showed less than 1.5 VSWR. Now I don¡¯t even need to use the tuner for WSPR. The absorptive bridge will keep the QMX safe during tuning. The best tuner that I have found for the QMX is the ZM-2 ATU. It has a tune option that has a resistor that keeps the QMX from seeing a high SWR. Just adjust the two knobs until the light goes out and then switch to Operation mode.?
|
I second Joe¡¯s selection. I used the ZM-2 to tune my EFHW antenna so all the 40-10 meter WSPR frequencies showed less than 1.5 VSWR. Now I don¡¯t even need to use the tuner for WSPR. The absorptive bridge will keep the QMX safe during tuning.
The best tuner that I have found for the QMX is the ZM-2 ATU. It has a tune option that has a resistor that keeps the QMX from seeing a high SWR. Just adjust the two knobs until the light goes out and then switch to Operation mode.?
|
The best tuner that I have found for the QMX is the ZM-2 ATU. It has a tune option that has a resistor that keeps the QMX from seeing a high SWR. Just adjust the two knobs until the light goes out and then switch to Operation mode.?
|
Chris,
As I've stated many times before, NEVER exceed the absolute?maximum rating of a semiconductor as spec'd by the manufacturer. Doing so is asking for trouble, possibly short term, or maybe a gradual weakening of the device under stress until it fails long term. Furthermore, relying on a manufacturer?underrating their device is a bad idea. I've seen a few manufacturers underrate by as much as 20% but this is not at all typical across the semiconductor industry. You'll find that many devices have very little safety margin relative to their maximum ratings so my advice is to never exceed?the manufacturer's specs under any conditions.
As far as current-limiting to 100 uA or so during breakdown testing, I don't think it will save you from?permanent?damage but you can certainly try this on your first DUT and see what happens. If you see some odd behavior after the FET breaks down, you should consider it unreliable even if it seems to work when the voltage is lowered. I always consider breakdown voltage testing as destructive. If you test multiple devices, I'm not sure what the stats will look like unless you're willing to test hundreds of them. The same goes for any TN0110 testing you consider trying.
Your zener fix was one of a few that were discussed way back when. Hans added diode?pads to the schematic so he was OK with this workaround.?
How many BS170's are you planning to perform breakdown testing on?
Tony
On Tue, Sep 17, 2024 at 12:48?PM Chris KB1NLW via <chrisrey1= [email protected]> wrote: Thanks Tony;
?
After I get the dissipation concluded I am thinking of measuring breakdown of my batch of BS170.? I think the QMX design is relying on the typical breakdown being significantly higher than the speck.? If I do this I should also check TN0110.
?
I'm thinking of limiting current during breakdown to ~0.1ma (concerned about damage), any thoughts?
It does seem that Hans should be more concerned.
?
?
I have three QMXs that have zeners across the BS170s.? Before I added the Zeners I had multiple failures, none since!
Of course I don't have them in my QMX+ so I could do these test.??
|
Hi Chris
That is pretty bad luck. I have been operating QRP labs radios without any mods almost constantly since 2018 and never had any issues with finals. One was also working WSPR non stop.
It is very odd why only some folks suffer from blowning finals, given the vast number of sets Hans has sold.
I would be curious to learn about your antenna and PSU. Do you use an ATM?
I really hope you find the cause.
73s Ted
2E0THH
??
|
Have you tried another tuner?? I find that mine is pretty much a dud, even after hours wasted troubleshooting it.? Maybe that's the issue.?
-Bill? W3WJ
|
Thanks Tony;
?
After I get the dissipation concluded I am thinking of measuring breakdown of my batch of BS170.? I think the QMX design is relying on the typical breakdown being significantly higher than the speck.? If I do this I should also check TN0110.
?
I'm thinking of limiting current during breakdown to ~0.1ma (concerned about damage), any thoughts?
It does seem that Hans should be more concerned.
?
?
I have three QMXs that have zeners across the BS170s.? Before I added the Zeners I had multiple failures, none since!
Of course I don't have them in my QMX+ so I could do these test.??
|
Hi Chris,
Long ago, JZ noted high-voltage peaks exceeding 100V at the BS170 drains. They showed up in simulations we ran as well. There were some fixes suggested which seem to calm things down a bit but based on our results, I recommended TN0110 as an alternative device for two reasons - 100V Vds max and lower Vgs threshold to enable harder saturation with the existing 5V drive level. With the BS170, I had to drive the simulation from about 6.5V at the gate to get the same output as TN0110 with 5V at the gate.
Your first graph highlights the dangers of reflections but I also conclude that the higher frequency bands are ¡°safer¡± which is likely due to less gain at those frequencies.
Tony AC9QY
On Mon, Sep 16, 2024 at 5:30?PM Chris KB1NLW via <chrisrey1= [email protected]> wrote: I'm doing a deep dive into cause of BS-170 failures.??
Currently I have finished evaluating peak BS-170 Drain to Source voltage and am in the middle of power dissipation for various SWRs.
- On 160m and 80m Peak voltage exceeds the BS-170 60v spec.? Probably should not use at 12V operation, particularly at 160m
- Using an Elecraft TI ATU there are spikes during tuning of over 100v!
I will am putting together a detail presentation.? Here are a couple of charts:
?
?
|
Donald>
I totally agree phase of the reflected power is critical for each of the harmonics, thus affecting how the two components from the LPF and SWR will add or subtract,? and affect the peak voltage across the BS-170.? I have probably 40 scope photos and a few hundred measurements showing this.? Ultimately I will publish much of this info.
?
In the past I have stated that BS-170 failures are most likely from excessive D-S voltage.? My effort demonstrates this is the case for both some bands and steady SWR and very bad for ATUs during tuning.??
Anyone with a good scope and the ability to make dummy loads of various impedances can replicate this.?
Can you do similar test?
(FYI I have the same digital scope as Hans).
?
Currently I'm in the middle of determining BS170 power dissipation -? a little more difficult? since it require differencing various power dissipations.? So far I have used two approaches and now looking at a third approach to see if I can get more consistent results.
?
(In the past I have been responsible for many failure analyses for NASA and the Military so I may be a little anal for Ham applications)
?
?
?
?
|
On Mon, Sep 16, 2024 at 06:38 PM, Tisha Hayes, AA4HA wrote:
Modes of operation, duration, voltage peaks (current peaks) all interplay
Phase of the reflected power is also a crucial factor, not simply its magnitude.??
73, Don N2VGU
|
Both the LPFs and non matched load reflect power back to the BS170s.? This chart shows the effect of the LPFs:
?
|
Very well put together beginnings of a slide deck.
?
Trying to come up with a fixed SWR value is difficult; Partially because like another poster said, impedance is a complex relationship and it is much more than just saying 1.5:1 or 2.0:1.
Modes of operation, duration, voltage peaks (current peaks) all interplay. Then you also have batch variations within a transistor part number (I have seen some really crazy numbers in transistors that were all supposed to be in-spec but maybe 1/2 were not).
--
Tisha, AA4HA
|
I'm doing a deep dive into cause of BS-170 failures.??
Currently I have finished evaluating peak BS-170 Drain to Source voltage and am in the middle of power dissipation for various SWRs.
- On 160m and 80m Peak voltage exceeds the BS-170 60v spec.? Probably should not use at 12V operation, particularly at 160m
- Using an Elecraft TI ATU there are spikes during tuning of over 100v!
I will am putting together a detail presentation.? Here are a couple of charts:
?
?
|
Joe
Ted 2E0THH
Donald S Brant Jr
Tisha Hayes, AA4HA