Simple Class H Modulator

[Gilly]

Hi Steve

Going off to another subject ... your Simple Class H ... is there any changes to the design that you posted on AMFONE...

Are currently building it and Bruce commented to me that there could be some small changes etc:


Thanks


Wayne

VK3AIY

[steve_qix]

To answer Wayne's original question - there may be some differences from what's on AMFONE... so here is the most recent stuff. I'll create a whole post here, to fill in any gaps and start a complete thread

This is the new class H modulator I've cooked up recently. I also have printed circuit boards for the implementation.

The on-board implementation (4 modulator MOSFETs) will support RF amplifiers up about to 3 Amperes at 30 volts carrier out. The board could be used with higher voltages (up to 45 volts DC at carrier).

Here is the board mounted on a 12 x 6 heat sink, and there is more than enough room to built the entire RF amplifier on the same heat sink (which is what I'm going to do), creating a compact transmiter.

To run higher power, MOSFETs must be connected "off board", and with more spacing between them. The on-board MOSFET spacing restricts the total power to about 100 watts.

I am currently running 180 watts (40 volts @ 4.5 amperes) input by outboarding 2 additional MOSFETs in each bank. This is very simple to do - just run wires over to the board from the MOSFETs and source and gate resistors. A bus may also be used.

Rich KB2AM is running a 12 MOSFET implementation using this board, with all 12 MOSFETs (6 in each bank) off-boarded. Works like a charm.

I do have extra boards if anyone is interested.

Note:R33 (shown as 33k on the schematic) is voltage dependent. If you are running 40 volts, 33k is a good number. If you are running 30 volts, 22k is better. This could easily be installed as a variable resistor (50k or so).



Here is the latest schematic. It is also available as a .PDF at http://www.classeradio.com/easy_class_h_rev_b.pdf

For printing: http://www.classeradio.com/easy_class_h_rev_b_left_half.pdf
http://www.classeradio.com/easy_class_h_rev_b_right_half.pdf

[K6IC]

Hey Steve,

Looks great ... I' like one board if you have one to spare. Think I have almost all of the other parts.

73 Thanks again, Vic

[Gilly]

Thanks Steve

Just a question ..... Will be using a carrier voltage of 30VDC and have 2 Power Transformers to choise from for the HV...

One has 60VAC with a CT and the other 45VAC with a CT...

For 30VDC Carrier which would be the better choice ?

Also for Q1 and Q2 was going to use IRF840 FETS.......they should be fine do you think...?


Wayne

[steve_qix]

Thanks Steve

Just a question ..... Will be using a carrier voltage of 30VDC and have 2 Power Transformers to choise from for the HV...

One has 60VAC with a CT and the other 45VAC with a CT...

For 30VDC Carrier which would be the better choice ?

Also for Q1 and Q2 was going to use IRF840 FETS.......they should be fine do you think...?


Wayne

Ok, the transformer will work very well, I believe, for 30 VDC. Use the 60 VCT winding in the center tapped mode, which will give you around 40 volts (maybe a little less, but close). This will be the carrier supply. Then use the 45 volt winding in bridge mode which will give you about 60 volts DC. Put these in series, and the high voltage output will be the positive peak supply.

On the IRF840, I would not use that small device unless you are going to run fairly low power. The power dissipation (and current) is small for analog service. Consider this: If you run 30 volts at 3.5 amperes (about the max you can run with 2 board mounted banks and without "outboarding" more MOSFETs) with the power supply voltages proposed, the average dissipation (under full modulation) in the top bank of MOSFETs will be around 165 watts. But the peak dissipation will be about twice this value - exceeding the device dissipation of 2 devices.

You are also getting close to the maximum current rating at full modulation - about 12 to 13 amperes - split between 2 devices is getting close to 8 amperes (each - the max DC current).

Unfortunately, class H is still analog service, and as such, there is a lot of power dissipation under modulation. If you use 4 of them in each bank (and run 30 volts at 3.5 amperes), it will probably work.

Anyway, just some thoughts. I picked the IRFP260n because it has such a high current rating that we will never exceed it, and the dissipation is high enough to withstand the riggors of analog service There are probably other high dissipation devices that will work as well !

[Gilly]

Hi Steve


I made a typing mistake...problems when posting late at night...

The other Power Transformer was 90VAC with a CT....not 45VAC..
So have 2 ... one at 60V and the other 90V....

OK on the IRF840's....have plenty of others IRFP450 so will use them....


Thanks


Wayne

PS: Just a couple of other things :

R46 Cur/Voltage
R41 current zero
R49 current gain

How are these adjusted doing testing.......thats all...

Thanks


Wayne

[steve_qix]

Ok on everything !

The power transformer will still work. Use the 90 VCT winding in ceter tapped mode, which will give you the same voltage as a 45 volt winding in bridge mode, and put the DC outputs in series (as talked about above). That should work out nicely!

The IRFP450 appears to be less than 1/3 of an IRFP260 with respect to current, but the dissipation is high enough that you could most likely use 2 IRFP450s to replace 1 IRFP260. Double the value of the source resistors for the IRFP450. Really, to be absolutely safe, you could use 3 IRFP450s to replace 1 IRFP260 becuase of the lower current rating of the IRFP450 - and triple the source resistors. Either way, it's an off-board solution (if you use the board).

The setup of the overload calibration resistors relatively easy. I'm writing documentation to go with the board, and the setup will be in there I haven't completed the documentation yet, else I would copy that section and post it!

Talk later and Regards,

Steve

[steve_qix]

Here's the section from the documentation about calibrating the overload circuit:

To calibrate the overload shutdown circuit:

1) With the high voltage turned off and no current flowing to the output, adjust R41
(Current Zero) for 0 VDC at U3, pin 8. Set R49 (Current Gain) full on (lowest
resistance)

2) Set R46 (Cur/Voltage Balance) to the mid-point

3) With the high voltage applied, the carrier level set to the desired output voltage,
and the RF amplifier set for its normal operating current, adjust R46 (Cur/Voltage
Balance) so there is a slight positive voltage at U3, Pin 14.

These settings should allow the transmitter to be fully modulated without tripping the
overload sensor, while at the same time generating an overload if the RF amplifier
current is more than about 25% of normal. If false overloads are generated under modulation,
adjust R46 for a slightly higher voltage at U3, P14.

Note: when making the above measurements, it may be necessary to place a resistor in
series with the meter probe, at the very end of the probe, to prevent RF from getting into
the circuitry. A 2k to 4k resistor works well in series with a standard 20,000 ohms-per-
volt VOM.

[Gilly]

Hi Steve


Thanks for all the information...
Will order some IRFP260's ..... and use them and not the IRFP450's......
There certainly is a big difference in the current handing between the two...didn't realize that actually....

Thanks again..


Wayne