Post by NikeofNargothrond on 02/16/26 at 22:46:16
Finished up my ZKit-1 build on wood with practically no hum. A lot of lessons learned during the process and wanted to share some things with the community in case anyone else runs into similar issues building on wood. Some things were silly oversights, but others a little more complex. Keep in mind, some of these items are specific to building on wood with no chassis ground.
1. Be sure to use the most recent Schematic. I've found easily a half dozen various versions online and in the Decware forums. Some things on the original BOM don't match the specs on the Schematic so be sure to compare the two. For example, the original BOM was still using 33uF caps for the first two caps in the power supply, but the Schematic and most recent PCB (revision 2025) have 47uF on them.
2. The original BOM has some parts that are now obsolete or being phased out at the time of this post. I've provided the actual BOM that I used and any notes for upgraded parts or alternatives that I used. Again, many older versions of this floating around out there. Lol. I sure wish it only cost about $300 to build one of these today as it did not so long ago! See attached for my BOM and actual costs.
3. I highly recommend upgrading the 6.8k Ohm 2w Metal Oxide Resistor to a 5w resistor of the same spec. Can't quite tell you why or how, but I had two of them go nuclear and melt down on me. Never had a single issue with the 5w.
4. Upgrade the Coupling Capacitors. I chose to use Mundorf MCap EVO Oil Aluminum as I like those sound properties. The original BOM calls for Cornell Dubilier 400v caps. I bumped up to 650v here as well. This is arguably the single most impactful component you can change to help meet your sonic requirements. Choose wisely.
5. If you opt to bypass the potentiometer entirely, be sure to run a 100k Ohm resistor (or similar) from the signal path to the ground holes where the potentiometer would be. This maintains the signal path ground plane reference. I missed this on my first power up and had crazy oscillations. Once I got my ground plane back in check, the power stabilized enough to move on. I ended putting the potentiometer back in anyways. Be sure to run a ground jumper from the PCB marked "Ground Case of Vol. Control" to the casing of the potentiometer. As Steve calls out in a post somewhere in the forums, the pot acts like an antenna if you don't do this and picks up all kinds of unwanted noise.
6. Once built, I pulled the input tube and had a signal side hum. Found an old forum post where someone ran two heavy gauge solid copper wires from each negative RCA directly to the ground holes where the potentiometer mounted, to their respective channels. You do this in parallel with the PCB circuit. This COMPLETELY eliminated the signal side hum. I opted to use solid core strand of romex I had laying around and ran it underneath my PCB so it wasn't having to dodge around other components. Be prepared for this. You can always use gator clips and place temporary jumpers to test it out. My original hum went from -52dB to -74dB with this alone.
7. Binding Posts - this was a silly miss. Apparently, binding posts come with more than one standard sized diameter shafts. Naturally, I purchased the larger size instead of the 1/8" posts. Mine did NOT fit into the PCB. If you plan to float your binding posts off the PCB, this is not an issue. Lost time here having to get replacements that fit.
8. RCA jacks - again, not a standard pin layout for PCB mounted RCA jacks. Didn't pay attention when ordering mine. There are a few different layouts of how the negative legs and positive leg are arranged. If floating these off the PCB, not really an issue but a heads up if you plan to mount to the PCB. If you do float them, be sure to use shielded wire with a drain to run from the rca's to the PCB. Steve notes in the Manual where to connect the drain/shield on the PCB.
9. Earth Ground - don't hesitate to use a thick gauge solid core copper from the Earth pin on the power inlet to the "Chassis Ground" on the PCB.
10. More Hum Reduction - you may not have the same exact issues or same causes as I did, but I want to share what I did that worked. Once I put a scope and began checking for elevated AC ripples in my power side, I noticed other things that were strange. Eventually, I found that my ground planes were very different from each other. I was still dealing with a -74dB hum at this point. I ended lifting the negative leg of the 150 Ohm Cathode Bias Resistor from the PCB and ran a solid core wire to the negative leg of the first 47uF capacitor in the power node. This brought my audible hum down to -85dB and helped to balance out the grounding floor on that side of the board. Yes, I double, triple checked my bias after this change to ensure the bias circuit was still self biasing correctly. It never even hesitated.
11. Remaining Hum Reduction - Lastly, finding more variance from the ground plane around the power node, I ran another strand of the solid core romex from the negative leg of the SECOND 20uF capacitor (don't forget these two caps are in a series!) to the negative leg of the first 47uF capacitor and my hum dropped from -84dB to nearly -99dB. I had to put my ear next to my speakers to hear anything at all, even then it is very faint and there is no impact to the hum frequency nor loudness, even when turning the volume all the way up on the amp. I am satisfied with the minimal amount of hum I now have.
12. Transformer wires - twist them all. At first, I only twisted the speaker wires for the OPT but was seeing a 240Hz and 360Hz spikes on the spectrometer. Once I twisted all of the paired wires on the OPT, the 360Hz peaks disappeared entirely the 240Hz got much quieter. Interesting. I ended up bypassing the PCB connections for the speaker wires and ran the OPT directly to the Binding Posts. This got rid of the 240Hz peak on the spectrometer. I twisted all of the PTF wires from the start as well.
13. Soldering - use quality solder. I used 37/63 and felt it did just fine. I didn't like the 40/60 I have laying around. It wasn't flowing properly with the PCB. Steve gives very solid advice in the Manual about soldering, read it... twice. Clean and tin your tip often. Nothing worse than trying to troubleshoot a circuit with a pesky intermittent connection giving you strange readings. Keep a copper mesh wick around to help clean up any mistakes or desoldering you need to do. Don't be the person who uses a drill bit to clean solder out of the holes on your PCB... yes, we've all been there at some point.
14. DMM - I ended needing 3 different ones... Ideally, you have electronics meters that can accurately read voltages up to 600vac/vdc, capacitance down to 0.0001uF, resistance down at least 1 megaOhm, a low range in millivolts, and is capable of properly measuring diodes for leakage. Unfortunately, I did not have one meter that could do all these things, but among 3, I sure did. I was sure to check each and every part for measuring to spec BEFORE I installed it on the PCB then again AFTER installing it. I ended up burning up three diodes when soldering them in... they are sensitive to heat.
15. Oscilloscope - for the love of God, get a scope. You can get a budget friendly one that doesn't have to break the bank, but I don't know I would have ever figured out the ground issues without the scope. Using the correct tools to take measurements is always the best way to ensure you are getting accurate readings. I've never been able to accurately measure AC Ripple on a DMM... but I have tried. Lol.
16. Patience. The actual assembly and installing parts shouldn't take more than 4 or 5 hours, even for a novice at soldering, but take your time and double check everything as you go. You can spend hours troubleshooting something that could have easily been caught earlier. I definitely got frustrated a few times with build, but looking back, I still had fun doing and would do it again.
So now we are up and running the amp in and enjoying it immensely. I think I would have definitely chose to build on a chassis had I known all the ground issues I would encounter on the wood alone, but I learned a few things too.
Here are some final pics and a few more details of my build:
- Built on 2x8x22 Black Walnut
Opted for 61NP-EV input tube
Opted for 6P15P-EB output tubes - purchased directly from the Ukraine
Opted for Edcor XPWR-008 power transformer. Same spec as BOM but adds 5v 3a taps and 200ma rating - did this in case I ever decide to add tube rectification
Driving refurbished Sansui SP-3500 Speakers (8 Ohm, 5-way, 99db 1w/1m)
Source is a Wiim Ultra - mostly stream high res audio or Phono
Project Debut III TT with Ortofon Stylus 10 cartridge
Blue Jeans Interconnects and speaker wires
I also found my source output at 800 mVrms sounds best to me on the amp. Here's a scope capture for the output with no clipping at about 0.9w. My amp volume is about 50% up and source volume at 50% as well.
Again, I hope this post can help someone else. Good luck and enjoy!