jamesmking

For anyone building a blue hawaii, here are the current draws for my build after 5 minutes of power on, rounded up to the nearest mA. These figures provide no margin for variation, to headroom for inrush on power up and should not be considered minimum specs for the transformer... PSU board set to 400V output (actual around 404V) constant current on the amp boards set to 20mA per valve inputs shorted to ground no headphones attached groove tubes el34 kemet long life 470uF caps @ 550V for B+ and B- Boards used: amp Blue Hawaii kgbhver6sbipolarinc PSU kgsshvpssicfetdual2new (single board version 1.6 with no delayed start cpc1117n that appeared in ver 1.7)total for both channels: -15VDC line draws 10mA +15VDC line draws 11mA +400VDC line draws 81mA -400VDC line draws 128mA EACH 6.3VAC line draws 3.1A Primary side current draw 0.67A @230VAC All Measurements made on a Brymen bm869s multimeter. NOTE With @230V primary the inrush current is significant and will blow a 3A fast blow fuse on power up, and blow a 4A fast blow after half a dozen power ups (tested with no B+- delay circuit). With a delay circuit 3.15A Timed fuse looks good. transformer I used (which is overkill, but as a result the transformer generates very little heat. ) 330VAC 300mA 330Vac 300mA 36V AC centre tapped 200mA 18v-0-18v 6.3VAC 4.5A 6.3VAC 4.5A Now I have measurements I would say that the transformer could be scaled down to somthing like 330V 250mA 330V 150mA +18V centre tap -18V 50mA 2x 6.3V 4.5A I hope this post will help someone. regards James

Here’s the problem, when you measure the wrong thing the wrong way because its relatively easy to measure and analyse you end up optimising the wrong thing the wrong way... and end up with worse ENJOYMENT of the sound. But objectivist don’t care about the actual sound they care about the measurements of simple, fake not existing in nature sounds like 1Khz pure sine. If they don’t know how to measure it e.g. real complex music, it does not exist... what you actually end up with is products where the enjoyment does not exist because the music is not priority. THD + noise has lead to DACs that disconnect/mute their outputs when then recieve an input which is a stream of 0s i.e. silence. As a result the noise looks realy low with silence as an input so the signal to noise ratio will look good. But you are not measuing the noise generated by the device! the output of the device is not connected to the rest of the device. All because the designer of the chip wants to have better *measured* signal to noise but not actually better real world noise when there is a musical signal present... again measuring the wrong thing the wrong way AND cheating to make it look even better.[https://www.evaluationengineering.com/instrumentation/article/13011757/testing-audio-adcs-and-dacs] Now its recognised the designers cheat by recognising test signals and so the measurement devices dont use silence rhey use an idle tone.... (which are still simple tones that dont exist in nature but exists just to stop the dac muting and are simple enough to be easily filtered out... eventually the dac designers will put logic in the chips to sense these idle tones and still switch on the mute circuit)... then the measurement comunity will need to devleop new idle tones.... I see no point comparing devices with 140db signal to noise vs 135db signal to noise at these kind of levels it tells you nothing at all useful about the sound enjoyment. I remember the very early days of CD.... A rich friend read a hifi review of the first Philips player and purchased it motivated by specifications and measurements alone. He demoed it to me and the first peice he played was classical with violin(s) (I am afraid I dont remember the peice of music only the effect it had). The hasrshness, shrillness and sterility of the sound almost took the enamel off my teeth. I said to him that’s unlistenable. He said the specifications of the CD player and CD system are far superior to LP and the CD is more ACCURATE. He even contended that violins sound like that in real life. I said to him I don’t care if everyone else thinks that’s how violins should or does sound; I literally could not stay in the room with the music playing. I stuck with my LPs and ignored CD for about another 20 years. It was so painful for me there was almost zero musicality. Next time I visited he had sold the CD player and was back on LP. he admitted he could not live with the CD sound. I’m not saying ALL digital replay is terrible but my experiences of the *first* CD systems were uniformly frightening. Now we both have digital systems. But I have only been a serious digital listener over the last 15+ years or so. For me iits taken digital a long time to the musicality levels of LP *regardless* of the measurements showing it being "better" from day one...The bottom line is we need measurements that actually correlate to precieved sound enjoyment and we simply do not have that.

I’m sorry, I don’t want to start a flame war and I am not against measurements, they can tell you if something is broken or almost breaking. For example, if it has 50% distortion or 0 signal to noise ratio and it’s not a guitar amp then its probably broken... But I don’t think I am as sensitive as a measurement meter I know I can’t hear a 0.5db difference or the difference between 0.0001% distortion and 0.000001% and yet personally, I have never purchased a piece of hifi *only* based upon the measurements and would never want to. I buy hifi to get *pleasure* out of listening to music and until someone can *model and measure" how my pleasure system from my ear to my brain works I will stick to trying to please my ears, emotions and mind by listening subjectively. The bottom line is most measurements use inputs that as so simple they simply don’t exist in nature e.g. single frequency repetitive sine waves... why - because it makes the analysis simple. Analysis of actual music being played... no way impossibly complex... As sure as heck I can listen to a lot of hifi that measures well and my brain says very quickly that it is NOT musical or lacking in bass or has glassy hard nasty treble. The device is a non-starter for me after that unless there is no alternative and even then, I will never be happy with it. What are the *actual **measurement* *differences* between say a blue hawaii and a stax srm-006 that *show* how much better the blue hawaii is? If you want an amp to measure well do what the Japanese did in the 70 and 80s that decimated the UK hifi industry... use as much feedback as possible to iron out all the nonlinearities, convince people to stop listening and only look at the measurements and complete just on measurements and watch you hifi industry die... The hifi industry responded by claiming reviewers with golden ears could hear these measured differences... the fact was ordinary people could not and they bought the garbage sounding stack systems and the uk hifi industry and ultimately hifi press and hifi shops died. Im not saying the reason the hifi industry died was solely due to measure and don't listen culture but it was a major contributing factor. I am not against measurements *in* *theory*. I would *love* someone to come up with objective measurements that will tell me that *I* *will* *enjoy* item X Y percent more than item Z. Then I can make value for money judgements and not waste any more money on kit that initially seduces me because it has more detail or more dynamics or more whatever, but ultimately I find myself listening to less music and enjoying what little I do listen to less... that’s not what I want from my hifi and yes I have made some horrendous hifi buying mistakes over the years. Its taken me a long time to understand that more enjoyment is more important than more X or Z. (regards james already starting to dig a bunker and put on a tin hat). My apologies if I offend anyone and you are welcome to think I’m a complete nutcase.

I second the good iron.... I bought a cheap iron with crap thermal regulation and little thermal mass... so as soon as it touched a joint it would cool down. So I increased the amount of time the iron was in contact with the joint.... this increases the time the joint has to oxidize, the flux in the solder would all burn off and the joints were horrible. The solder would not stick to the oxidized joints and it was frustrating. Some years ago, I purchased a hakko fx888d and instantly my joints improved thanks to the good thermal regulation and thermal mass I only needed apply the heat for a short time. So, the flux built into the solder not run out before the was made and cooled. (There is a wealth of good information on you tube) I high recommend the eevblog I would also say a good desoldering station like a duratool d00627 makes like much easier. It has a built-in heater and vacuum pump. You place the tip over the joint, wait for the joint to melt, hit the trigger and the solder is gone. Its far easier and more reliable than trying to hear the tip with a soldering iron and use a spring loaded pump in your other hand, or using solder braid. It gives you the confidence that if you screw up (and I have multiple times) you can fix the mess easily and without lifting tracks off the pcb. Its so useful. I had a problem with the 400V- on my alpha centauri power supply. I figured it had to be one of the transistors, so I desoldered all of them on the -400V side tested each one found a transistor that had decided it was two back to back diodes, put all the other transistors back, replaced the bad transistor and had the psu up and running in half an hour.) I also second the good multimeter with input protection in all modes is also a must. I have this bad habbit of working on amps after work when im tired so I make mestakes... I managed to have my multimeter in ohms mode and tried to measure +400V.... I got a spark just before the probes touched the screw terminal. The psu went into current limit mode and the multimeter survived... a bit of a drama but no harm done. A cheap multimeter without input protection would probably have died. I have a brymen bm869s, its overkill for features and accuracy but its very robust . The brymen being abiused Regards and please consider DIY... James

you should not have any trouble with the amp then.... (the SS in KGSS stands for sharp silcer depending upon how sharp the heatsinks are...? )

I also have a BSc and MSc in computer science, but for the last few years I have been lecturing Cisco networking and computer network security rather than programming. If you have the time and the space, DIY is a lot cheaper than buying... My total cost for the blue hawaii build as less than 1/4 of the price I could buy one for new. Even second hand I probably could not afford to buy one. (OK the case will not look as nice and I did not go for a crazily expensive volume pot, but I could add one later). Apart from the satisfaction when its built and working, I know that in the future I can upgrade/modify/repair/service the amp. The blue hawaii is only my second high voltage DIY hifi build. (before this I built the hi-amp alpha centauri and have built a DC electronic load and a few other low voltage DIY projects). I didn’t hate electronics but I don’t "get it" the same way I "get" programming.: I can write programs from scratch in my head... I can't design more than a simple circuit full stop... although I am getting better at understanding and reverse engineering other people’s circuits and modifying them a little bit... The alpha build was tough, there where errors in the pcb, silk screen, component list, I killed a transistor adjusting the psu, the psu specs were marginal, as was the temperatures on the DC valve heater transistors. I think so few people built it that the information did not get debugged. It took some problem solving to get it working but it was a learning experience. In contrast the blue hawaii has been almost plain sailing. Enough people have built them that any errors seem to have been ironed out. The psu worked first time as did the right channel. The left channel hand one issue which I spotted with my magnified visor: a almost solder bridge which would arc at about 65V. Trivial to fix and the amp now works fine. My biggest issue with the Kevin Gilmore designs was getting my head around all the terminology used in the filename of the gerber files used for the PCB creation. I sent a few private messages to Kevin Gilmore and he was very helpful and responsive. To sum up I’m so glad I decided to go DIY, I have learnt a lot and the satisfaction outweighs the frustration when things go wrong.... (just dont ask me to confirm this when im in the middle of something being wrong ) regards and sorry for my rambling James

The blue hawaii is going to be too large for you, I managed to get it into a single case but that case is 400mm deep, 85mm tall and 19" rack mount wide. It needs a lot of heatsinks and the power supply is large.... and the wiring needs to be carefully routed to avoid hum from the rather large mains transformer (120mm diameter, 65mm height) in my case... The alpha centauri is not much smaller - same case but 300mm deep Are you buying pre built or DIYing your amp? It looks like you dont have enough space for valve... the tiny 6cg7 in the stax amps have very little drive, all the better valve amps use more poweful and therefore larger and more power demanding valves like 6sn7 (alpha) or el34s (hawaii).... But that means a bigger psu, bigger transformer and more heatsinks.... Personally I think the stax'es own modern energizers sound absolutely terrible in every respect. They don’t have enough drive to drive the headphones properly and as a result the bass is just muddy and flat and gets worse as you increase the volume. I had a stax srm006t and 404 signature combo and found them to be bright, fatiguing, not very musical and poor bass. I simply did not enjoy most music with them. I felt that my brain was on over drive and I could not relax into the music the way I could with my electrostatic main speakers. I spent a lot of money (for me) upgrading to a set of sr-007a and ran them on a stock stax energizer and the sound was better but I was still never happy with the sound. It was still a bit flabby, lacking in warmth in the bass and got worse as the volume increased. I felt the upgrade from 404 sig to sr-007a was bearly worth it. I felt my Quad 2905 speakers and Lemma Hydra amp sounded way better, more relaxing, better bass and far more musical. I tried parametric eq to try to make the stock stax amp sound more musical and failed, if I increased the bass all I got was more mud and the mud got thicker sooner, it sounded different but not ultimately better. The issue was the amp design. I built a alpha centauri stax amp http://www.high-amp.de/html/hybrid-amp_eng.html and found it to be a *massive* improvement on any current production stax energizer from stax. Beter and deeper bass, more control, less flabby and muddy, far far better dynamics and stereo imaging, nicer trebble - it was an improvement in every respect except size and power consumption - neither of which I cared about. My only criticisms of the alpha was that I still could not relax into the music and the bass was not rounded or warm enough - it went deep but the mid and upper bass was lacking. I also found that there was very little depth to the stereo imaging. It was far better but still I could not fully relax and just let the music wash over me... still my brain was analysing the sound. I just finished the electricals on a blue hawaii build... case is still to be machined and milled but I have had two days to listen to it. My word its good! without any audiophile components except for cardas chasis wire and ptfe tube sockets it sounds amazing. Its as much of a improvement over the alpha as the alpha was over the stock stax. The blue hawaii is fantastic - I can just listen without strain, without my brain trying, the detail is amazing but its integrated and does not just throw detail at you. The treble is sweet and the bass is rounded, warm and full. It sounds so much like my quad 2905/leema hydra combination that I seriously wonder if its worth running a conventional amp and speakers anymore. I tried my sr-404 signatures on the hawaii and it made the signatures sound very good too - better than the sr--07 on the stax OR alpha! I am convinced that most of the stax headphones are not bad but the stax stock energizers are just complete junk. Anyway, sorry for the rambling. But I think what I am trying to say is that to get descent sound from the stax headphones get a realty good energizer, then later you can think about upgrading the headphones. Upgrading the headphones on a stock stax energizer gets you better sound BUT all the weaknesses of the stock energizers - the lack of drive remain. I have not built or listened to an KGSS but my thoughts are that if its close to the blue hawaii in sound then go for it - and there is not much reason to think it will not be given the simularities in the circuit, power supply and the designers ear. The blue hawaii murders, stomps, crushes, destroys and dismembers the stock stax amps. In my opinion electrostatics are not easy to drive and they demand a top flight amp to have good bass and musicality. regards James

Hi, im planning t build a blue hawaii using kgbhver6sbipolar amp boards which has constant current and the kgsshvpssicfetdual2new power supply - which does not have delayed start but I will make my own... I am trying to get a reasonably definitive answer (if any exists) on the transformer specification... the power supply board and amp baords say on the silkscreen 2x 6V 4A for the heaters, I purpose 2x 6.3V 4.5A for the transformer 18-0-18V for the low voltage I purpose 0.2A.... is thie going to be sufficent??? 2x 360V for the high voltage. It has been suggested to me that if I am going to run at +-400V DC then 330V is fine for the transformer... so I purpose 330V 0.3A is this going to be sufficent? regards and thank you james

updated soft start, slightly shorter with 3 more standoffs and a vent hole for the 1n4007

Thanks Kevin. Im sticking to 6.3VAC non directly heated

Well spotted I didnt notice that, your amp board is slightly different from my one... I wish there was version numbers on the pcb for the amp board... my board comes from the file kgbhver6sbipolar.zip. your board has 5V ac for the heaters rather than 6.3... so I think it might be the directly heated triode version of kgbhver6sbipolar.zip which is kgbhver6sbipolardht.zip..... ugh howmanywordscantheyputonafilenamebeforeusingunderscores??? maybe this thread needs renaming to "blue_hawaii_newbees_trying_to_understand_the_different_versions"... ?

Yes that is the blue hawaii amp board I plan to use. I have no checked component availability for it yet. I want to build the PSU first and have ordered all the parts fot eh psu. I have 5 psu and 5 amp boards on order from JLPCB, the boards will be black and with gold pads ? . Its too late for me to cancel the psu order so my psu will not have the cpc1117N for delayed start ? . But I plan to put a 555N and relay before the 360VAC inputs and implement delayed start that way - its the method my alpha centauri amp uses. I have updated the 1.7 psu board to have cpc1117N soft start, I did it *late* last night but have not had a chance to check it yet... Im off to rock climbing in a few minutes. You are welcome to the gerber files if you want them, if you can check them even better.... if they are correct maybe the stax mafia could add them to the archive.... (I think its only fair us dual psu people get the same functionality as the singles) ? ? ..... I also have a reverse enginer of the circuit for the amp board... again not checked.... if someone wants to check the schematics maybe the stax mafia would be interested in them too? (when I do a build I like to have circuit diagrams)... makes troubleshooting far far easier. regards James This is my reverse engineer of the amp board thuytn posted. it is NOT checked. This is the version of the Blue Hawaii Kevin Gilmore recommended I build....

Thanks, I found it, v1.8 has the cpc117n the board I reverse enginerred is v1.7... There seems to be no dual rail single board version with the cpc117n v 1.7 is the latest I can find. But looking at the differences in the +400 "right board" v 1.7 and 1.8 the only differences are the cpc117n and a few component moves to fit it in. so it should not be too hard to modify the version 1.7 dual to 1.8...

Thank you for the quick reply. This is a more or less direct reverse engineer of the psu pcb. I have only just completed it so there may well be errors....

You may be happy with a basic stax system but the problem is that if you are not used to electrostatic sound (I also run Quad 2805 electrostatic speakers) there is a big difference in sound to most moving coils. The electrostatics offer very good transparency and speed but for me bass and overall warmth has been an issue that seems to require throwing money at to solve.... I have a stax srm006T and sr4040 which I found to be rather bright and the bass was rather mushy. As the volume is turned up the bass gets mushier but no louder and the treble just dominates. It sounds fast and detailed this way but I am very sensitive to poor bass.... I managed to buy a pair of sr007a second hand but could not afford to upgrade the srm006T. to an srm007T. The overall ballance was a bit better than the sr4040s and there was a more natural, less bright and hard sound, but the bass was still mushy and it just ran out of steam even quicker - I guess it could not drive the sr007a... I built a alpha centauri (http://www.high-amp.de/html/hybrid-amp_eng.html) earlier this year - The alpha bettered the srm00t in every way It does not run out of steam, macro and micro dynamics are better imaging is better - it was my first serious diy build and I learnt a lot from building it. I was especially grateful that the power supply has full short circuit protection on the +-400V and was extremely robust.... I managed to "measure" the +400V supply with my multimeter in resistance mode ? . Some sparking at the test probes and the 400V went into over current protection... no damage done and months later its still working fine... (the multimeter survived too - a brymen bm869s which can handle 1000V in any imode). The designer of the alpha claims that you can even short the +400 and -400 together and the psu will survive. My only issue with the alpha is that the sound stage is rather flat and has no real depth and mid and upper bass is a bit lacking. . I read reviews of the blue hawaii and the descriptions made me think it would be my kind of amp - but there was no way I was ever going to afford to buy a prebuilt.... So I’m up for building one... To summarise, from my experience the modern stax amps from Stax are not good - and they certainly do not have a good reputation online. I think they are over-priced especially given the build quality - thin bent steel. I don’t think the 6cg7 has enough drive to get the job done. The 6sn7 of the alpha is better and I am hopping the el34 of the blue hawaii will be better still... If you have the time, equipment and skills to build a diy amp I highly recommend it. You can get far better sound for far less than the commercial offerings and learn a lot in the process. regards James

Hi all, Im planning to build a blue hawaiii using the version 6s bipolar amp boards (kgbhver6sbipolar.zip) and the kgsshv dual 2 new power supply, (kgsshvpssicfetdual2new.zip) Gerber files are in https://drive.google.com/drive/folders/0B_iJFfZStuVhSE5nOHBVdTByR1k Has anyone tried this combination? will the psu provide sufficent current (the voltages are fine +-400V -+15V and 580V) Does anyone know what ratings I need for the transformer Im guessing for the secondaries 2x 6.3 5A (el34s draw about 1.5A each plus I want plenty of headroom) 2x 360V 230mA 18-0-18V 200mA Does the kgsshvpssicfetdual2new offer delayed start for the +-400V? Is it robust and have any form of short circuit protection on its outputs? Sorry to ask so many questions. I could not find circuit diagrams for the kgbhver6sbipolar or the kgsshvpssicfetdual2new.. so I have been reverse engineering from the gerber files... which is not fun and a slow process... regards James