KGSSHV Carbon Build Thread

[spritzer]

5 hours ago, TMoney said:

A brilliant little headphone, I agree! It was the surprise of @rdreyer's little menagerie. Beautifully clear and transparent in the treble and mids. I thought it lacked in the bass, but prehaps that is due to it being earpad headphone and not getting any kind of a seal rather than being circumaural?

 

They are lacking in the bass but that was on purpose.  The Stax Pro series (so SR-X and 4070) all trade bass amount for depth and insight into the mix. 

[wink]

14 hours ago, JoaMat said:

Me to find that blue plastic water can very interesting. What worries me most is why Vikings like Soren and me are more interested in plastic head phone stands than the head phones???????????????

Impeccable breeding, perhaps.....   ??    

[insanity]

I just suffered a pop and all the lights of the amp went out. Seconds later I cut power to it. Checked the fuses in the IEC, 2A fuses were intact. Then I suspected the toroid simply because all the lights were out, and not just HV or LV. 

Removed the Toroid and measured all the windings. 

Dual Primary measured 4.5ohms each. So 9ohms against 230V = 25A WTF. Probably wasn't long enough for the fuses?

Secondary 

HV@330V measured around 55 to 52 ohms for the two windings = 6A...

LV@18 measured 1.5 each = 12A...

I didn't measured the resistance of the toroid before, but when I built the amp voltages were good. If I'm not totally brain dead this looks like a burnt through transformer. Confirm?

 

Transformer specs from toroidy were: 

TS200VA PRI: 2x 117V SEC:

> 2x 330V @ 250mA

> 2x 18V @500mA

> 16-0-16V@500mA forget this winding, it was not in use.

Any ideas?

[sorenb]

Was recently asked about the same, someone had measured:

For the 220V input, R=6.2ohm. For the 16-0-16V output, R=0.8+0.8ohm. However, for the 330V output, R=45ohm!

And asked whether it was too high, the ToroIDY I had at hand measured:

For the 220V input, R=14ohm. For the 16-0-16V output, R=(4+4) and 93ohm for 390V windings.

 

Have you measured the voltages from the transformer unloaded?

[insanity]

No I have not yet measured the transformer unloaded. I is too late in the evening to start working with live voltages... ;-)

So far I have only removed the toroid and measured the windings. 

A preliminary visual inspection of the amp boards top sides, has not revealed anything. I will have to do that another day.

 

Btw: do you happen to know if the GRLV boards need a minimal load to properly regulated the voltages?

 

Edited August 22, 2016 by insanity

[kevin gilmore]

grlv boards will run perfectly without any load

[insanity]

Ok then I think I have a bigger problem. I took out an LV board and tested it out of circuit without a load and it measured way above the 15v it should have... 

This sucks... Looks like a major investigation... 

[headinclouds]

Hi insanity,

I have been staring at my computer all day compiling an order for a new project so I am pretty frazzled too..  However those resistance values you've measured don't look so unusual.  The dc values are not the same as the ac impedances. It won't draw those currents when on 50/60Hz.

We await your testing of the transformer

Good luck

PS have you checked the switch? but it must be working if you have trouble in the downstream supplies.

Edited August 23, 2016 by headinclouds
another thought

[insanity]

Looks like I have found the culprit today. There was some charring on the B- header of one amp board... See for yourself. 

To me it looked like there might have been a bit of flux from the B- pad to the trace adjacent to it. I cleaned all the board on the back side as well as possible, but how the f*** was I supposed to get that flux under the header? Do you guys always soak the boards on both sides? But even then, no brushing possible there. 

The good news is that the transformer is alright. Unloaded voltages are fine. 

The bad news is that the amp board is trash (I got two spare boards luckily). Unfortunately the two GRLV boards don't work properly anymore. I guess I will have to replace the amp board and replace all the active parts on the GRLV. 

I still have to check the other amp board. Hopefully there was no damage to it. The GRHV seem to work properly (without load). 

 

FYI the amp was used approx. 50h before this happend and ran about 15min on the day it happened. 

 

What should I do to prevent this from happening with a new board and the other board?

Edited August 23, 2016 by insanity

[jose]

My KGSSHV C is Alive!!!

 

 

[UFN]

Congrats! Can we see some more inside pics?

 

//UFN

[jose]

 

Ok, I know my amps are not very pretty , but I like the simple front. I made the 20mA versión so it need a big heatsinks

Edited August 23, 2016 by jose
photo

[JoaMat]

1 hour ago, insanity said:

.....
What should I do to prevent this from happening with a new board and the other board?

This maybe?

I use this type of terminals. Better isolated to the board.
 

[insanity]

9 minutes ago, JoaMat said:

This maybe?

I use this type of terminals. Better isolated to the board.
 

What brand is that? The ones I used are rated 400V. As discussed earlier in the forum there are no terminal block with higher voltage and the same size.

[JoaMat]

Bought from Elfa in Sweden. Sorry, but I can’t find them on Elfa now.

I’m sure you would find some similar if looking around.

[mypasswordis]

Did you solder the big collector tabs on all the SOT223 package transistors for better heat transfer? I would wick off some of that excess solder clump on the smaller tabs, too. It looks like you use general amounts of solder in general, less would help (including on the terminal blocks). What temperature did you set your iron to?

[kevin gilmore]

so that actually looks like an overheated pin due to poorly tinned wires (or no tinning) or not screwing down the wire tight enough.

Seen this on too much expensive lab equipment. So the pin overheated, then all sorts of bad. The trace it was shorting to is the feedback line and is close to 0 volts

 

[insanity]

I didn't solder the big SOT223 tabs. Is it recommended to do that? Thought of easier removal in case of a problem. I agree that there might be a bit too much solder on the other SOT223 pins, but being small they are difficult to solder. Considering the general amount of solder used on the other joints, I was aiming to achieve "ok" joints as seen here: 

The iron is around 370c or 430c. Don't remember exacty. What is recommended?

@Kevin For the HV lines, I use silverplated wire with teflon, they tin very easily and I doubt that this was the problem. When I removed the wires from the terminal blocks I did not have the impression of not tightened screws. But it's hard to exclude of course. 

[mypasswordis]

It is easy to remove the SOT223 transistors, easier than through hole components. all you need is some wick/braid. Well, maybe not so easy if there is a resistor or terminal block in the way, which is probably what you mean...

Check out Hakko's site also, there are some pics and info on temps: https://www.hakko.com/english/lead_free/pages/

I realized you are in the EU so most likely use lead-free solder, and the Hakko site says around 380C for that. I would go more with the first OK solder joint in your linked pic and also the satisfactory example in Hakko's site. Lead-free solder is more likely to develop tin whiskers IIRC, which further increases likelihood of arcing. That shouldn't have been the problem in this case because they usually take months or years to grow.

I was just wondering about temperature because there are brown spots on some of your joints, which may mean not all of the flux is melting? That wouldn't make sense especially if you set it to 430C, which is way too hot. Not enough thermal recovery doesn't make sense, either, because it's happening even with the resistors and a lot of solder is clearly flowing from the other side. Maybe too hot or too much time soldering the joint?

[kevin gilmore]

probably does not matter on this amp, but soldering the tabs on those transistors on the ssdynalo makes a fairly big difference to getting the heat into the heatsink plane

[Pars]

lead free is bollocks, based on bollocks, IMHO

[insanity]

I didn't use lead free solder. I used cardas quad eutectic.

[mypasswordis]

Cool. In that case, eutectic solder has the lowest melting point, so 370 and 430C are too high. Try cranking the temp down, applying heat for less time, and using less solder and see if your joints look better. Hopefully that means when you solder the new terminal block you won't be leaking that flux residue to the feedback trace again. You really want the lowest temperature that can still work, because higher temperatures mean shorter tip and heating element life, burnt flux, more thermal stress on the components and pads, etc. Also, I don't know what flux you are using, but make sure if it is rosin activated/corrosive that you clean all of it off everywhere you can. I use standard Kester 44, which has somewhat alarmingly strong claims to contain no-clean activated rosin core:

Quote

Kester 44 possesses excellent fluxing ability, the flux residue is non-corrosive and non-conductive under normal

conditions of use. When exposed to an elevated temperature and humidity environment (38°C, 94% RH) for

72 hours, there is no evidence of corrosion caused by the flux residue. Throughout its many years of wide

usage, 44 Rosin Flux has produced many billions of soldered connections. In all these billions of solder

joints, involving the most delicate and critical of electrical and electronic components, there has never been

an authentic instance of corrosion by the flux residue under normal conditions of use. This mild property of

the residue permits leaving the flux on the assembly for many applications./

 

For extreme overkill, you can also use a terminal block with more isolated pins like JoaMat posted or put the block in from the other side if possible, but I'd play around with soldering technique a bit more first because it's just good practice.

Edited August 24, 2016 by mypasswordis

[chinsettawong]

You can also put a thick coat of woman's clear nail paint over the nearby copper traces.  The nail paint is a good insulator.

 

 

[headinclouds]

Remember the craze for kapton tape back in the early KGSShv days?  I did buy a reel like this ,

http://www.ebay.co.uk/sch/i.html?_from=R40&_trksid=p2050601.m570.l2632.R2.TR11.TRC1.A0.H0.Xkap.TRS0&_nkw=kapton+tape&_sacat=92074

Haven't used much to be honest.