Snail Phono Pre

little eddy · Post by **little eddy** » Thu Oct 13, 2011 10:57 am

Progress as per the name is at a snails pace but just trying to get motivated again.

Just about to order some bits from Rapid so my first question is what is the maximum heat dissipation of the 10m45 without a heatsink? I'm tight with space using my current chasis so no heatsinks would be very beneficial.

I'm assuming the stated 40W is with a heat sink. I am nominally looing at 5mA and around 135V so around 0.7W dissipation required.

Also from a safety perspective, is the metal tab that the heat sink would connect to electrically isolated from the device itself or would the heatsinks need to be insulated?

Does anyone know where I can get hold of some BSP135 because not in stock at RS or Farnell. Depending on how far I go could need between 2 and 8. Is there an equally suitable alternative, or does anyone have a stash that they are looking to sell?

With the parallel ECC83/5751 phono 1st stage, I propose to have a common load/CCS but can I also have a shared cathode resistor and possibly cap. Some designs I've seen share the cathode components and others have their own separate legs. Gut feel is that separate cathode components would be electrically superior but I don't want to double up on these components from a space perspective unless they are likely to be beneficial.

Finally does anyone have a pair of 12ay7/6072a valves they are willing to sell?

Mike H · Post by **Mike H** » Thu Oct 13, 2011 1:22 pm

little eddy wrote:what is the maximum heat dissipation of the 10m45 without a heatsink? I'm tight with space using my current chasis so no heatsinks would be very beneficial.

I'm assuming the stated 40W is with a heat sink. I am nominally looing at 5mA and around 135V so around 0.7W dissipation required.

Yes you're right there so not a lot of heat no.

Also from a safety perspective, is the metal tab that the heat sink would connect to electrically isolated from the device itself or would the heatsinks need to be insulated?

The tab is connected to pin 2, = positive terminal, so they all need to be insulated from each other and from chassis. You've got a few of them as well (8 all together) best way I can think of doing it is for example screw 4 each onto 2 pieces of matrix board, which may then be mounted vertically in the chassis to save space, this will make absolutely sure the tabs can't move and touch things making shorts.

Does anyone know where I can get hold of some BSP135

Bad news this keeps coming up "obsolete", and Farnell's page says discontinued. Also found it on digikey, 'qty 0'. Think you might be unlucky

jack · Post by **jack** » Thu Oct 13, 2011 5:40 pm

Mike H wrote:
Does anyone know where I can get hold of some BSP135
Bad news this keeps coming up "obsolete", and Farnell's page says discontinued. Also found it on digikey, 'qty 0'. Think you might be unlucky

Can you use a better rated device, e.g. BSP300 ? The BSP135's specs are pretty rubbish by modern standards. Rds(on) is particularly poor... The BSP300 has a thinner gate region, hence far lower Rds(on) of 20 ohms, higher Vds (800V vs. 600v) etc. Farnell 147-1698 - looking at the schematic, the BSP300 would probably do fine.

Mike H · Post by **Mike H** » Thu Oct 13, 2011 9:36 pm

Ah trouble is that's enhancement mode, he wants depletion mode.

That's the problem hardly any depletion mode FET's around. Makes me wonder about ways of making an enhancement mode one work as depletion mode.

little eddy · Post by **little eddy** » Thu Oct 13, 2011 9:46 pm

nickds1 wrote:Can you use a better rated device, e.g. BSP300 ?

nickds1 - You seem to know more than me but who doesn't?

The BSP135 was used by a few people on the 1st stage of the WAD/WD phono amps so perhaps one of those knowlegeable chappies might be able to comment on the likely suitability of the BSP300 in this particular application.

Heat dissipation looks fine at around 0.14W and 0.18W for 1st and 2nd stages respecively (70V @ 2mA and 120V @ 1.5mA).

I also did a search for BSP300 on Maplins and it came up with these: http://www.maplin.co.uk/mos-fets-33843 Are any of these suitable - looks like they are all under-rated in terms of voltage - shame with them being from 29p!

Mike H · Post by **Mike H** » Thu Oct 13, 2011 10:04 pm

BSP300 is no good, it's enhancement mode MOSFET, you need depletion mode MOSFET, the gate biasing polarities are exact opposites. BSP135 uses negative bias, the 300 positive bias, i.e. needs turning on like a transistor, else it does nuffink.

Although I might mention that as the signal levels in those positions are really quite small, you could still use anode resistors, at least to get it going.

little eddy · Post by **little eddy** » Thu Oct 13, 2011 11:50 pm

Mike H wrote:.... you could still use anode resistors, at least to get it going.

Yes I took note of this in an earlier post so temporary resistors will be used at least to get the thing up and running. The only thing is that moving between resistors and SS CSS will alter the RIAA components when I get round to the phono.

With my smallish voltages, BSP129 and BSP149 might also be possibilities but again are out of stock at RS/Farnell.

Would these do? http://www.ebay.co.uk/itm/5-Stuck-BSP12 ... 415d7720f6

jack · Post by **jack** » Fri Oct 14, 2011 8:28 am

Mike H wrote:BSP300 is no good, it's enhancement mode MOSFET, you need depletion mode MOSFET, the gate biasing polarities are exact opposites. BSP135 uses negative bias, the 300 positive bias, i.e. needs turning on like a transistor, else it does nuffink.

Doh! Just shows how rusty I am now to miss something simple like that.

Post by **Nick** » Fri Oct 14, 2011 8:32 am

Yep, thats why I first used that device, its was a cheap simple CCS, low noise and enough current for a ECC83. A C4S can be used instead if needed.

Mike H · Post by **Mike H** » Fri Oct 14, 2011 3:22 pm

I was tossing and turning in bed last night trying to figure how to make a enhancement MOSFET work like a depletion MOSFET.

No answer yet

Except, involving a bipolar transistor as well, but means a leakage current is added from some resistor that biases the gate, though it could be a large value.

pre65 · Post by **pre65** » Fri Oct 14, 2011 3:27 pm

Would a 10M45s Ixys not do the job ?

OR

http://www.ebay.co.uk/itm/2-x-Supertex- ... _645wt_932

Mike H · Post by **Mike H** » Fri Oct 14, 2011 3:42 pm

Well quite, presumably because it doesn't need to be that much high power.

Mike H · Post by **Mike H** » Fri Oct 14, 2011 4:44 pm

Waitaminute waitaminute waitaminute wait wait wait

Dave the bass · Post by **Dave the bass** » Fri Oct 14, 2011 9:31 pm

Mike H wrote:Waitaminute waitaminute waitaminute wait wait wait

<Question from the dopey kid at the back of the class>

Balanced LV heater supply?

DTB

audio-talk

Snail Phono Pre

#91

#92

#93

#94

#95

#96

#97

#98

#99

#100

#101

#102

#103

#104

#105