There is a Mezmerize board available on the diya store. Get one and recycle most of your parts to populate it. Get a few NEC EA2 non latching relays too somewhere on line. Maybe its the switches. Alternatively just copy the relays switching part to add it to your now realization. I have made one such part for a JC type preamp I made to a friend not long ago which is also DC and no noises again.
It will produce higher THD and high output impedance as a buffer due to its low Yfs. Also more noise when fed from the lowish source impedance that the audio sources have. Optimized for 100kOhm sources regarding its best noise figure. It can work in other words but not with as good results.
Thank you Salas, i find 2sk170 @ notebooks, PC components, harddrives and electronic at reichelt but i don't know if are fakes or not? any suggestions?
The B1 puts 18V across the two jFETs.
If the jFETs have the same Idss then you find that 9V drops across each jFET and each jFET passes it's Idss.
The jFET dissipation is 9*Idss and the answer comes out in mW.
The LSK170 has a maximum dissipation capability of 400mW when Ta=25°C
I suggest you never run at more than 50% of that and they will be hot. <20% to 25% would be cooler.
That 50% will limit your Idss to 200/9 = ~22mA
20% gives ~8.9mA
The DCB1 works at +-10Vdc i.e. 10V across each jFET.
If the jFETs have the same Idss then you find that 9V drops across each jFET and each jFET passes it's Idss.
The jFET dissipation is 9*Idss and the answer comes out in mW.
The LSK170 has a maximum dissipation capability of 400mW when Ta=25°C
I suggest you never run at more than 50% of that and they will be hot. <20% to 25% would be cooler.
That 50% will limit your Idss to 200/9 = ~22mA
20% gives ~8.9mA
The DCB1 works at +-10Vdc i.e. 10V across each jFET.
Its a German retail firm with good reputation
You would need grade B http://www.diyaudio.com/store/matched-jfet-pairs-grade-b.html
Help! DCB1 PS issue.
First and foremost, THANK YOU to all who give so freely to further the DIY Audio communities! Also, thanks to all who posted questions, pictures and support for this project!
I know this is a late post on this thread, but I have just finished a DCB1 Hypnotize kit from Tea Bag and ran into a problem. The build went well and I was very careful. I have it wired with a 50K Alps pot for volume control.
I powered up the board and the test points for the PS output were at 9.99 and 9.97 vdc. The relay operated correctly and the DC offset was about 2mV per channel. I hooked up an FM tuner and let it run for a couple of hours. I then hooked up a pair of chip amps I have. The sound was bright with a bit of sibilance. I let it run for a few hours and it smoothed out a little. Along the way, I checked PS outputs and they stayed right at the previous values.
This morning, I powered it up to get some more time on it, checked PS values and DC offset, all good. After about an hour, I noticed the bank of 5 LEDs on one channel was not lit. I had not notice any change in sound and both channels were still on. I measured the test points and the lit side was still 9.99, but the unlit side was 19.8 volts. I unplugged it immediately.
I felt around, no noticeable hot spots, no smell and no visually damaged parts.
May I request help as to where to look to chase out this problem?
Thank you!
First and foremost, THANK YOU to all who give so freely to further the DIY Audio communities! Also, thanks to all who posted questions, pictures and support for this project!
I know this is a late post on this thread, but I have just finished a DCB1 Hypnotize kit from Tea Bag and ran into a problem. The build went well and I was very careful. I have it wired with a 50K Alps pot for volume control.
I powered up the board and the test points for the PS output were at 9.99 and 9.97 vdc. The relay operated correctly and the DC offset was about 2mV per channel. I hooked up an FM tuner and let it run for a couple of hours. I then hooked up a pair of chip amps I have. The sound was bright with a bit of sibilance. I let it run for a few hours and it smoothed out a little. Along the way, I checked PS outputs and they stayed right at the previous values.
This morning, I powered it up to get some more time on it, checked PS values and DC offset, all good. After about an hour, I noticed the bank of 5 LEDs on one channel was not lit. I had not notice any change in sound and both channels were still on. I measured the test points and the lit side was still 9.99, but the unlit side was 19.8 volts. I unplugged it immediately.
I felt around, no noticeable hot spots, no smell and no visually damaged parts.
May I request help as to where to look to chase out this problem?
Thank you!
Look for bad joints in all parts or for bad LEDs in the problematic polarity side first. Try to light up the gone dark LEDs one by one with a battery. No desoldering them is needed to check that. If they do work one by one and no bad joints or cracks are visible anywhere, especially at the Mosfet joints due to mechanical stress during sinking assembly, then it must be some bad small semi or a bad MOSFET itself. Do the first inspections and we will see. If finding a bad LED, replace it.
Bad LED located and replaced, burning in nicely.
Salas,
Thanks for the reply.
I did find a bad LED in the suspect string. I replaced it and the voltages are now balanced withing 2mV.
With regard to the Mosfets, I mounted them to the heat sink plate and the board to its stand offs before I soldered the leads. So I should be okay there. Im using a 10K Alps pot for now.
The sound was still pretty harsh at first, as though it was over-driving the circuit. I kept an eye on the voltages and watched for hot components and let it burn in.
I am at about 100 hours and the sound is smoothing out very well; the harshness is all but gone. Imaging is great and it is DEAD Quiet on my ChipAmps.
I appreciate your help and would welcome any further thoughts you may have.
Best regards,
Tracy
Salas,
Thanks for the reply.
I did find a bad LED in the suspect string. I replaced it and the voltages are now balanced withing 2mV.
With regard to the Mosfets, I mounted them to the heat sink plate and the board to its stand offs before I soldered the leads. So I should be okay there. Im using a 10K Alps pot for now.
The sound was still pretty harsh at first, as though it was over-driving the circuit. I kept an eye on the voltages and watched for hot components and let it burn in.
I am at about 100 hours and the sound is smoothing out very well; the harshness is all but gone. Imaging is great and it is DEAD Quiet on my ChipAmps.
I appreciate your help and would welcome any further thoughts you may have.
Best regards,
Tracy
Harsh so much is not something usually reported in any of its testing stages. See that the contacts are good if using Molex signal wiring connectors. Try a 20K pot and/or 100uF good electrolytics instead of the Wima MKP10s. See if it goes any deeper tone with those changes only in case its still annoying on some songs that it rather shouldn't. Congrats for fixing it.
That one I use in the DCSTB PSU already so I am practically familiar with it. In my batch it was chancing between 4-11mA and its not lending itself easily to find pairs for great offset in applications like here because its not offered subdivided in IDSS brackets. If you will use it for a DCB1 the performance will be worse than with the original but decent. Closer to the original or slightly better can be done with the BF861A but its SMD (SOT-23).
Better use lowest VDS possible. I don't know the application you have in mind that's using 15V but DCB1 is using 10V so less mW here. Max is 250mW at 25C. You derate for inside the box ambient and then you have to keep a safe distance from the new max also. It takes largest allowable pads area design to help dissipating the SMDs heat significantly better when you need to push them powerwise. Heavier copper FR4 will not make some valuable change.