Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › [SripBoard Design] Starving Student Millett Hybrid Vacum Tube Amp
New Posts  All Forums:Forum Nav:

[SripBoard Design] Starving Student Millett Hybrid Vacum Tube Amp - Page 17

post #241 of 269



Thanks to everyone for their help today, I managed to get the amp up and running about an hour ago, after a few false starts. It's nice and quiet, in fact with nothing playing I can turn it up to about 80% before I even get a slight hiss. It sounds very nice as many people have said, but I do have one concern, it gets hot, after about 20 minutes the heatsinks are very hot (you can touch them but you can't keep your fingers there for long) the tubes also get warmer than I would expect. Is this normal?

Edited by Frodo - 11/24/12 at 9:20am
post #242 of 269

Totally normal, it's a very hot amp due to its unique use of the gets, I wouldn't worry about the heat. Tube filaments are also of course like lightbulbs and get similarly hot. That's a very nice looking amp you've built and your a lucky bugger for having it nice and quiet ;). I have been and still am having problems quieting mine, are you using the stock power supply?


What cans have you plugged into it?


I have to say that that retro look you've got looks really nice with those JJ tubes as well.




post #243 of 269

Nice job! It looks cute, if such a word can be used to describe an amplifier. biggrin.gif


So you changed C6 and C3 around before turning it on? They were indeed reversed in the picture you posted.


I am too very surprised your amplifier is so quiet, especially considering the messy wiring... I think it's simply your headphones, which do you use with this amp? Are all of your headphones as quiet?


I've spent hundreds of hours and dollars on my build to make it quiet, and it's still too noisy to be used with my Shure SRH-440. It's usable with my Q701 though. For some reason the AKG headphone is quieter.


You could easily cut 1/3 to 1/2 the length of all the wires. I see you twisted each input signal wire around a ground wire, this is great. The next step would be to clean up the tube's wiring, as those too carry sensitive signal. I'm just not sure how you should do it... maybe someone with more experience can tell.


The MSSH do get very hot. MOSFET needs to be biases with a relatively large current in order to be usable as audio amplifiers. Your heat sinks should be just good enough, as they seem to be what most people use. To verify if your amplifier is not getting hotter than it should, simply measure the heater voltage once everything is warmed up. It should be around 14v. Anything higher than this may be too much. Check both heaters and report back, we'll see then if tweaks are necessary.

post #244 of 269

My reasoning as to why the akg is quieter than your shure is purely due to the large amounts of power it takes to drive. Looking at innerfidelity's datasheets:





we can see that the shure requires 50mV and 0.06mW to reach 90dB

the AKG on the other hand requires a whopping 318mV and 1.68mW of power to reach the same output levels.


Obviously you have to turn the amp up louder to use the akgs at the same SPL, however I know that in my build (and I'm taking the liberty of assuming that yours is the same) the noise stays more or less the same until I get into the last 1/4 of the pot's turn, therefore it would be orders of magnitude less noticeable in your akg.


I'm still looking at mine but it seems pretty safe to say that the SRH440s and the SSMH don't seem to get on.

post #245 of 269

Hey thanks for the information! I never noticed those measurements, but they do explain what I observed. The 440 indeed need very little power. I actually had to modify the low gain setting of my NFB-12 to be even lower, because it was too loud with my 440.


In my build, the noise does get higher as I turn the volume up. In fact, it's almost silent when the pot is at minimum volume. The SNR stays more or less the same at every volume level. My 440 are dead silent out of my NFB-12 so it's possible for an amplifier to be silent even with these. I'm just clueless if it's possible for a tube amplifier to be as silent, though. In the meantime, I'll keep on making researches an tweak my MSSH until the SNR is lowered to an acceptable level.

post #246 of 269

Thanks for the encouraging words. Still a few things to do, such as Kim mentioned,shortening wires, and moving R16/17 and R14/15 closer to the pot and tube socket. It was just luck that I had used 2 ground wires and twisted them around the signal wires, I didn't expect any advantage over such short distances, (and yep also had to change C3 & C6 to the opposite polarity)


I noticed that the pot also has the occasional crackle when I move it, probably because it was cheap, I'm not sure how it will effect overall sound, or if there will be any major gains by using a more expensive audio pot.


The headphones I was using to test the amp yesterday were just an old pair of cheap JVCs. I've now tested it with my HD595 and ATH-M50s, it's still very quiet, although now I can detect a sight hiss at the very edge of my hearing (slightly more noise just after power on, but this drops away quickly, interestingly there is more noise with the case open), I wasn't sure if I was imagining it at first. I know mine are a long way from audiophile cans and are easy to drive without amp. I'm hoping to get some HD650's soon, so will give my thoughts later using something with a bit more impedance. At the moment  1/4 to 1/3 volume is a comfortable listen level for me (using an iPod classic as the source).


I'm very pleased with the sound so far, I don't have great deal to base it on, but it's definitely sounds better than the Cmoy and Cmoy Tube hybrid amps I previously built (not to take anything away from Fred_fred2004's kits, which are excellent), but the SSMH seems to have better resolution, I'm hearing new things in recordings that I had never noticed before. Looking forward to making an Objective 2 amp at some point in the future.


The case is just cheap metal one from an electronics shop, I have the same type I used for my alien DAC, I bought them to match, but I read afterwards that the SSMH might destroy the Alien so I haven't tried them together yet. I'm building a Skeleton Dac next which is a little cheaper so I might try the SSMH with that.

The power supply I am using is 48v 0.52A (rated 100-240v) made by FSP, Model # 025-1AD207A http://www.fspgroupusa.com/fsp0251ad207a/p/445.html It was $8 on ebay. (tested with a digital volt meter the voltage switches between 47.7 & 47.8 continusly (anyone know why?). The amperage is a little higher than the CISCO, I thought this might be good if I wanted to use LEDs. At the moment though I'm happy with the glow from the fillaments.


When I switch it off, the volume drops away to nothing and then comes back again for a second before dropping away completly (the power Led also follows this pattern), is this just the caps discharging?


The voltage at the heaters (measured between pins 4&5) is: 12.5 volts for both tubes.


The heat on the heat sinks still worries me,


I'm using standard 34x42x25,4mm heat sinks, I'm going to change to these http://www.aavid.com/products/standard/530002b02500g which are 30mm taller, hopefully it will improve things.


All the best, Andy

Edited by Frodo - 11/25/12 at 7:54am
post #247 of 269

12.5 V is as low as you can safely go, so your build is not wasting more power than it should. As for the heat sink I had to use a fanless slot CPU heat-sink to keep everything cool. I used to have a smaller heat-sink, but it was getting too hot. I burned my fingers a few times just by touching the heat sink, which is usually not a good sign.



My freshly finished SSMH on it's first test drive. The pot, knob and output jack were salvaged from a vintage stereo. There's bad scratching sound from the pot, I'll have to find a better one.



My Millett


Now everything gets equality hot. The MOSFET, the linear regulator and the transformer all get very warm. I rarely use the amp more than an hour or so at a time, just to be safe. Note that the MOSFET are probably the toughest of the bunch. They can safely operate up to 175 C. A normal MSSH dissipates around 15W as heat, which is quite significant for an headphone amplifier. I think the higher heat sinks would be good enough though. If I compared the data correctly, they should stay cool enough (around 50 C).


I think one of the reasons why your build is so silent is that you used a ferrous enclosure. As you noticed, the amp got more noisy when you opened the case, which tells me the steel case really helps. But then some people built them in plastic enclosure and report they are silent...


It's normal that you have ~47.7 V from your power supply. The product datasheet says the regulation is within 5% of 48 V, so 47.7 is actually quite good. It's nothing to worry about. Your DMM "jumps" between 47.7 and 47.8 is also nothing to worry about. It's probably the voltage oscillating by a fraction of a volt, which happens to go up and down the threshold your DMM has to round to 0.1 V. I'd worry if it oscillated over 0.5 V.


And yes, the amp slowly turns off as the capacitors discharge. It's recommended to unplug your headphone before turning it off though, as the discharge can place DC on the drivers which can damage them.

post #248 of 269

That is is one very cool looking amp! I remember reading on the original SSMH thread about about the problems you had with your clear case and heat, which is one of the reasons why I stayed away from plastic and wood. I listened to mine for about 1 1/2 hours last night and heat sinks got hot, but not hot enough to burn, but I will still be happier with the taller ones I think. I also noticed that I can no longer hear the slight hiss that was at the edge of my hearing, it's noise free apart from the crackling pot up until 80%, perhaps there is some sort of settling in time? I don't know if there is a techincal word for it.


I want to build another now :-) it's a shame the 19J6 tubes are so expensive these days, I know the original amp ran at much cooler temperatures. The stripboard method definitely puts a nice sounding amp in the hands of beginners, thanks for sorting out the final layout Kim, if you hadn't I might have ended up shelving the whole thing.


Couldn't resist adding the LEDs across the heaters, I know they are frowned upon, but I'm a noob so I'm still at that stage where LEDs still look cool to me in everything :-)






For anyone who is interested, the case size of my case is 120x78x50mm, it was a tight fit to get everything in so I wouldn't recommend using anything much smaller with the stripboard design and vertically soldered  components. Saying that, the switch I used is pretty big so you could save a fair amount of space if you used something different, and used maybe just a 1/4" jack and 3.5mm for the input, I also have RCA connectors and maybe having both is a bit of overkill.

Edited by Frodo - 11/25/12 at 4:03pm
post #249 of 269
Thread Starter 

First off well done Frodo and excuse my messy diagrams... It was some time ago and I had little experience. Not that I am an expert now but anyways.


As for the forever helpful comunity thank you. I have put my head-fi notification off and I was not there when Frodo needed help.


@KimLaroux Thanks for the fix. Would you mind if I put your diagram on the front page? Is so I will remove it.


Also I was a bit worried of the heat the amp produced specially since I had pretty small heatsink, after using an infrared thermometer, it is safely within the range of the TO-220s. Just watch your fingers haha.


EDIT: OP has been updated.

Edited by TestSubject - 12/16/12 at 2:48pm
post #250 of 269

Great thread and lovely builds! smile_phones.gif I have been doing some research towards doing my own amp and have settled on SSMH (I think). Before I order parts and get the iron hot I have a few questions.

Source: Laptop

Phones: Sennheiser HD598 (50 Ohm)



1. Is there any advantages to using a ground plate and point to point wire it over a strip board? As I see it strip board is a sure way to reduce the length of wire used. Of course I will wire the rest as tidy as I can to my ability to reduce noise.

2. I mostly sit at the computer when I use my headphones, which includes gaming, movies and music=long hours. Can I use this amp for long sessions at the time since it gets hotter than 19J6 builds?

3. What should I be looking for in heat sinks. What values do you recommend since I want to use it for long sessions at a time?

4. Would it be feasible to hide the heat sinks inside the build with openings to let air in and out? (mainly for aesthetics)

5. I have heard turning the amp on or off could release DC voltages straight to the phones damaging them. Is this true? If so is there a way to build in protection for this? I want the amp to sit pretty on my desk and just be turned on and off by anyone without problems.


Appreciate any answers! Oh any particular advice to a first time builder? Looking forward to the build so much I am dreaming about it beyersmile.png

post #251 of 269

Hey Grumus, I'm not as qualified as the others to answer your questions but I'll give it a try.


1. I think the only advantage using a stripboard has over point to point is that it's easier to build it that way for those of us who are not used to following circuit diagrams or who are not confident in point to point soldering. You might find that you actually end up using more wire with the stripboard design than point to point (I'm sure I did).

2. I use my amp just to listen to 1 or 2 albums at a time, it does get hot, it probably won't be a problem for longer listens, but I really only use it for special occasions when I can really sit down and enjoy it. It has a very unique and enjoyable sound. Other might disagree, but think you are better off with a small solid state amp for gaming and movies. My SSMH also takes a couple of minutes to warm up, it still crackles a little at first.

3. I changed my heatsink for larger ones, the ones I have now are 20mm taller than the ones in the photo. They still get hot but they wont burn you, my guess the temperature stabilizes at about 55oc, which I think is about what KimLaroux calculated . I'm sure you could use bigger heatsinks, but there is probably some law of diminishing returns. I'm sure someone with a physics background could tell us about that.

4. Personally I wouldn't put them inside the case, they need to dissipate a lot of heat. Perhaps if your casing was some sort of mesh cage or you had fans it might be ok. To me though I think it would over complicate things.

5. Yep, you can get a bad spike, always remove them before turning the power on or off. I think you can buy/build a suppression circuit, but it just complicates things. Maybe think about mods later. If you need an amp for everyone to use possibly a valve amp is probably not the best choice, have you looked at kit amps like the O2. I have one and it's a great little everyday desktop amp.


My advice, just take your time with the starving student, plan the layout well and use very cheap headphones for the first tests. It's a very satisfying amp to build, I still love the sound of mine. In fact I think I'd like to build another.


Oh, and don't go too small with the case if you decide on using the stripboard.


Hope to see some photos of your build soon!

post #252 of 269

Thanks for the answer! 


1. Alright, I am confident with soldering and confident enough to follow diagrams. So I might go for point to point anyway. I will order it with stripboard anyway.

2. Ok, I will get quality heat sinks for it anyway. And if it's only adequate for shorter periods I will have a good reason to build a solid state :D

3. Big(ish) quality sinks check!

4. Ok, will think hard before I place them in a bad spot.

5. Might use it as my personal amp then if I cant find a reasonable way of adding protection. I have looked at the O2 and it's abit pricey since I am actually a starving student!


If I go for point to point I am thinking of having a big ground plate and mounting everything on that (like the bottlehead crack). Is it a bad idea to have the common ground open like this safety wise? I know it's not a deadly amp but I want to do it right so I wont make mistakes in the future.


Yeah I am thinking of doing a CAD model of the entire amp before I build it to get a better feel of the project (and a bit of CAD fun!) Will post pics if I decide to CAD it for others to get a feel for the layout.


Thanks again! 

post #253 of 269
Originally Posted by Grumus View Post

Great thread and lovely builds! smile_phones.gif I have been doing some research towards doing my own amp and have settled on SSMH (I think). Before I order parts and get the iron hot I have a few questions.

Source: Laptop

Phones: Sennheiser HD598 (50 Ohm)

You aren't going to get very good results amping the output from a laptop. Laptops are typically the worst quality sources available. You will want to get a USB DAC to make the amp building exercise worthwhile. Sorry to say it to a bonafide starving student.


2 - the overall heat is the same between the 2 builds, both have 150mA through the output stage at 48V. The difference is how the power is split between mosfet and tube heater (power = heat). The 19J6 has 19V across its heater leaving 29V dropped by the mosfet, and its 12.6V and 35.4V with the 12AU7, so the mosfet is dissipating about 1W more heat (6.4V difference x 0.15A).

The extra power dissipated in the mosfets is still well within their limits, the IRF510 happily run hot compared to other devices, so it just comes down to how hot you want your heatsinks.

In terms of listening time, heatsink temps stabilise at a set temperature above ambient, there is no limit to how long you can safely run the amp, unless its not getting fresh air (shut in a drawer or cupboard etc)


4. if putting inside the case, then the case has to be able to easily dissipate the 11W that the 2 heatsinks are releasing into the case, otherwise everything else inside is going to get hot too. You would need heaps of airflow for this to be feasible.

Another option would be to use other heatsinks you prefer the look of, the possibilities are endless.


5. easy enough to use a protection circuit. amb's epsillon 12  is one such item. You could also copy the relay output protection from the minimax schematic which i believe is a simplified version of the same circuit.


If you really need to keep costs down then have the headphone output going through a switch to setup a manual delay.

 - turn on the power switch

 - wait 30 seconds

 - turn on the headphone output.

and turn off in the other order (but no need to wait 30 seconds to turn off)

post #254 of 269

Maybe I should build a DAC as well and put it in the case? Do you have any good tips of DACs that will work well with the starving student? I have heard that it can destroy DACs.


2. I actually got my hand on 19J6 tubes for 6 dollars each at a radio museum right next to my school. Had no idea I had a source for tubes that close to me until yesterday! Good to hear that the listening times can be long! And I will get a big heat sink to keep me if not the mosfets happy.


4. Maybe I will just put them on the back behind the amp.


5. I have been searching around and this is what I need (I think). 


It's the e12 delay circuit from the Millet Max. This would seem like a good way of protecting my phones. I would need some help in how to connect this in the amp and what values each component should have. Would be great if someone could put this together with the amp scheme. 


Maybe I just should go with a switch system but I would be very proud of my self if I made it user friendly for everyone dt880smile.png

post #255 of 269
Originally Posted by Grumus View Post



2. I actually got my hand on 19J6 tubes for 6 dollars each at a radio museum right next to my school. Had no idea I had a source for tubes that close to me until yesterday! Good to hear that the listening times can be long! And I will get a big heat sink to keep me if not the mosfets happy.




Do you know how many they have in stock? For the price they let them go, I'd guess they have a significant amount... And if that's the case, I know someone who'll buy the lot. Just don't go and tell them they are actually rare and wanted tube, cause they may jack the price up.



New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › [SripBoard Design] Starving Student Millett Hybrid Vacum Tube Amp