A Practical Milliwatt Mod for the K2

By, Randy Hargenrader WJ4P

I am one of those nuts that likes to run QRPp for the fun of it and Waynes’ "white paper" on the Apps note pages was an inspiration to take my K2 to a "lower level"! Waynes’ mod suggestions involve an op-amp circuit to add to the K2 to get it to go below 100mW. I built this circuit in both gains (3.16dB and 10dB) with some success. The problems I had with making this circuit work had nothing to do with Waynes’ suggestion. In retrospect, it was fortunate that I did have difficulty making it work. In the process of discovering why the op-amp solution wasn’t stable, I found a simpler way to get the K2 to do a /10 power reduction with just a few resistors added to the power control circuitry. This mod uses a double pole switch to exchange the stock resistors with the proper resistors needed for smooth control of the power output of the K2 in the range of 10mW to 1.5W.

I have had this operating in my K2 for some time now and it is quite stable and fun to use. This mod is not for the faint of heart since you have to mount a switch somewhere and that means modifying the case! (And, I don’t have a step by step procedure prepared yet, either.) But, I do have a .jpg that shows the mod in full color and is annotated so you can see what was done. One more caveat – Wayne cautions that the spurious emissions may not be as well suppressed at these low levels. The published specs are for the 5 to 10 watt power out levels. Also this modification is not for the K2’s with the ATU installed.

Perhaps someone can develop a low power modification for that crowd.


R-68 on the RF board is a 220 ohm resistor in a voltage divider involved in sampling the RF volts at the antenna jack. (R67 and R68 form this divider) The subsequent "scaled" RF is rectified by D9 and passed on to the power control arrangement in the K2. The K2 power control algorithm expects to see a certain voltage from D9 for a certain power level out to the antenna. The K2 then "makes" more or less power out depending on what it "sees" from the sense circuitry (D9, R67, etc…) to conform to the power level selected by the front panel knob. By raising the resistance of R68 to a higher value the RF sensing circuitry makes more volts for a lower power out, "fooling" the power out control circuitry into making less power than is requested by the front panel knob. Carefully selecting the correct value for R68 (approx. 1500 ohms) results in a /10 scaling of the requested output power.

The other area that needs to be "helped" with running lower power is the transmit power control circuitry near Q24. This circuit is controlled by a DAC, producing a voltage to Q24 affecting the power out. At lower power levels and on certain bands, the DAC operates in a very coarse range (low numbers) which caused the original op-amp mod circuit and this arrangement to "hunt" very actively for the correct power out to satisfy the front panel setting. (If you decide to do the op-amp version from Waynes’ write up, you will find this area requires attention also to make it work.) This hunting often left the power setting now where near what was desired since eventually the algorithm in the K2 "gave up" and set it as close as it could with the resolution is had to work with. So, by "scaling" this voltage from the DAC by adding a resistor that made a voltage divider in the DAC circuit, we force the DAC to use a higher voltage (and more bits!) to get to the desired power level. More bits gives us better resolution. I switch in a 270 ohm resistor to ground from the junction of R98 and C168 to make the voltage divider "appear" on demand. Note: I highly recommend that you do the "K2 Optional Power Control Mod" from the Builders Resource page under "Mods". This post from Wayne describes getting finer resolution at QRP (5 watts) from your K2. This mod adjusts R98, and my K2 has a 1K ohm in that position now.

Both resistors are switched in by the use of a miniature double-pole slide switch I garnered from a defunct cassette player. For the resistor change at R68 requires a double pole switch with the common point the junction of R67 and D9 as the common connection. R68 and the added 1.5K resistor are connected to ground and switched in as needed for "high" or "low" power. The other pole of the slide switch is used to ground the added 270 resistor connected to R98/C168 when running "low" power. I also added a .01uF capacitor to the switch side of the added 270 resistor to keep any RF that might be riding on that circuitry from being carried across the bottom of the RF PCB by the connecting wire.

Installing the switch was quite unsettling! I was afraid to "butcher" my baby! I consoled myself that I could always buy a new bottom panel if need be. I looked for a spot that would not be readily seen and settled on notching a spot on one corner of the bottom plate. This also put the switch right at the RF power sensing circuitry area allowing short runs and point to point connections. You will see in the .jpg that I moved the RF sensing parts to the bottom of the RF board to make it easier to work with while I was investigating this mod. Also, moving these parts to the bottom allowed me to insert the modification without cutting any lands. I left D9 and other parts there since they weren’t in the way after-all. The added parts to the power control circuitry are also mounted on the bottom of the RF board and tack soldered in place. The "long" run of hook-up wire from this area looked safe since it should be just a switched DC connection.

One last word of advice: When running "Cal Fil", be sure to have the new "power out switch" set to "Hi" power. The internal frequency counter needs more RF than the low power mod produces when it is selected.

Email Randy - wj4p@mindspring.com

K2 mW mod pictures -

Back to index -