In a nice turn of events, my anniversary of 2 years as a member of the DIYAudio.com forum was coincidentally marked with a nice little gift. Nelson Pass and the good people at DIYAudio host an event called Burning Amp annually. It allows people to see and hear designs made by other DIY audio enthusiasts and attend demonstrations and lectures. What makes this event so special though are the building sessions. Here, Nelson brings a project he, or one of his team, has designed and have kits and tuition available so that attendees can make their own equipment under expert guidance to take home. This event was where the famous Amp Camp Amp had its genesis, evolving into a favourite for novice builders including myself. Last year the Whammy headphone amp was the attraction and this year a preamp / headphone amp made its debut. Alongside this came a late addition, a 2nd Harmonic Generator, the H2. In a spirit of generosity, Nelson donated a number of kits to be distributed for free to DIYAudio members around the world, and I was lucky enough to be one of the recipients.
So, what is the H2 and why would I want to build it? Nelson Pass wrote an article about the H2 and published it on his First Watt website. Here, he acknowledges that the key to fidelity in reproducing sound has been to reduce distortion and that this has largely been accomplished (in the realm of electronic hardware at least). The argument runs that this historic process of ever decreasing distortion has left some enthusiasts with an unsatisfactory experience, the supposed ‘clinical’ sound of modern amplifiers, and a nostalgia for the distortion inherent in tube amplifiers. The H2 then is a response to this, a way of injecting a supposedly euphonic 2nd order harmonic distortion back into the signal chain for solid state amplifiers.
Whilst I am of the opinion that equipment design should aspire to minimal distortion and faithful reproduction of an inputted signal, I am certainly not averse to altering what comes out of the speakers. There may be many reasons why you might want to do this. I have used tone controls in one of my projects and I occasionally use DSP. However, these are always options, I can still revert to my low distortion signal at any time. I’ve never heard a tube amp, have no idea what 2nd harmonic distortion sounds like and I have never considered building a tube amp. The H2 to me seems like one of things that’s too easy to do to pass up on. The kits were free and postage and packing at cost was all I paid to get a PCB and all the parts. A note at the end of the article sums it up nicely…
This circuit is fun to play with, but inevitably some people will take it seriously, especially if they like what they hear. There will be plenty of people who won’t care for the effect, and that’s perfectly OK. I’m giving these things away, so no whining.
Nelson Pass (2019)
The circuit itself looks very simple with a correspondingly low part count. A few resistors and capacitors, a pair of potentiometers and LM317 regulators and a pair of J112 Jfets which have been matched by Pass, and that’s it. You have to provide your own 24VDC 500mA regulated power supply, RCA connectors and enclosure. Construction is straightforward, but be careful about choosing the correct placement for the Jfets, there are two options depending on pinout (the Pass supplied Jfets go in the lower connections for Q1 & Q2 labelled D S G). The supply voltage has to be set to match the value supplied with the Jfets usings the potentiometers but the harmonic content can be altered from this calibrated value.





The voltage is set by comparing the Vp value of the Jfet against the distortion curves in the image below and reading off the T2 value where these lines intersect. In my case the Vp value is 2.70V, so I drew a vertical line from that value. It intersects with the 1% curve at approximately 15.4V so the corresponding potentiometer is adjusted until the meter, connected to T2, reads 15.4V.

All that remains is to hook up some I/O connectors and the power supply and get it in the system. For my quick test I put it between my Topping D30 DAC and ACA. There was a bit of buzzing going on so it really needs to be in an enclosure, but my first impressions certainly weren’t unfavourable. I’ll get a cheap enclosure sorted soon; I have a few ugly stomp box types that I got cheap when Maplins closed down which will be suitable, I just need the rain to stop for five minutes so that I can get outside to drill them. Once this is boxed up I’ll try it through my other systems with better speakers so that I can get a better idea of what it sounds like.

Finally, thanks to Nelson Pass for making this available at no cost and for DIYAudio member Coolnose for arranging the European distribution.
UPDATE: I have now completed this and put it into an enclosure. More details here.
UPDATE: 8th July 2020: The results of Archimago’s latest blind test – relating to the audibility and possible preference for harmonic distortion – are in. It would seem to point to a preference for lower distortion where people could discern a difference.
Thank you for sharing this experience! Very nice!
Please tell me, how do you arrange these resistors so admirably accurate _in_the_air_ ?
thanks
dave
Hi David,
Thank you, I hope to have it in an enclosure soon.
To raise the resistors off the PCB, I first bend the leads using a tool such as this. Then I just use blu-tack to hold them in place at the correct height. Some people use a pencil underneath the resistor body to make sure they are all the same, but I just do it by sight.
Hi, great work! I love this project. I have a few questions…. 1. Nelson’s original version had one LM317 for both channels. Your new version has 2. I see component values are different as well. So is yours a refibed version or os your version what Mr. Pass called V2?
I’m keen to make this thing. I have a Bryston 14B which lacks warmth, driven direct by a Benchmark DAC-Pre. Its not warm either.
Any clarification about the two versions would be very helpful.
Best regards,
Brad from Canada
Hi Brad,
Thanks for your kind comment.
The revised schematic with the new resistor values and dual LM317s appeared in the article I linked to and Nelson calls this the DIY version 1.
Good luck with the build.