Transorbs

Anything goes here, but abusive posts will be deleted.

Transorbs

Postby K9ACT » Thu Sep 23, 2010 6:47 pm

Does anyone have or know where I can get 1.5ke540a transorbs? The usual sources don't even list them.

Aside from that, is there any reason not to use two 250V jobs in series?

Next, after installing the D Limit on the modulator, I see that we should put a Transorb in the RF deck at this end of the line.

What is the difference in function of these two devices?

js
K9ACT
Class E Enthusiast
 
Posts: 146
Joined: Sun Jun 27, 2010 12:03 am

Re: Transorbs

Postby n4lta » Fri Sep 24, 2010 7:34 pm

I have had the same problem recently. I tried to get samples - I usually can get them through my engineering business - but couldn't get any.

BTW - I love your web site - I also do astronomy and sausage making. Have used some of your recipes.

Pat
N4LTA
n4lta
Member
 
Posts: 20
Joined: Thu Aug 05, 2010 8:53 pm

Re: Transorbs

Postby steve_qix » Mon Sep 27, 2010 8:10 am

They're hard to find in franchise distribution. I have some a bunch of them here ( I guess I should check my stock to get an exact number), and they are in production according to the manufacturers.
Visit the class E web site at:

http://www.classeradio.com
User avatar
steve_qix
Class E Guru
 
Posts: 434
Joined: Fri Jan 31, 2003 11:03 am
Location: Townsend MA

Re: Transorbs

Postby Gilly » Mon Sep 27, 2010 8:20 am

Hi Steve

Just on the Class H modulator.....last time we spoke you were fine tuning some component values within the same circuit...etc:

Just wondering if its all finished now.....?



Wayne
VK3ALK
Gilly
Class E Enthusiast
 
Posts: 167
Joined: Tue May 15, 2007 6:59 pm
Location: Mt.Eliza Victoria Australia

Re: Transorbs

Postby K9ACT » Mon Sep 27, 2010 8:51 am

steve_qix wrote:They're hard to find in franchise distribution. I have some a bunch of them here ( I guess I should check my stock to get an exact number), and they are in production according to the manufacturers.


Please send me 10 of them if you can and I will PayPal by return email.

My rig runs forever at 100W but at the 200W level, it sooner or later fails with full modulation and it's always drain to source short.

js
K9ACT
Class E Enthusiast
 
Posts: 146
Joined: Sun Jun 27, 2010 12:03 am

Re: Transorbs

Postby steve_qix » Mon Sep 27, 2010 4:24 pm

K9ACT wrote:
steve_qix wrote:They're hard to find in franchise distribution. I have some a bunch of them here ( I guess I should check my stock to get an exact number), and they are in production according to the manufacturers.


Please send me 10 of them if you can and I will PayPal by return email.

My rig runs forever at 100W but at the 200W level, it sooner or later fails with full modulation and it's always drain to source short.

js


You shouldn't have device failures. Does the failure occur when you're just talking, or when going from receive to transmit (or visa versa)?
Visit the class E web site at:

http://www.classeradio.com
User avatar
steve_qix
Class E Guru
 
Posts: 434
Joined: Fri Jan 31, 2003 11:03 am
Location: Townsend MA

Re: Transorbs

Postby K9ACT » Mon Sep 27, 2010 7:00 pm

The fets only fail with modulation. Interestingly, it seems to be only when I am speaking into the mic.

I can run the CQ robot into a dummy all day.

jd
K9ACT
Class E Enthusiast
 
Posts: 146
Joined: Sun Jun 27, 2010 12:03 am

Re: Transorbs

Postby steve_qix » Mon Sep 27, 2010 9:30 pm

If it never occurs on transition (from receive to transmit or the other way), you may have a parasitic. Class D is rather different from class E. I would carefully monitor the drain waveform(s), as an envelope and as RF, to see if there are ever any high voltage spikes, grass or other anomalies present particularly under high positive modulation. Intermittent parasitics are not often easy to diagnose, but I've seen plenty of them over the years. Almost always due to layout.

Regards,

Steve
Visit the class E web site at:

http://www.classeradio.com
User avatar
steve_qix
Class E Guru
 
Posts: 434
Joined: Fri Jan 31, 2003 11:03 am
Location: Townsend MA

Re: Transorbs

Postby kf1z » Mon Sep 27, 2010 10:23 pm

The "CQ robot" does not present things such as breath-pops to the modulator.
At least I wouldn't think so...

Not sure how it was recorded, or how it's piped into the modulator.


Will it fail if you use the mcrophone into the dummy load ?

Or only through an antenna?

If you don't have a good quality shunt capacitor from drain to drain, or from each drain to ground to quell the high voltage peaks,
then that can cause the problems too...

With my class-D playing around, I have seen peaks on the drains of well over 10 to 12 times the V+ input.
Popped 100 volt mosfets, with only 12 volts total dc supply !

In your case ( I believe 90 volts? HV DC supply )

That could make the peak voltage at the drains be 900 + volts. on highest modulation peaks.

So yes... need to look at those drains with the scope.
User avatar
kf1z
Senior Contributer
 
Posts: 558
Joined: Sat Mar 18, 2006 7:07 am
Location: Ely, NV.

Re: Transorbs

Postby steve_qix » Mon Sep 27, 2010 10:50 pm

kf1z wrote:With my class-D playing around, I have seen peaks on the drains of well over 10 to 12 times the V+ input.
Popped 100 volt mosfets, with only 12 volts total dc supply !


Wow, what did you do to get those high peaks? You can't get that with class E, even when worked into an open circuit. That would definitely cause problems.
Visit the class E web site at:

http://www.classeradio.com
User avatar
steve_qix
Class E Guru
 
Posts: 434
Joined: Fri Jan 31, 2003 11:03 am
Location: Townsend MA

Re: Transorbs

Postby K9ACT » Tue Sep 28, 2010 12:08 am

Here are some interesting factoids:

I used the rig at 100W for the net today and it worked fine for almost two hours.

Later on, I ran it up to 150W and ran it for 30 minutes on the CQ robot/dummy with no problems. I then switched to the mic and within a minute or so there was a loud pop and the line fuse blew.

Damage assement: one RF FET, one Mod Fet and one gate driver.

Probable cause: Old Man Stuff

Yesterday, I could not get it to work at all on the antenna because of parasitics. After trying all the usual tricks, I put the shunt cap back across the drains and it all got better. I added it with clip leads so when I had it back on the bench today, I soldered it back in "properly". After the bomb damage, I noticed that I had one side of the cap on the drain buss but the other was soldered to the gate buss. Guess which side all the blown components were on?

The CQ robot was recorded using the same audio chain as the mic. I does however, go back through it again but it's not obvious what difference that would make relative to this problem.

Anyway, with all this scary talk about high voltages being generated, sounds like the transorbs would solve my problem.

js
K9ACT
Class E Enthusiast
 
Posts: 146
Joined: Sun Jun 27, 2010 12:03 am

Re: Transorbs

Postby steve_qix » Tue Sep 28, 2010 6:15 am

Hi Jack,

Welcome to the world of research and development ! That's what you're doing at the moment :D because you're design and implementation is not a tested or proven configuration !

First: You would eliminate your catastrophic, cascading failures if you used the overload board. Your modulator would never fail, and probably the RF driver ICs would also be spared. If you ever had an RF MOSFET failure, the system would shut down fast enough to protect the modulator and quite probably the driver too. Really, do put that overolad system properly, as documented, into your modulator/power supply. This will keep any failures local.

Second: The transzorbs will NOT fix this problem. The transzorbs are there to stop back-spikes from things like antenna disconnection during transmisions, keying up into no load and OCCASIONAL things like this - for VERY SHORT durations until the overload system can get things shut down (if there's a current overload from the "event"). They cannot possibly absorb the energy contained in a sustained parasitic oscillation or high voltages from a bad, missing or improper shunt or RF amplifier design/implementation.

I worked on this stuff on and off for approximately 20 years before coming up with something that was reliable enough to release to others. During this R & D process, I tried numerous designs (class D, and class E) and discovered what worked consistantly and what didn't. Ultimately, the end process of all this were the current class E designs that are proven reliable.

If I had had the FQA11N90 30 years ago (when I started the development project), maybe I would only have needed a few years to get a stable design - and the IXDD414 also really make the implementation very predictable! Hard to say exactly if the development time would have been really short, because, due to the lack of "prior art" at the time, EVERYTHING had to be figured out from scratch, so to speak. I do have class e transmitters that are 10 years old, and have never had a failure....and they've been run into no load and short circuits at full power many times.

Anyway Jack, you're certainly rediscovering through your unplanned "R & D" the reason for things like the overload shutdown circuits, sequenced T/R and PLENTY of headroom in component specifications!!!! 8)

Regards,

Steve
Visit the class E web site at:

http://www.classeradio.com
User avatar
steve_qix
Class E Guru
 
Posts: 434
Joined: Fri Jan 31, 2003 11:03 am
Location: Townsend MA

Re: Transorbs

Postby K9ACT » Tue Sep 28, 2010 11:49 am

After reading Steve's last, I was mentally composing a response about the "brewing" aspects of home brewing vs kit assembly and other philosophical notions. But I did download the overload schematic just to start thinking about it.

In the mean time, I had corrected all the known problems and tested with and without an antenna for about 30 minutes with no problems when a friend found me hiding way down the band. We talked for a few minutes and he asked me to move back up to the net freq. I moved, tuned up on low power and within about 5 seconds at 150W, it crashed again.

I am becoming a believer.

As a point of interest, unlike comments here, my rig draws very little current with no antenna or a mis-tuned antenna. Is this a Class D thing or dumb luck?

js
K9ACT
Class E Enthusiast
 
Posts: 146
Joined: Sun Jun 27, 2010 12:03 am

Re: Transorbs

Postby K9ACT » Tue Sep 28, 2010 2:50 pm

Upon further investigation, it seems that I may be missing the forest for the trees.

Ponder this photo...Image

When I first noticed this pattern I assumed that it was just an imprint from the pattern of the silpad.

Under the microscope it is clearly reformed aluminum and not just rubber. These things got so hot, the aluminum melted; so it obviously is not a transient problem but a more basic heat sink problem. Notice how the pattern does not appear where the screw applies direct pressure but away from the screw.

These two would indicate that the one will less of the pattern was screwed down more tightly than the other. This sort of presents a quandry as screwing down too tight can cock the device and produce the same problem.

BTW, these are the last two devices to fail.

Any thoughts?

js
K9ACT
Class E Enthusiast
 
Posts: 146
Joined: Sun Jun 27, 2010 12:03 am

Re: Transorbs

Postby steve_qix » Tue Sep 28, 2010 3:34 pm

I don't remember how many devices you are using. Using FQA11N90s in class E, you can safely run about 55 watts power input PER DEVICE, and expect to get 50 watts output per device. So, 8 devices will do 400 watts OUTPUT with no problem, or about 45 watts input (which is what I run them at all the time). The MOSFETs do not get hot - they get warm only.

It may be an efficiency problem - have you measured the efficiency? Still, I'd also look for parasitics. But, you should see 90+ percent device efficiency, and an overall efficiency (including tank circuit, transformer and other losses) of around 90% (plus or minus a couple percent). I've seen 92 or 93%, and I've also seen 88% - all are within the range of normal.

I don't have any direct experience with class D, but it is very easy to set class E up correctly and get high efficiency. The class E waveform tells all ! If the waveform is bad, just adjust the variable capacitors in the tank and make it right. I don't even bother looking at the waveform any more - just use the drain current meter and RF output meter. Works like a charm !

Regards,

Steve
Visit the class E web site at:

http://www.classeradio.com
User avatar
steve_qix
Class E Guru
 
Posts: 434
Joined: Fri Jan 31, 2003 11:03 am
Location: Townsend MA

Next

Return to General Chit-Chat

Who is online

Users browsing this forum: No registered users and 1 guest

cron