The diyAudio First Watt M2x

I removed the Tucson boards and put a jumper across 2 and 4. Using my phone as a signal source the M2X now plays much louder than when the boards were in place. Finally sweet, sweet music! While I think about what could be wrong with my Tucson boards, which by the way are using OPA1611 op-amps, I may build a different IPS such as Mtn. View.

Even with the IPS boards removed and bypassed my M2X still doesn’t work with the Pioneer receiver as the signal source; I can’t get any sound out of it. The Pioneer has no problem driving my Hafler amp. It’s got me stumped…

To follow up, my M2X is working now and sounds AMAZING. Thank you for all the help!

The cause of the problem was dead Tucson boards. I don't know if I fried the OPA1611 SMD op-amp while soldering or something happened when they were installed in the amp. Bypassing the daughter board section with a jumper between posts 2 and 4 allowed signal to get through the amp, so it was clear where the issue was. At first I thought my Pioneer preamp wasn't able to drive the M2X with the input stage bypassed but I realized that I just needed to turn the gain WAY UP until working daughter boards were installed.

I'm going to rebuild the Tucson boards with the OPA604 DIP. In the meantime I've completed the Mtn. View boards which are sounding great.

By the way I used the SLB PSU and mounted the Antek AN-3222 transformer vertically (like 6L6). The Edcor audio transformers, without Mu Metal shielding, are approximately 6" from the Antek. I can't hear any hum in one channel and it's only audible in the other when I put my ear 8" from the speaker (90.5 dB sensitivity).

I'll post pictures when I finish cleaning up the wire routing, hardware bits, etc.
 
Last edited:
To follow up, my M2X is working now and sounds AMAZING. Thank you for all the help!

The cause of the problem was dead Tucson boards. I don't know if I fried the OPA1611 SMD op-amp while soldering or something happened when they were installed in the amp. Bypassing the daughter board section with a jumper between posts 2 and 4 allowed signal to get through the amp, so it was clear where the issue was. At first I thought my Pioneer preamp wasn't able to drive the M2X with the input stage bypassed but I realized that I just needed to turn the gain WAY UP until working daughter boards were installed.

I'm going to rebuild the Tucson boards with the OPA604 DIP. In the meantime I've completed the Mtn. View boards which are sounding great.

By the way I used the SLB PSU and mounted the Antek AN-3222 transformer vertically (like 6L6). The Edcor audio transformers, without Mu Metal shielding, are approximately 6" from the Antek. I can't hear any hum in one channel and it's only audible in the other when I put my ear 8" from the speaker (90.5 dB sensitivity).

I'll post pictures when I finish cleaning up the wire routing, hardware bits, etc.

I found the Tuscon OPA604 to be pretty meh. The OPA1611 sounded far better to me, but it is my least favorite board.

Do yourself a favor and get a DIP8 socket, and an SOIC to DIP adaptor so you can try both.
LCQT-SOIC8-8 Aries Electronics | Mouser
4808-3004-CP 3M Electronic Solutions Division | Mouser

The DIP socket will not only prevent you from cooking the IC, it will let you swap out the opamp. You can even try different SMD opamps and compare them :)

BTW Mountainview is prolly my favorite board, though the jury is still out - all of them sound good.
 
Member
Joined 2011
Paid Member
The problem with the single stage filter from Schurter, linked in #2220, is that it is rated for ten amperes of mains current. M2x draws about 1.3 amps from the mains in 115VAC countries, and about 0.65 amps in 230VAC countries. Check your builder's notes to remind yourself which mains fuse you installed; in my M2x it's a 2.5 amp slow-blow @ 115V. Why use an inline filter rated for 4x more current than you can ever possibly draw?

It's not just wasted money. The filters themselves perform worse and worse as the current rating rises. Schurter's datasheet shows performance curves, take a look: the best performance comes from the lowest current rated filters. They even include a table of data that tells you WHY. To handle higher currents, Schurter's designers are forced to use larger diameter wires in their coils, so they cannot wind as many "turns" on the ferrite core. Adding to their difficulty, higher current increases the magnetic flux density in the coil core, so they must reduce the number of turns to avoid saturating the ferrite. Result: lower inductance. It's right there in their tables. Lower inductance means lower performance RFI filtering.

To get the largest amount of RFI filtering (highest attenuation), follow these simple guidelines:

  1. Use two-stage filters rather than single stage
  2. Use the lowest current rating filter, which is just barely higher than your equipment's maximum draw
  3. If you are truly crazy, consider connecting "N" number of filters in series
 
RF/EMI Schaffner has the greater attenuation graphs. Years ago I researched deeply and that was my conclusion. Schurter, for the IEC connectors without filters is my other choice.

As Mark says, you have to choose the one with the tightest amperage because the attenuation is greater. I have a lot of 1A and 3A. And one of 6A and other of 8A (power strip). Spain: 230 Vac, 50 Hz. With 115 Vac you need double amperage.

Schaffner FN9244B with B = medical

https://www.mouser.es/ProductDetail/Schaffner/FN9244B-3-06/?qs=%2Fha2pyFaduizvpSn1rUp%2FhCeqfVzEXqxiooWwv4TR5mv9p1P2yZfKA%3D%3D

An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.
 
Last edited:
I have been advising for a few years, with not much success, as with the minimum aluminum thickness of 4 mm, star-aquad cables and...

Depending on the quality of the filtering of the device in question, sometimes two cascaded Schaffner filters are optimal. Others, only one.

Also, hurrying, one more a Würth 150 kHz ferrite. Let us say that a 150 kHz ferrite is like half (attenuation) of a FN9244B filter.

It is trial and error.

BTW: I have a lot of these ferrites, 8 mm. And one of 12.5 mm (power strip).
 
Last edited:
Other interesting alternative for the audio systems are wall galwanic separation transformer with power handling ~ 1k Watts or +
Tamura are not the only one , many of ordinary electrical producers brands have them in stock
for medical equipements one example is french " Le Grand " industrial model I use. See principe on the schematic
 

Attachments

  • 4DE31B12-0EE1-4023-8CFF-EFE98FD9684E.JPG
    4DE31B12-0EE1-4023-8CFF-EFE98FD9684E.JPG
    295.3 KB · Views: 380
OK!
The chassis I have for Whammy use a normal size IEC without fuse, switch etc. So I have to add this…...at least the fuse.


I wonder what the low A-rating filters like 1A can handle in inrush current if you use "oversized" transformers but only use a small part of its "VA". Is that number mentioned in the data sheets for the filters?
 
Which "size" of rail fuse will be appropriate with M2X?
I have 3.15 A (F-type) in my mind. It should just be a kind of short circuit protection and protection if something goes "mad" so I don't decharge the capacitor bank on M2X PCB.
Will M2X leave class A in low impedance loads? ….e.g. if I test with e.g. 1 ohm load....what will happen if I don't go beyond the max. limit of the output transistors (and fuses).
 
I powered up one of my M2X mono blocks for the first time. Input/output wiring is not yet 100% finished so it was just a quick test to check if everything seems to work as expected. I measured DC offset with one DMM and bias with the other (red DMM in picture by measuring voltage over one of the 0.47 ohm resistors). Bias stabilized at approx. 0.61V which is approx. 1.3A (as far as I remember this is within spec). Rail is +-23V. I made a raw adjustment of offset and it is about 3 mV in the moment. I switched to AC mV and it just shows 0.0 mV so not much noise at output. Shielding of Edcor and toroid and additional filtering using DC filter chokes and 4-pole Jensen caps as last step seems to work well. So.....so far so good. Will finish the wiring and go ahead with last par of wiring of the 2nd mono block. The heatsinks of this 5U deluxe does not get warmer than you can keep your hands on it forever. About 60W of dissipation.
 

Attachments

  • DSC_4004_00001.jpg
    DSC_4004_00001.jpg
    922.6 KB · Views: 521
I also look forward to a listening test but still some work to do until then. Finish building and a lot of testing.
I powered it up without the vario transformer and just connected it to the wall outlet so I also tested the inrush. The 2AT fuses together with NTC seems to "tame" the 500 VA toroid. Without NTC inrush limiter I think a 5AT or 6.3AT was needed. A 1000 VA needs a 10AT with any inrush limitation.
 
Congratulations on getting your amp working again. Now that the dust has settled, would you describe what pieces actually failed when you accidentally swapped SupplyPosRail and SupplyNegRail , and how you repaired those failures?

It seems it was only the daughter board that got damaged.
I only replaced parts on channel A while I was sorting things out, channel B worked fine after I used a different daughter board.
As to which parts exactly got damaged, I do not know. I posted a pic of one of the boards a few posts back, and it shows it that it got hot enough to melt the solder joint at one of the diodes.

I plan to just replace all the parts on both daughter boards, the opamps are very likely blown, and they are the expensive parts - the rest are fairly cheap. IIRC, Norwood is the most expensive board after Ishikawa and it's unobtainium JFETs.

I will also replace the IC's on channel B when I replace the heat sink to be on the safe side.

So, about $40.00 in parts for the daughter boards, and another $20.00 for amp board parts.

The bright side that switching to discreet diodes for rectification made a much bigger impact than I expected it would. The old rectifiers got pretty hot.

Thank you all