Mezmerize DCB1 Building Thread

[Salas]

Originally Posted by rokoko

Salas, can you please clarify:

It is a radial type of can, both flying leads on the same side (end), not the axial type of cap?
About the potentiometer-mutistep switch: what DIY-selfers use the most as a good aesthetitac approach and most nice rotation feeling, is it a Alps type 20kOhm potentiometer or DACT, GOLDPOINT Stepped Attenuator?

That capacitor model is as you described.

Alps is smooth rotating with good log taper (volume rises intuitively louder as you turn it up). Its not harsh but not that transparent or focused. For better results you would need to spend on a good stepped attenuator. There are cheap SMD resistor stepped ones on ebay for around $20 also. They sound good but their long term reliability has been questioned some times.

[rokoko]

Those on your picture are the "Obbligato MKP oiler" as you recoomened as DC blocking caps, and would be even beter than axial Gold series?
The pot or switch should be linear or Log. I tought Log is what should be used for volume regulator, but others are using linear.

[udailey]

Hes not suggesting its better, hes suggesting its decent and inexpensive.
Log raises volume slow and then faster. Linear is fast at low volume. Log is best. Linear might be okay with a really low gain amp.

[Salas]

Log. Linear will open up the volume weirdly fast for the human ear loudness perception curve. The Obbligato MKP oil are excellent value for money and they come in big uF. Plus they are not harsh. Can be mounted sturdily with a bracket. There are step ups from there, including their own gold line claimed to be better which is still fairly well priced. There are many much more expensive choices of course. Read around, ask members. We use the DCB1 as DC coupled by majority.

[Renron]

bIG BLACK can! Think 10oz. Red Bull size. I've got them in my speaker XO, and in my tube preamp. Nice warm sound, not sharp. Clean but not crisp like a teflon cap would be. Ron

[McQuaide]

Originally Posted by Renron

An easy, easy extension shaft for the pot is a short piece of rubber tubing (automotive, vacuum or gas line) slipped over the pot shaft with another metal shaft on the other end. You can use a grommet or plastic tube (straw) where it penetrates the faceplate so it's not metal on metal. The knob will hide the penetration grommet. Hobby shops, and hardware stores have all different sizes of metal tubing. Doesn't have to be solid. Set screws on the knob will hold just fine.

[]--==----

pot >shaft>tubing>shaft

These ascii drawings don't usually work out well, but I tried.
Ron

Thanks! Exactly the kind of advice I needed.

[rokoko]

Thanks for all of your advises, I will keep building!

[BackinBlack]

Hi Salas and fellow Mezmerize builders. My first post here, so please excuse any duplication of comments.
After what seems like months of reading through the various Mezmerize/DCB1 threads I finally ordered the pcb, promptly delivered to me here in the UK, components sourced from the likes of Mouser, Farnell and eBay, hours of matching the 2sk170s and LEDs and eaarlier this week I completed assembly of the Mezmerize. Powered up and everything worked first time, but despite careful matching of the LED strings the rail voltages were +10.92 and -10.17, also the strings of five LEDs were slightly different brightness. Measuring the volt drop across the 10ohm resistor at the base of the string showed that the negative supply string wewre carrying 70mA, whilst the positive supply string carried 171mA. A bit of investigation and I noticed that the schematics show that the 2sk170s at the base of each string are configured differently. For a normal CCS configuration gate and drain are connected together, however the schematic shows the 2sk170 CCS for the positive LED string with gate and source connected together, thus not working as a CCS. The pcb is also configured this way. I can't find any reference to this anywhere in te various threads, has any one else noticed this?
Anyway, a slight modification to the pcb and now I have the CCS working correctly.
After about 6 hours of "burning in", current in each string of 5 is now 6.8mA, the strings of 3 have 8.7mA, the power rails +10.18 -10.14, and the output offsets are 0.5 & 0.1mV. I'm very pleased with this!
Next step is to add the Lighter Note attenuator that I've ordered from Uriah Dailey and get on to listening to it. I shall try to get a listen before I add the Lighter Note and report further.
Meanwhile, Thanks to Salas, crt, Teabag all all other contributors for their contributions to the Mezmerize.

[Salas]

Hi, maybe 7.1mA and 17.1mA? its been discussed some times before but its naturally difficult to locate over an old long thread. The sinks for the 5strong LED groups are configured as you found on PCB not by error but on purpose. Its not on the Hypnotize HR though. That was to add some more 2nd harmonic in the Mez as it was designed to run at only 60-70mA and it was found worthwhile tonally. It creates a steady difference in rails also that helps the not very well matched audio JFET pairs to show less offset when tried around in each pair. Its easy to run it in ''orthodox'' mode as you found, and it has been described there in the old black non HR Hypno alternative use as general PSU. They show a slight subjective difference between orthodox and heretic. Congratulations for your build, photos are always welcome. For how much current you set the rails CCS? Let us know how you liked it when you will put it to service. Thanks and regards.

[BackinBlack]

Salas, Thanks for your response.
You are correct the current was 7 & 17.1mA, I quoted the volt drop, but noted it as current!
Thanks for the explanation regarding the different loads in the led strings. I had picked up on the difference in rail voltages to add more 2nd harmonic, but thought that referred to earlier versions. I'll have a listen to see which I prefer.
For now I have set the rails CCS at the "normal" 60mA with 2 x 68R. As soon as I have mounted the MOSFETS on reasonable size heatsinks I shall increase the current to perhaps 120mA or so.
Photos, will of course, follow.