Completed attenuator module with straight through (10W), 3dB (20W), 7dB(50W) and 10dB(100W) attenuators and switching relays for selecting two inputs into amp or into internal 100W dummy loads.
My quick and ugly construction method with lots of not so quick to strip coax. Not looking so good here, but it works. It's a facehugger!!!!
Trying to show the thick film power resistors sandwiched between the pcb and heatsink. The heatsink was tapped.
Last edited by mw0uzo; 02-12-2013 at 06:56 PM.
Progress:
- Attenuator module fitted and working. Wiring error meant it had to be taken out and corrected. Now I can use my ts850. Yay!
- Significant upgrade to firmware. Peak hold mode for all measured values, all user settings stored in eeprom and settings stored for each input selection.
Bad swr on 40m lpf turns out not to be 40m lpf at all. The amplifier is distorting or generating a larger harmonic on 40m resulting in the higher readings. While output is clean after the lpf - its doing its job properly - the source of the distortion has not been found. Increasing bias reduces the swr by a few points. 40m is the only band affected by this. Input is clean. Does anyone have any thoughts as to why this could be?
I may have found an answer. Apparently at 7Mhz, the 3rd harmonic is only 12dB down. This equates to 40w reflected at 600w, giving a swr of 1.7, which is the best of what I am seeing, 1.7-2.0. Perhaps normal... I hope...
Last edited by mw0uzo; 02-19-2013 at 08:18 PM.
More progress on the firmware - its at V1.0!!
- Monitor modes refined and fixed. All bar graph modes work properly now.
- Plenty of bugfixes
Used spare phono socket at rear for external bypass/standby switch for remote operation when used in a rack.
Image of amp with front panel removed
Image of completed LPF and attenuator modules
The amp is pretty much complete now. YEAH!
Some questions:
ALC - does it really need it, with the selection of attenuators available and the comprehensive protection?
SWR fault level - At the moment it warns at 2.0 and faults at 2.5 after 0.5s of continuous detection of bad swr. Is this too relaxed? (If there is a large spike of Vreflected, then it faults instantly)
Last edited by mw0uzo; 02-25-2013 at 05:03 PM.
That's a very nice piece of work. It's taken you quite awhile from start to finish, and you should be proud of the results.
A clear conscience is usually a sign of a bad memory
RIP ALBI-W3MIV RIP RUSS-W5RB RIP BOB-VK3ZL
ThanksOriginally Posted by K7SGJ
A few pics showing whats on the display now:
Bypass mode screen showing power through, swr, antenna and input settings.
Monitor mode 1. 3 line mode, all relevant information in numerical form, menu visible always
Mode 2 - Power bar with no scale
Mode 3 - 4 line mode with power bar, scale and relevent numerical information.
The other modes are for debug and other monitor purposes, I got rid of the others as they were pretty much useless and poor resolution.
Last edited by mw0uzo; 02-26-2013 at 12:26 PM.
Note for the bias setting. Now the amplifier is complete there seems to be no problem with spurious output. Bias has been set to 1A total, which reduces SWR at LPF on 40m. Also, a 4.7k resistor has been added in series with the thermistor, the bias reduction was *way* too sensitive. The thermistor has also been moved from on top of one of the MOSFETs to the heatspreader. This gives sensible, flat operation of the bias wrt temperature. Interestingly a lot of people recommend that the thermistor should be in direct contact with the mosfet, e.g. on top, but in the original eb104 article it is recommended to be in contact with the heatspreader. From what I have seen, the bias behaves correctly when the thermistor is in contact with the heatspreader, not the MOSFET.
Last edited by mw0uzo; 03-01-2013 at 08:08 PM.
Also amp has survived extended TX of CW duty cycle at power level of 500W in my recent tests, so its looking good for plenty of 400W operation :)
Now only time will tell, I will post updates to this 'log' as it is used.
Last edited by mw0uzo; 03-01-2013 at 08:06 PM.
Contacts this afternoon on 80m at 400W pep. MW0KST, GW0JAI, M0CSV and G3IUW. Band suffering from lots of fading. No smoke though and cool operation of heatsink ~25-30 degrees.
Can't wait till you get some DX there.
Posting Permissions
Reply With Quote
