The Allan Holdsworth "HARNESS" - Part II 

Thanks. Happy to hear that you got something out of these videos! To add to that, I think he may have been trying to simulate the inductance of a speaker moving a lot of air - in other words at high sound levels. Came to that realization after I uploaded part 3.

Maybe in the future. It's as far as I can take it for the moment.

Yes I was and am stoked idk why I love hearing people analyze and discuss circuits

It's not something you could hear, especially on RU-vid. It would probably just sound like a cranked amp. You have to plug one in between an amp you are familiar with, and a cabinet. Somewhere around 1997 or 98, I had tried all the various load resistors, and they all lost something. Your amp just wasn't your amp any more. Touch response was always compromised, too. So I had the Harness II recommended to me. I had a blackface tremolux at the time, but had a rack rig. I'd given up on using the Tremolux as a pre-amp, and had gone throught the Mesa pre-amp, the GT, Kasha, Kitty Hawk and a few others. None felt right. The Harness was the answer, because I could use the Tremolux as a pre-amp for clean sounds, and it kept all of the tone, and *all* of the feel. I used the Harness II for many years, not knowing it was risky. Guess I got lucky. But boy, did it sound and feel fantastic. It didn't add anything to the sound, it just allowed my amp to be what it was, and I could then run it through effects and into a power amp.

Thanks! There may be a part 3 sometime in the future if I round up enough info. Not finished with this thing yet, just at an impasse at the moment.

I look forward to pt 3@@VegasCyclingFreak

Yeah I think there is a very small time delay involved, like microseconds, which isn't really detectable to our ears. Since there's seven inductors, there may also be frequency dependent phase shifting (small amounts). I don't know how to test for such things, not yet anyway. Probably involves expensive equipment that I don't possess. 🙁 The idea behind the Harness is this: This replaces the speaker of a tube amp with a power rating of up to 100 watts. It places more load on the output transformer by roughly doubling the inductance, which forces the output transformer to sag/saturate and the power tubes to work harder. The nichrome wires handle most of the peak current and heat dissipation. This was Allan's way of silently getting a cranked amp sound at line level, silently.

@@VegasCyclingFreak Aye, I appreciate the thorough response. I'm still wrapping my mind around it now. So, he's got the harness between his effects and the tube amp? No no I get it now, the harness replaced the speaker so that he gets a line out and that goes into the venue PA? So even tho much more is going on, it also has to do with the impedance matching for the tube amp, which I understand to be different than the impedance matching with a transitor amp...I think I'm starting to get it. Maybe. The testing you showed so far was still kind of impressive for me. I think maybe it would involve a frequency analyzer. If it helps I can ask an EE , he's MIT so he could probably tell how to test it and he's Guitarist too so he might be stoked honestly. Although there's plenty of EE folks here already maybe they will say how. I imagine that the way the harness affects tone will make it hard to detect all of what is happening. A bit advanced for me to comprehend. Thanks again, this is cool stuff, and thanks for sharing it means a lot when someone puts in all that work and then they put even more work to disseminate knowledge freely. If I had some money I'd give you some, (not that you asked for any.). :) Edit: I missed the very end. He achieves this sound silently? Maybe implying there's another way to go about it that isn't silent? I be a bit tired.

@@SimEon-jt3sr At the time Allan was using the Harness the signal path was as follows: Guitar into Mesa Boogie head, spkr out went to the Harness. Output of the Harness went to stereo parametric EQ, then to eight parallel delay lines, various other things and then to a power amp which in turn went to two pairs of amps, one for clean sounds and one for distortion sounds. The speakers on the amps were mic'd and sent to the PA. This was Allan's elaborate way of getting power amp distortion at very low volume levels. He was not a guitar pedal guy. He, like myself, did not like super loud amps. He liked to get that sound but at low volume levels. Exactly how I like to do things too. In a studio environment he could record with just headphones if he wanted to.

Yeah if one ever comes up for sale again I expect the asking price to be like $2000

Ha ha I'm going with from another planet!

@@VegasCyclingFreak yea man I’ve been fascinated with this mans way of seeing the world for a long time. My high school teacher was friends w/ his wife. It was my birthday so she goes I’m gonna bring you a gift your gonna love it. It ended up being a sixteen men of tain cd singed by him with a little message. I was blown away.

@@ERYKJACKNIFE Super cool! I never got to meet him or see him play in person, but what a well of inspiration he has been. 16 Men Of Tain is so good!

Yeah, that's a good way to think of those nichrome wire coils. They can easily handle 3-5 times the 100 watts rating Allan gave his units. Don't know for sure but it seems that maybe the inductors Allan used were rated for 100 watts. In my "replica" I used 200 watt inductors with 20 AWG wire. Still undecided about how safe it is to use. In part III I will go over why.

@@VegasCyclingFreak I'm sure you could explain it better than me but as far as the amp/output transformer is concerned, so long as the cold DCR of the unit adds up to around what the speaker tap wants to see, that's a good start. Looking at the schematic of it on your absolutely fantastic website I look forward to reading through more of later, the inductors serve the roll as the primary resistive element so they need to be capable of withstanding the inrush current at start up before the nichrome gets a chance to warm up. 20awg is probably about right if it's you're using a heat resistant enameled wire; the sand off kind, not solderon type. The inductors only need to be rated to withstand this brief spike until the nichrome kicks in to moderate the amount of current draw like a shock absorber to take the burden off of the inductors while in operation. The only concern I'd have is if the nichrome DCR goes too high it would under-load the output transformer at the moment of maximum stress but that's pretty unlikely judging by the heaft of the nichrome coils. And also you don't even necessarily have to test it on a tube amp either. Take a piece of crap solidstate amp or stereo system and see what it takes to kill it or if it kills it at all. I doubt it would but one experiment is worth a thousand expert opinions.

@@WillelmMacguyver I used 200 watt inductors - they are made with 20 AWG enameled copper. I'm concerned about the impedance part of this thing. Per my testing yesterday, it's basically 65 t0 90 ohms between 100Hz & 1kHz. I suppose I ought to try a different method for measuring impedance to verify that my test results are accurate. Long story but I discovered that my LCR meter was giving me false readings on the capacitance so what I said about the impedance in part 2 was wrong. 😒

Yeah I did that in the descriptions... but I wonder how many people look at that. If I'm on my phone I almost never look at the description but if I'm on the laptop I very often do. Hmm... I just remembered there are those cards things I can use. I'll look into it. Glad you found the videos interesting. Part 3 goes online tomorrow. I thought I was done in part 2 but the need for one more video arose.

@@VegasCyclingFreak I started reading your website too. Cool stuff!

Yes, that’s what I was saying… in the simplest of terms.

Yeah it's not a transformer but similar principle. What does not make sense to me is that high frequencies do not appear to be attenuated at all... it only goes up up up from 200 Hz. Completely backwards of what I was expecting to see. Maybe I get these odd results because it's not on the output side of an output transformer?

No idea. If he did he likely didn’t share it with anyone.

Yes. Back in the day you had to get creative. People are frankly spoiled today by simply selecting some preset or "plug-in", but I come from the same era as Allan when all this didn't exist. When you have to work to find and achieve your sound it's a lot more meaningful.

This not the same as a guitar pickup, it's not microphonics either. "Potting" it won't do anything to change it. It's called "mutual inductance", where the field generated by the current going thru one of the coils influences the field of the other coil. It's likely a very small factor in the overall scheme of things. How you orient them in relation to each other makes a difference and it's possible to even eliminate it, but not possible in a one rack space configuration. This page here goes into more detail about it: www.audioholics.com/loudspeaker-design/inductor-coil-crosstalk-basics

Happy New Year. Maybe this year it can finally happen?

@@VegasCyclingFreak: Hey, there's a community named Providence up above Summerlin and the 215. What do you think about it?

@@carbonsnail014 All I know is that everything here is built as cheaply as they can get away with. That area of the city isn't bad though.

@@VegasCyclingFreak: Right on, thanks. Be safe over there.

I think it's basically acting like one big inductor with a resistive load designed to take the most of the current and dissipate the heat generated.

Line level output looks same as input, just scaled down. A 50 watt amp would make the circuit see about 20V RMS and something like 2-3 amps of peak current. It seems to me that everything I've demonstrated would just scale up with an increase in voltage. Probably the only way to really test it properly is connected to a tube amp output transformer.... I think that's the missing piece in all of this. All the stuff on the other side of the output transformer will affect what this thing does.

​@@VegasCyclingFreakcould you get your hands on a cheap tube amp maybe from a Chinese website, there are some available for around $150, it would be a shame to destroy output transformers. But in the name of tone it must be done!!!!

This replaces the speaker of a tube amp with a power rating of up to 100 watts. It places more load on the output transformer by roughly doubling the inductance, which forces the output transformer to sag/saturate and the power tubes to work harder. The line level out is taken at the midpoint between this circuit and the secondary of the output transformer and is in parallel with the circuit. The line level out goes to whatever you want. If there was no line level output you'd be feeding whatever is connected something like 35-40 volts RMS and you could kiss them goodbye. This was Allan's way of silently getting a cranked amp sound at line level, silently. There's a lot to explain when it comes to Allan's sound processing... in the late 80s/early 90s it was pretty complex.

I tried that. The only thing I could find was a patent by Ibanez for Allan’s signature guitar. I’ve seen a copy of instructions that came with one of Allan’s custom units that said “patent applied for” numerous times… hard to know if a white lie or if he just didn’t follow thru on it.

I don’t know about the capacitance… it’s just what I measured. Doesn’t make a lot of sense to me. Maybe my LCR is confused by this thing? Yeah, the stuff at my website doesn’t include the impedance stuff… I just simulated that earlier this week. I think I need to connect this thing to an output transformer to see what it’s actually doing, or devise some other kind of test(s).

@@VegasCyclingFreak Looking at the circuit diagram, you've got a pretty simple circuit. about 7mH in series with 5 ohms. That's a lot of inductance: typical voice coils for speakers are not more than a few 10's of microhenries. This is because the inductive reactance at say 1KHz is a whopping 50 ohms due to that 7mH, swamping out the nichrome wire resistors. Basically the load is entirely reactive over 1kHz. If there is a mystery here, it's what AH was contemplating with this design choice. I'm pretty sure you won't hurt anything by testing it on an amp, as basically you are running the amp open circuit at higher frequencies...

@@kpnconsulting8739 Interesting point you bring up there. Seems that most 12" guitar speakers have a voice coil inductance between 0.1-ish mH and roughly 0.5mH. With 7mh x 7, that is nearly 5mH, plus the total apparent inductance goes up about 40% when they are placed on the baseplate & get the screws and washers attached. Maybe in Allan's unit it's lower, but the DC resistance on each of my 200 watt 1mH inductors is 0.73 ohms. It seems that the nichrome was mostly there for the current handling/heat dissipation capacity. Some of the other "load box" type schematics I've seen have up to a 50mH inductor. I can't really compare Allan's circuit to anything else out there... it's pretty unique. It would have been so interesting if he could have done a deep dive on the whole process of how he came up with the design and settled on those components.

@@VegasCyclingFreak It just now struck me what AH was doing. The circuit is a matching network to get you from the output transformer secondary to your line input. I just unplugged the speaker from my 30W Marshall Artist and measured the inductance of the output transformer secondary. 6mH! Right on the nose. So he's matching the resistance and the inductance of the output transformer with the "harness" resistance and inductance. Then he's looking at the divided down voltage drop across the stack as the output. V/ clever.

@@kpnconsulting8739 Eureka! Hmm... that makes sense. So it's mirroring what the secondary of the OT is doing with a line level output arrangement in parallel? Almost seems like it's turning the OT into a back to back transformer arrangement of sorts, thru the mutual inductance coupling of the inductors in the Harness. This doesn't make the OT work harder? I spoke to an owner of a custom Harness. He told me that you can't use it for extended lengths of time... too hard on tubes and output transformers. He apparently burnt up the OT in his Blues Breaker. That was his experience anyway.

lol I don't have the $10K of equipment needed to go after it to make it sound good. Not the purpose of the video anyway. May happen in a part 3, still be determined if that happens or not. Likely but I can't say when.

Well it was output of a Mesa Boogie head into the Harness, then into parametric EQs, then to eight delays, and then to a stereo power amp. I don't know if he always did it like that but seems in early 90s that was his rig. It got a lot simpler in later years.

@@VegasCyclingFreak Very cool info - thanks! The person who wrote that must have confused Allan's Harness and thought it was another amp. For me, he had the best lead tone ever. Best wishes.

Yeah, hopefully I'll get some answers down the road. This is a little out of my wheel house, so maybe I'm not fully understanding what is going on with this thing. This thing has turned out to be much more interesting than what I thought initially. I'm used to peanut galleries, I deal with that virtually every day at work in the consulting engineering world. ha ha

I don’t think he was interested in it.

Frank Zappa might've been. though.@@VegasCyclingFreak

@@sharktroubles Maybe. Frank was a guitar pedal sort of guy. I've never seen Allan use a guitar pedal, the J. Rockett "signature" pedal included.

Yeah that ended up being wrong. LCR meter gave me false reading on capacitance. I will address that in part 3.

😆

Because Rocktron was using the Juice Extractor name for their unit that had Holdsworth's name on it, which had nothing to do with his circuit other than the concept and Allan's name.

There's not many of them but the total hardcore Holdsworth fans that play guitar will be the ones who have them, or possibly a few of Allan's contemporaries. I understand Metallica used Allan's creation at one time, but I am not sure if that was Allan's hand made unit(s) or Rocktron's completely different circuit.

@@VegasCyclingFreak a friend of mine had a rocktron juice extractor and i used it quite a bit helping him set up his rig, seemed to work fine, the extra outs were great for parallel effects (a concept i learned reading Allan's stuff) but obviously i'd be way more interested in trying a harness. i currently use a rivera rockcrusher as a harness and it works great but again i would walk 500 miles to try a harness haha.

@@musterionsurly The Rocktton units are like 1000 times easier to acquire though. I read from someone that knew Allan that he could never dial it in to his liking. On one of Allan’s personal prototypes he had eight outputs on the back, because at the time he was using eight parallel delays.

Does anything to do with a guitar, amp or guitar on YoiTube have to be a bloody sound demo?!?!?! I don't even know if I have this circuit correct. I suppose you would like to watch my amp blowing up but I don't. Sorry Charlie.

By eye. The pictures I have to work with are too low resolution for anything that accurate. I've had someone else look at these pictures, too. They felt like it might be 18 gage. That can be compensated for though. I just need to know what it measures in a real unit that Allan made.

@@VegasCyclingFreak do you know the ratings on the coils, could one extrapolate the winding count from gauge and Henry's if there was a way to get a close enough estimate on the weight of the coils. I'm sure you already thought of that kind of thing. Wow this is quite the under taking. I have only recently even learned who Holtzworth is and this makes me understand just why everyone says he's so major in the guitar sphere. And this project is going to be a main reason why I can now get closer to fully appreciating his style/catalog or contribution.

@@SimEon-jt3sr He didn't use this thing for his whole career, but for at least 4-5 years as far as I can tell. I believe he used this on Secrets, Hard Hat Area and Chad Wackerman's 40 Reasons. Allan used a bazillion different rigs in his career but always had his sound. This Harness thing isn't the secret to his sound but I think it may have contributed to what people feel was his best "tones". Personally I love his sound on 16 Men Of Tain and his contributions on Softworks "Abracadabra" album (you should check out live performance of pieces on that album - Allan's playing is just amazing to me). The inductors in his units were rated for at least 100 watts. In my replica I have 200 watters, only because easily available for reasonable cost. I determined they were 1mH from markings Allan made on one of his personal units.

Yeah about the resonant frequency... I discovered that my LCR meter was giving me false readings - live and learn. Consulting with some people that are more knowledgeable than myself, I now know that in reality the capacitance should be very small, maybe a few hundred pF. Don't really know how to measure it with a reactive circuit like this. Maybe I need a different LCR meter. In any case, the consensus is that the capacitance is so small that it's really just academic. The corrected LTSpice model & actual impedance curve look a lot different than what I showed in this video. I'm currently measuring the impedance at various frequencies, etc. Will go over that in part 3. I have not found a resonant frequency... maybe it's so subtle that I pass right over it.

​@@VegasCyclingFreak A few pF seems very reasonable... I made the same assumption and calculated a resonance in excess of 100kHz, based on your other values.

@@joebasswizard Seems reasonable. No resonant peaks to be found in between 20Hz & 20kHz.

Thanks

Yeah I don't plan on using this but I wanted to reverse engineer it as far as I could. I don't know that the wire would break when it's only seeing 100 watts-ish, but as a manufacturer I guess the concern would be a mechanical connection coming loose and then we would have the situation you presented - no load on the output transformer. And your sound goes with it because it's all in series.

@@VegasCyclingFreak You did a fantastic job reverse engineering and building this. I am very impressed that Allan had the skills to design this himself and have it go to production! Thank you I learned a lot from your video. Enjoy.

@@buzzedalldrink9131 Yeah about that... Rocktron's circuit is totally different but the units that Allan made himself used the nichrome wire. Rocktron unit used ceramic resistors. Anyway, yes Allan I believe was a genius.

@@VegasCyclingFreak agreed he was special

I used a Harness II for years back in the mid to late 90s. Maybe I just got lucky. I used it to load down a Tremolux that had been converted into a head. From there it went into a Rocktron Patchmate that switched in and out various pedals and a couple of rack mount multi-effects, then into a power amp. Curious that they are so risky and Holdsworth used one. He had to have known. Maybe not. All I can say is I had no problems, and it sounded and felt *great*. Far better than all the various tube pre-amps I had tried.

Hey let me know how it works. It's still not 100% verified... had to make some guesses. Also keep in mind he designed this circuit for tube amps with a master volume. I have a hunch he got the idea from messing around with crossovers in hi fi systems, or maybe from tube radios. Your guess is as good as mine!

@@VegasCyclingFreak I was sitting here racking my brain and I think I know where he got the idea from. When I was a kid back in the 90s my pop pop had a old top shelf hifi in the basement my nana wanted out. Long story he was a repairman, chief engineer of a b24d liberator in the service, and I got all of his college books from the 40s which is how i learned this stuff. Now back to this hifi, this one was stereo AND component so it had 2 12s that have almost like a felt cone, 2 5 or 6s with bakelite covers and I believe the other was a 3" speaker in there too. Now what I remember the smaller speakers were hooked to a central board and I remember it had wound bobbins on it I think were cardboard like a sewing spool and there was quite a few. I have the speakers out of it and this weird spinning reverb air funnel thing.I know that back in the day guys would take the 12s out of those hifis cause they have incredible breakup at low volume. Looking back on it now, that board was probably to balance out the speakers in terms of load and volume/presence so all 6 speakers would have around the same spl output and would sound like one big full range unit. That's also what I suspected the weird air funnel thing is. It was designed to move the air in the cabinet out and filll the room with sound. This was all cutting edge back then so they were trying new things. Figures the only other guy that's seen one is another wacky junk collector named Allan. Guess that makes me Alan Depreciationovertime

lol I'm not telling people to go out and make one and it's not something that I will ever use. This is documenting the process of reverse engineering something with very little info to go on. You might see it as OCD but I want to get it right. In my line of work, when you don't get it right you might end up in a lawsuit. Welcome to engineering. Also, this is done in the spirit of true DIY. Not the kind of DIY today where you just buy a kit and assemble it. Yawn.

lol cornductor? Did I make a spelling error somewhere? I'm just a dumb mech engineer that knows how to push air thru ducts. 😁 I think the way to get some more concrete answers on this is to put it on the other side of an output transformer. Stay tuned....

@@VegasCyclingFreak Can't wait! "Remember, folks - no flipping!" 😁📺

I believe in part 3 I go over this capacitance thing. We are talking about a total of only 10-15mH or so of inductance in this circuit, and not very many turns of pretty heavy gage wire on each inductor. That's a very tiny amount of inductance and it will have almost no self-capacitance. It would only be picofarads, probably less than 100-200pF (and some think I am way high in that estimate). Same with the nichrome wire, almost no self-capacitance there with a total DC resistance of only around 6-7 ohms and an air core. Allan's unit is pretty much a pure resistive load. No LRC circuit like most of the "attenuators" made today. Closest thing to it is the original "Power Soak", but that uses a truckload of ceramic capacitors. Actually is closer to the "Altair" which I believe used the same nichrome wire as you'd find in a toaster. Still trying to see what the inside of one of those looks like. Anyway, if anything, the steel baseplate will slightly cut the higher frequencies but it won't be anything dramatic. It would be like the difference between a 10 and 20 ft guitar cable. It appears that Allan used both steel and aluminum baseplates, and some of his units were in an aluminum enclosure and some were steel (as far as I can ascertain). Also some of them were "custom" units, with volume and low/high tone controls. That tone control scheme Allan came up with is simple, clever and very effective.

@@VegasCyclingFreak This is one of the oldest errors made with impedance measurements. Any misunderstanding of which component is the dominant contributor to the impedance will lead to large errors in the other components. E..g. misunderstanding dominance can turn a 20uf capacitor into a 100pF capacitor easily. The exact same thing was made in papers in a related field by a gov researcher who were just a bit sloppy with impedance measurements yielding huge capacitances as a result. This turned out to be false as I proved later on and allowed me to file patents solving a 50 year old problem. This was a highly qualified individual. I am sure you make the same mistake. Your actual transfer function measurements shows that you are making a mistake, and to me it is clear that your impedance measurements are wrong. So which impedance/LCR meter did you use to measure the output impedance as in your video where you got these huge capacitances ?