Clarification on Blue Hawaii Power Supply

[alex_o rules]

I recently purchased a Blue Hawaii (DIY) and it looks great. I am ensuring that all is well, and I am performing some measurements on the amplifier prior to plugging in my headphones (SR-Omega.) The amplifier was built in a single chassis and uses Headamp's boards from 2004. One thing I noticed while examining Kevin's project page for the Blue Hawaii is that it calls for +-450v and +-15v outputs on the PSU. However, the Headamp KGSS/Blue Hawaii (as labeled) supply boards seems to output +/-360v and +/-15v. This seems to be closer to the KGSS specification. Does this throw off the amplifier sections in any way? Finally, when I measures from Bias output to Ground on the supply board, I measured ~260v. My feeling is this should be ~580v for Stax Pro bias. I am able to change this measurement by using a 1M trimpot on the supply board, but am worried that something is amiss. Keep in mind I took all these measurements without headphones attached. Moreover, when I plugged my HE60 in, they seemed to play music fine. Mind you, I have never heard any other Blue Hawaii amps, but found the bass somewhat unimpressive. I'm going on memory versus the SRM-007t Mark II.

[justin]

The Blue Hawaii article says +/-400V

The board may be labeled +/-350V because that was the voltage for the KGSS, which is what the boards were used for before the Blue Hawaii was ever built -- if your Blue Hawaii has one of my boards in it. If you are actually measuring +/-350V from the boards, and this is a DIY amp, then the person must have built it that way for any of the following reasons

- Because the board was labeled for a 50V zener diode and they didn't understand what they were doing

- Power consumption/heat was too much at +/-400V

The bias voltage will measure low because the 5 meg resistor in series with it forms a voltage divider with the input of your multimeter. You have to measure before the 5M resistor. Goodbye headphones...

[alex_o rules]

Hi Justin,

The amplifier was built by Naaman, so I suspect he knew what he was doing. Mind you, I do not know how many hands it passed through to get to me! With that said, why did you say 'goodbye headphones'? Also- how would +/-360V affect the functioning of the amplifier? Is the Zener's value the only factor in controlling the output of the power supply section? Lastly, is your KGSS/BH power supply more similar to Kevin's BH supply or KGSS supply? Are they much different?

Thanks!

Trevor

The Blue Hawaii article says +/-400V

The board may be labeled +/-350V because that was the voltage for the KGSS, which is what the boards were used for before the Blue Hawaii was ever built -- if your Blue Hawaii has one of my boards in it. If you are actually measuring +/-350V from the boards, and this is a DIY amp, then the person must have built it that way for any of the following reasons

- Because the board was labeled for a 50V zener diode and they didn't understand what they were doing

- Power consumption/heat was too much at +/-400V

The bias voltage will measure low because the 5 meg resistor in series with it forms a voltage divider with the input of your multimeter. You have to measure before the 5M resistor. Goodbye headphones...

[justin]

Have you measured the PSU voltage or are you just going by what the board says?

Building it at +/-350V would result in less max voltage swing. Yes, the sum of the string of zener diodes controls the max output voltage. However, if the transformer secondaries are not high enough to support +/-400V you would end up with an unregulated power supply. Also, if you do change the diodes you'd want to make sure that the power supply capacitors installed are OK for the higher voltage (they would need to be at least 450VDC caps)

What i meant was if you adjusted the bias supply so it measures 560V at the headphone jack, with your voltmeter that's either a 5M or 10M input, then you would have actually have set it around 1000V (which i dont think is possible) and I assume you could easily damage the headphones

[alex_o rules]

I measured the PSU output voltage to be +/-360V approximately.

I will examine the string of zeners to see if they are 150+150+100 or 150+150+50. I will also take a look at the caps, and the transformer. It appears to be a custom victoria magnetics torroid, but I'll take a look at its specs.

Finally, with regards to the bias, I am feeling a bit thick here. So, without headphones plugged in, measuring before the 5M to ground I should measure ~560V? (Isn't pro bias 580V?) I should then use the trimpot to adjust.

Have you measured the PSU voltage or are you just going by what the board says?

Building it at +/-350V would result in less max voltage swing. Yes, the sum of the string of zener diodes controls the max output voltage. However, if the transformer secondaries are not high enough to support +/-400V you would end up with an unregulated power supply. Also, if you do change the diodes you'd want to make sure that the power supply capacitors installed are OK for the higher voltage (they would need to be at least 450VDC caps)

What i meant was if you adjusted the bias supply so it measures 560V at the headphone jack, with your voltmeter that's either a 5M or 10M input, then you would have actually have set it around 1000V (which i dont think is possible) and I assume you could easily damage the headphones

[alex_o rules]

Ok. So after a little bit of discovery, I can confirm that the transformer meets the required specs for the BH on the secondaries. The caps are all 450V as well. I adjusted the bias pot referencing ground before the 5M, and got 560V without headphones plugged in. Examining the zeners, I was unable to determine if the silk screened 50V was used or the proper 100v was used. Would it be possible for you to provide some insight on determining the zener's value? Also, I have a gut feeling that something is not right in the supply. Can you provide some voltage measurement points to help me determine if there is an issue?

Thanks for bearing with me on this!

I measured the PSU output voltage to be +/-360V approximately.

I will examine the string of zeners to see if they are 150+150+100 or 150+150+50. I will also take a look at the caps, and the transformer. It appears to be a custom victoria magnetics torroid, but I'll take a look at its specs.

Finally, with regards to the bias, I am feeling a bit thick here. So, without headphones plugged in, measuring before the 5M to ground I should measure ~560V? (Isn't pro bias 580V?) I should then use the trimpot to adjust.

[luvdunhill]

Just measure the voltage across each Zener.

[alex_o rules]

Ahhh, so not ground reference. Just one side to the other should be equal to 150v, 150v, 100v if I understand correctly.

[alex_o rules]

Ok. I just measured across the zeners that are not between the heatsinks, and with positive on the cathode side and got ~156V, ~156V, ~51V. I did notice that the 150V zeners had the cathode on the right side, and the other had it on the left side.

[luvdunhill]

ok, hold my hand now...

that means the PSU is working as it was configured.

[nikongod]

I want a hug. Not from Marc. Just in general.

[alex_o rules]

Another observation is that the 4N60B MOSFETs get quite hot (Are these a IRFBC30 equivalent?) whereas the 2SA1968 transistors stay relatively cooler.

[alex_o rules]

I understand that it is working as configured other than the +/-350V rails being ~360V. However, my understanding is that for a BH to perform to spec it requires a total rail voltage of +/-400V and the 50V zener is supposed to replaced with a 100V for the BH.

[luvdunhill]

then replace it

do you also understand what heat sinks are required for it to run at spec?

[alex_o rules]

I am not entirely sure, no. Currently the heatsinks are aluminum and about 2"x1.5" on the mosfets and transistors. They are probably the same ones that Kevin shows in the single box Headamp configuration shown on the BH project page.

[n3rdling]

Isn't that the BH that momo guy from Texas messed up?

[alex_o rules]

Yes, I did purchase it from a guy in Texas, and his email address was momo- something, Is there some history I should be aware of?

[grawk]

I find it extremely unlikely that naaman would have built a blue hawaii PS board at 350v if it needed to be 400v. So either 1) he didn't build it, 2) someone screwed it up after he had it, or 3) you're not doing something right now.

[alex_o rules]

Well, the one I have in undoubtedly the same one in Naaman's pictures. Here are the pictures of the unit sent by the guy in Texas. I of course, cleaned the amplifier so it looks like new. http://img848.imageshack.us/g/img8888.jpg/

[alex_o rules]

Also, to verify my measurements I used two calibrated meters, a LEM LH1025 and a Fluke 179. Both produced the same measurements.

[luvdunhill]

I find it extremely unlikely that naaman would have built a blue hawaii PS board at 350v if it needed to be 400v. So either 1) he didn't build it, 2) someone screwed it up after he had it, or 3) you're not doing something right now.

depends what you mean by "needed"... anyways, I disagree with all those choices.

Edited April 15, 2011 by luvdunhill

[morphsci]

I agree with number three. Specifically AOR is currently only using his head as a hat rack.

[alex_o rules]

I've built amplifiers and other electronic devices before with success. Using a meter to measure DC voltage between two points is well within my capability. It's pretty simple to measure DC voltage between the +/-400V and ground. It's also simple to measure across the zeners. I would say that since I measured approximately ~150V on the two zeners labeled 150V, and then measured about 50V on the zeners in question would seem to imply I did not do anything wrong.

[justin]

you dealt with momo? good luck

[alex_o rules]

I'm getting a bit worried here. Some history please?