[BoilermakerFan]

Quote:

Originally Posted by WilCox /img/forum/go_quote.gif Nevermind! I just looked at the board layout of the V1 MAX and realized that there would be a lot of traces and ground planes that would need to be cut to make this work.

I plan to add a polyfuse, but I can't remember what points it spans and which trace gets cut...

 

[WilCox]

Quote:

Originally Posted by BoilermakerFan /img/forum/go_quote.gif I plan to add a polyfuse, but I can't remember what points it spans and which trace gets cut...

Actually, it may not be so bad! CR5 may not be needed and may be able to be removed. Then the anode of DR3 can be relocated to the + pad where CR5 was. Then cut the trace after the CR5 + pad before it joins the power buss and insert the Polyfuse there. Colin and Tom should confirm if that would work and (assuming they have tried it) that it makes any difference.

 

[Beefy]

tomb suggested a procedure to me to convert the V1 to V2, but it is quite involved and not for the faint hearted......

If it were me, I would either wait for V2 to build a new one. Or build a second if I already had one!

 

[tomb]

Quote:

Originally Posted by Beefy /img/forum/go_quote.gif tomb suggested a procedure to me to convert the V1 to V2, but it is quite involved and not for the faint hearted...... If it were me, I would either wait for V2 to build a new one. Or build a second if I already had one!

V2 is definitely looking up. The board is populated and the relays appear to work fine, although I haven't tested it with the Bantam, yet. The headphone relay is interesting - Colin has gone toward less parts, not more. Part of that philosophy dictated a change to 24V relays and the use of a 78L24 regulator instead of the zener diode/resistor combination. We had to rework the delays at the higher voltage, so you'll see a requirement for a 2M resistor. Either that, or it was upsizing to 1000uf caps or more. We didn't want to do that. Anyway, it works great - as a matter of fact, the offset that's always appeared on turn-off is completely gone. Most headphone delay circuits exhibit a brief excursion of a few volts under zero load, a bit less with a headphone connected. Even under zero load with this one, it's immeasureable on my Fluke. The parts selection we worked out will give you 45 seconds from a cold start, but at least 20-30 seconds on a quick re-start.

The power supply measures 45uV noise with the LNMP, so the MiniMAX power supply is a success for the MAX V2! This is also with 2x2000uf in the PS and 4x3300uf caps for CA4/5. I used a 0.75A polyfuse. I built it with MOSFETS, that's the reason for all the current bling.

Right now, I'm not happy with the sound of the MOSFETs, though. The JFET-mod is not up to the quality of the MiniMAX sound with BJT's. With 1-1/2" sinks, I have it biased up to 150ma, but it runs a little hot on the middle sinks.

I've tried dropping the bias down to 120ma, but it sounds noticeably worse - a hard, etched sound. One of the things Colin suggested is that the gate capacitance on the Z24/Z34's is too much for the tubes and this style of buffer. We talked about a lot of different things, though. Still in the next few days, I'm going to change those out to the IRF510/9510's and some Zetex's for the input JFETs. We'll see if that makes a difference.

I'll post some pics when I get a chance. I've been covered up with doing other stuff lately.

 

[Beefy]

Quote:

Originally Posted by tomb /img/forum/go_quote.gif This is also with 2x2000uf in the PS and 4x3300uf caps for CA4/5. I used a 0.75A polyfuse. I built it with MOSFETS, that's the reason for all the current bling.

Bling indeed! Those must be pretty tall caps to fit?

Quote:

Right now, I'm not happy with the sound of the MOSFETs, though.

How is it different from the old MOSFET Max that everyone seems to love......

 

[tomb]

Quote:

Originally Posted by Beefy /img/forum/go_quote.gif Bling indeed! Those must be pretty tall caps to fit?

They're all ~35mm.
Quote:

How is it different from the old MOSFET Max that everyone seems to love......

It shouldn't be - except that it has a cleaner power supply. Now either the MAX's old power supply muddied things up a bit so that the hard edges were softened out or it's been that way all along and no one had built enough of the BJT's to be familiar with the comparison. That's the only thing I can figure - because everything else works perfectly.

I ran it for quite awhile - some of this may be due to the KZ's I was using for output caps. Those obviously have a break-in: a lot less time than Black Gates, but the changes are striking. Many people talk in terms of the KZ's having a hard edge or "etching" to their sound, also. So maybe it's confluence of "etch", I don't know, but I ordered some Silmic's to replace them. That's the first thing I will try.

 

[Beefy]

Quote:

Originally Posted by tomb /img/forum/go_quote.gif Now either the MAX's old power supply muddied things up a bit so that the hard edges were softened out or it's been that way all along and no one had built enough of the BJT's to be familiar with the comparison.

I do know that sacd_lover has had quite the assortment of Maxes, and raves the most about his MOSFET Max. It might be interesting to see what configuration he has in his amps.

Quote:

I ran it for quite awhile - some of this may be due to the KZ's I was using for output caps. Those obviously have a break-in: a lot less time than Black Gates, but the changes are striking. Many people talk in terms of the KZ's having a hard edge or "etching" to their sound, also. So maybe it's confluence of "etch", I don't know, but I ordered some Silmic's to replace them. That's the first thing I will try.

If it were me, I would go with old-familiar. In your case, stick some Muse ES in there for some more direct comparisons to the combo you enjoy the most, and have the most experience with.

 

[soloz2]

Hmm... I've built several BJT versions and one MOSFET version and didn't notice the problems you're describing

 

[BoilermakerFan]

Quote:

Originally Posted by Beefy /img/forum/go_quote.gif If it were me, I would go with old-familiar. In your case, stick some Muse ES in there for some more direct comparisons to the combo you enjoy the most, and have the most experience with.

I second that Tom! But you already knew that...

Seriously though. Can you swap in the ES caps while you wait for the Silmics to show up? I think the KZs are causes at least half of your pain. I can send you my set of Cerafines and BC transistors too if you wish.

I have to prep all 5 of my NABU lower chassis for paint this week. My car was hit by my neighbor and my preferred shop is a direct bill with his insurance company. So I get to use my preferred shop and he has to buy Audi Brilliant Black Metallic paint anyways... So my lower chassis will all be sprayed Brilliant Black metallic instead of the satin since it will be $20 for all five this way.

Perf' ing the bottoms of the two V1.2 MHM chassis to line up with the new board cooling holes and adding holes in the bottom at the heatsink mounts to make my swaps faster too. The Aikido 24V will be prepped and painted along with 2 chassis for PSUs on the consoles. I'm going to try to get one top prepped with Om cut into the top to be backed by mesh for the cooling and two other will have blue agave plant silhouettes cut into them for the other two, but they won't be painted black...

My buddy also dropped off one of the most amazing pieces of figured walnut I have seen in a long time. It's old growth harvested off a customer's private farm and milled locally. Wow! Only finish it will get is carved lettering and an oiled finish. Going to be one awesome face plate for one of my amps. I have enough of the walnut to also make feet and use as detail in the maple knobs for it. Only cost me

for the wood too. I suspect my buddy who gave the wood will eventually become the owner of my third MHM and a set of SFI orthos I build for him, but we'll see. He's a vintage guitar player so I'll have to get with Colin on how to mod (if necessary) the input to accept an electric guitar source as one of the two inputs beside the BantamDAC.

 

[tomb]

Thanks, guys! I already have a pair of Silmics and will put those in first (got them at CA2 already). It may just be the KZ's, but I have my doubts that they really sound that bad on their own.

When I described the problem to Colin, of the things he suggested might be causing the issue, one was that it might be a symptom of too much gate capacitance in the MOSFETs. Turns out, the IRF510/9510's are way down in the capacitance values compared to other MOSFETs. Maybe that's why Pete used them on the Starving Student. We'll see what does what, though.

It will have to wait a few days longer, unfortunately - I'm fighting with a couple of Starving Students that need to have the production boards and casing design vetted.

EDIT: Sounds like some outstanding casework you have going there, BMF! Looking forward to seeing them!

 

[kugino]

both mouser and digikey are out of the CA4 and CA5 caps (1800uF 35V)...they won't be shipping until later this summer!! so, can someone recommend a replacement for these caps? thanks.

 

[tomb]

Quote:

Originally Posted by kugino /img/forum/go_quote.gif both mouser and digikey are out of the CA4 and CA5 caps (1800uF 35V)...they won't be shipping until later this summer!! so, can someone recommend a replacement for these caps? thanks.

Mouser has 91 of the UPW's in 1800uf 35V:
UPW1V182MHD6. These are slightly smaller than the 2200uf UPW's, but definitely in the ball park. They have 384 of the 2200uf 35V:
http://www.mouser.com/Search/Product...vfdOEj8hrJQ%3d
They also have a Nichicon UHD, which looks to be the best choice for ESR and ripple, very close to the Panasonic FM at 1800uf 35V, same physical size:
http://www.mouser.com/Search/Product...5QXNp7Ln7Ok%3d

Also, the 1000uf 50V Panasonic FM's at DigiKey are very close to the same ratings of ESR and ripple - exact same physical size.

 

[kugino]

Quote:

Originally Posted by tomb /img/forum/go_quote.gif Mouser has 91 of the UPW's in 1800uf 35V: UPW1V182MHD6. These are slightly smaller than the 2200uf UPW's, but definitely in the ball park. They have 384 of the 2200uf 35V: UPW1V222MHD6 They also have a Nichicon UHD, which looks to be the best choice for ESR and ripple, very close to the Panasonic FM at 1800uf 35V, same physical size: UHD1V182MHD Also, the 1000uf 50V Panasonic FM's at DigiKey are very close to the same ratings of ESR and ripple - exact same physical size.

thanks, tom. of these, which do you think would work/sound the best?

 

[tomb]

Quote:

Originally Posted by kugino /img/forum/go_quote.gif thanks, tom. of these, which do you think would work/sound the best?

For MAX or MiniMAX? I'm getting my threads mixed up these days. Those suggestions are for a MiniMAX where the height is restricted to 1" (25mm).

If you're building a MAX and can go as much as 1-1/2" height, then a whole range of other caps can be used. I put 3300uf 35V Panasonic FC's on the V1.2 prototype MAX.

 

[Beefy]

Quote:

Originally Posted by tomb /img/forum/go_quote.gif I put 3300uf 35V Panasonic FC's on the V1.2 prototype MAX.

It just clicked...... I'm guessing you don't have to worry about turn off thump with the new relay system?