A Clean Canvas - 24 Pill Class E Rig Construction

[K1JJ]

Tom,
Why do you have the duty cycle set so low? The two pots on the board should allow you to get closer then 50%

I want 99% final efficiency like some of the class E guys have....

 Actually, Steve sent me fixed resistors instead of pots for the duty cycle adjusments. I figgered they were optimized, but looks like I will have to dick around with them.

T

[WA1GFZ]

Do you have the correct frequency coming out? You could be double clocking the flip flops with internal noise.

[K1JJ]

Do you have the correct frequency coming out? You could be double clocking the flip flops with internal noise.

Yep, the input to the board is 7770 according to the counter and scope.  The output is 3885 according to the scope and counter. I hear the carrier loud and clear on 3885.

I double checked it with 7770 from an HP sine RF generator instead of the VFO.  I think a little adj with the duty cycle and some more VFO output and I'll be there.

T

[steve_qix]

Tom,
Why do you have the duty cycle set so low? The two pots on the board should allow you to get closer then 50%

I want 99% final efficiency like some of the class E guys have....

 Actually, Steve sent me fixed resistors instead of pots for the duty cycle adjusments. I figgered they were optimized, but looks like I will have to dick around with them.

T

What???  If you got fixed resistors, they were supposed to be pots.  You'll have to remove them and I'll send you the appropriate pots.  That was surely a mistake!!!!!!!!!!!!!!!  You *must* have pots.  The duty cycle has to be adjustable.

On the duty cycle:  The duty cycle coming from the VFO on 75 meters will be lower than 40% - even closer to 30%.  This is because the IXDD414 drivers in the transmitter add a bit of delay and tends to stretch the on pulse.  I've run a lot of experiments, and for optimum efficiency a duty cycle of 40-something (40 to 45 percent) is best at the gates of the RF amplifier MOSFETs.  I think I'm running about 40% on my rig that's like Tom's, although I'd have to measure it to refresh my memory.

Sorry about the parts mix-up.  I have already modified the parts list to call out 2k pots instead of 2k resistors - that's why the problem happened.  The word POT was omitted from the parts list. That's what field tests are for    (and Tom is a field test site - the VFO kit has not been officially released).

Regards,

Steve

[WA1GFZ]

Sounds like the flip flops are working fine. Now you just need to fix the duty cycle.

[WA1GFZ]

Tom when you remove the fixed resistors stick a tooth pick in the hole while the solder is soft so you can get the pot leads in.  This way you won't need solder wick and have the chance of damaged pads or traces. You can sharpen the tooth pick to the right diameter.

[K1JJ]

OK Steve -

No problem.  I'll wait for the pots.  While you're at it, please slip in a couple of 2N3904's and I'll swap the 2N2222 in the VFO too.  Then I'll play with the gimmick cap and the 8.2K feedback resistor to try for more output.


That's a neat idea to use a toothpick, Frank. I usually use wick.  But these boards are pretty rugged and I've never lost a trace with them yet. The plated thru holes are FB.

T

[WA1GFZ]

wick works fine if you can get it in the right position without over heating the board. I find boards take less damage when you just open up the hole to insert new component. The excess solder just becomes part of the new joint.

[KF1Z]

I blow em out with a Rat-Shack solder blowin' iron...

Supposed to suck.

but it blows betta.

[steve_qix]

I blow em out with a Rat-Shack solder blowin' iron...

Supposed to suck.

but it blows betta.

That's what I use, too.  You can remove an IC from a board and not damage the etch in any way.  Works great !

[David, K3TUE]

I blow em out with a Rat-Shack solder blowin' iron...

That's what I use, too.  You can remove an IC from a board and not damage the etch in any way.  Works great !

The right tool for the job is always a joy to use.
Love my HAKKO 808 Desoldering Gun.

[WA1GFZ]

rubber bulb Solder suckers  generate a lot of ESD. We threw them away at work about 15 years ago. wick is much safer.
A commerical solder sucker is fine as long as it is esd rated.

[KF1Z]

So does the dry air from the woodstove...

All the ground pins cut off the soldering irons...

Metal workdesk....

Cardboard to lay the pcb on while soldering.

ESD   CAN ruin some certain parts...

Doesn't happen very often in the home workshop.

Never here that I'm aware of.

So why get over-worried about it?

 

[K1JJ]

Is this the Radio shark gem you guys are talking about?  Not bad for $10.
http://www.radioshack.com/product/index.jsp?productId=2062731

That Hakko 808 looks quite nice in the ads. It's $200 on eBay.   I wonder what size tips are used for PCB work like the e-rig boards?  There's lots to choose from.

Steve, do you use the RS unit to suck the solder out or blow it thru like Bruce mentioned?


I bought a bundle of Wick on e-Bay a few years ago. Thought I'd be set for life. Turns out the bastard sold me a slug. It was plain copper shield with no flux in it. Tough to use. QC reject...   Hmmm... maybe I can soak it in some.

T

[steve_qix]

That's the baby !  Works great 

[WA1GFZ]

I've seen my share of blown parts due to ESD. It is a very big problem in the modern world. I never handle any HPSDR hardware without a ESD strap. Some guys have blown parts by not using the right equipment.
I've even seen RF fets like MRF150s damaged by ESD.
We get trained every year on ESD. We even recorded SEU events in our hardware. Not a problem with a 6146.

[KF1Z]

Is this the Radio shark gem you guys are talking about?  Not bad for $10.
http://www.radioshack.com/product/index.jsp?productId=2062731

Thats the one!..

Been using this one for 15 years plus....

Darn near ready to fall apart!

But still blows good!

[KF1Z]

Tom,

what value resistors do you have on your inputs to the IXDD414's ?

[K1JJ]

Tom,

what value resistors do you have on your inputs to the IXDD414's ?

On the 2-band VFO driver board?  There are none. Check the schematic on page four of this thread.

T

[N3DRB The Derb]

on my last job, getting caught with a board in your hand without your esd strap on your wrist and plugged into the mat on the bench OR wrapped in a ESD bag was a termination offense after the 2nd action. They let you have one chance, after that you was gone escorted out the door.

[steve_qix]

Tom,

what value resistors do you have on your inputs to the IXDD414's ?

On the 2-band VFO driver board?  There are none. Check the schematic on page four of this thread.

T

I'm guessing Bruce meant the IXDD414s at the RF amplifier that drive the RF MOSFETs.  There should be a 300 ohm resistor at the input of each IXDD414, and when combined will form a 50 ohm load (this is done for both phases).  Unless I missed something in the implementation and there are more than 6 IXDD414s per phase..

[K1JJ]

Yes, I'm using the 300 ohm resistors as per the design for 24 pills posted back on an earlier page here. This equals 300/6 = 50 ohms load per phase.  (One 414 per two fets)

Steve, I'm testing the power PDM boards now.  That floating charge pump is confusing. Can the modulator be tested as a stand alone unit considering the floating ground and charge pump?  I have a good signal going into the opti-coupler but nothing coming out at pin 5 - and the charge pump regulator is not regulating at 12V, but instead showing 16V at real ground or 0V at floating ground.   The charge pump seems to be the issue.  How is the unit tested in light of this or am I missing something?

T

[KA2QFX]

Correct me please Steve if I'm wrong but... 
With nothing connected to the modulator output there appears to be no ground return for the floating supply regulator. The RF deck would be the ground return in operation.  For test purposes I'd tie the floating and real grounds together with a 1K 10Watt resistor just to get that regulator biased on.

[steve_qix]

Correct me please Steve if I'm wrong but... 
With nothing connected to the modulator output there appears to be no ground return for the floating supply regulator. The RF deck would be the ground return in operation.  For test purposes I'd tie the floating and real grounds together with a 1K 10Watt resistor just to get that regulator biased on.

Correct     Tom, you can test the modulator drivers, etc. by simply grounding the output of your PWM filter (or the output of the modulator board).  Obviously, no hi voltage can be applied to the modulator !!!!  This will bring the charge pump diode into conduction, and the power supply on the PWM output board will function properly.

There's a nice feature of using the charge pump approach - if there is no load on the modulator, it doesn't produce any output!

[WA1GFZ]

You can also check it with 12 volts on the drains from the LV supply just to see it function. bootstrap driver takes some getting use to but they function well.  We used them in the F111 weapon system release drivers because N channel FETs had lower RDS on back in the day.