Register Account

Login Help

Forum Home - Rules - Help - Login - Forgot Password
Members can access, post and reply to the forums below. Before you do, please first read the RULES.

Power Supply Causing Dirty CW Signal?

Sep 19th 2014, 03:59


Joined: Aug 8th 2014, 13:19
Total Topics: 0
Total Posts: 0

I'm working on an Astron RS-35A linear power supply for a fellow ham. The power supply is said to be less than 10 years old.

He received a report of hum on his CW signal with a K3. I have confirmed the noise on CW only (no noticeable noise on SSB), using my FT-100D on his power supply. CW signal is clean with my power supply (new RS-35M).

Here's where I'm stumped:

Output voltage is 13.93 VDC, drops to 13.91 under full load (almost 20A). This is acceptable.

Ripple is less than 20mV peak to peak no load, increases to 30mV peak to peak under 20A load. Astron claims these supplies (new ones) get better than 5mV ripple at full load, but 30mV is still pretty good (or is 30mV horrible for CW?)

Floating/Grounding negative terminal makes no difference in ripple.

I could just 'shotgun' the supply, replace every electrolytic, 723, bridges, and pass transistors in it but I would like to know what is wrong and why. I'm not sure there is anything wrong with it... Or is there? Help!

Mike, KG4MSR
Sep 19th 2014, 13:23


Super Moderator

Joined: Apr 4th 1998, 00:00
Total Topics: 0
Total Posts: 0

Try using a faster scan rate on the scope. You may have much higher frequency noise riding on the DC voltage that the filter capacitor is not taking out all the way. Noise like this sometimes cannot be seen while viewing ripple.

It's possible that one of the solid state devices (diode or regulator) is generating this while under load.

You can always contact Astron and ask what they may think is causing it.

Bob Allison
ARRL Laboratory
Sep 23rd 2014, 03:36


Joined: Aug 8th 2014, 13:19
Total Topics: 0
Total Posts: 0
Thanks for the reply Bob,

I called Astron and they promptly transferred me to a tech who, after hearing the problem described, recommended replacing C1 and C3.

Problem solved.

Back to Top


Instragram     Facebook     Twitter     YouTube     LinkedIn