[sachu]

Very cool stuff Jazzy..very cool indeed.

In fact i just turned on my own EHHA after more than 2 months after being only on the stacker 2..the EHHA is one sweet amp for sure. Enjoy.

 

[TheShaman]

Looks cool! Any pics of the inside?

Did anyone have the chance to compare the EHHA to an Aikido (pref. high voltage)?

 

[Horio]

So I'm thinking about using one of the Glassware H-PS-1 heater supplies in my balanced build. I want to be able to run 6H30's which need roughly 1A each. So if I use this kit at 12.6V and wire the tubes in (2) parallel sets, my draw should be 2A.

If I go with a 6.3V-0-6.3V (wire in series) center tapped transformer, the rectified voltage is roughly 12.6V x 1.3 = 16.4V. My understanding is the LM1085 has about 0.6V drop, so power dissipated would be:

V drop => (16.4V-0.6V)-12.6V = 3.2V
P to dissipate => 3.2V*2A = 6.4Watts

Per the TubeCad.com blog, with the LM1085 with the 2.5" heatsink can shed about 8Watts.

Does this look correct to you guys? Please be kind, I am no EE guy but I am trying to learn...

 

[Lifthanger]

it's 12.6V x squareroot of 2 ( 1.414) = 17.82V

 

[Horio]

Quote:

Originally Posted by Lifthanger /img/forum/go_quote.gif it's 12.6V x squareroot of 2 ( 1.414) = 17.82V

Thanks Lifthanger. So the revised calc:

Vdrop => (17.8V-0.6V)-12.6V = 4.6V
P to dissipate => 4.6V x 2A = 9.2W

Looks like that puts me over, so I'd need another heatsink option or perhaps two supplies.

 

[mideel]

Quote:

Originally Posted by Horio /img/forum/go_quote.gif V drop => (16.4V-0.6V)-12.6V = 3.2V P to dissipate => 3.2V*2A = 6.4Watts Per the TubeCad.com blog, with the LM1085 with the 2.5" heatsink can shed about 8Watts.

You might be a little off on your calculations, using the guide from AMB I get:

P = (Vin - Vout) * I
Tj = Ta + (P * (Rjc + Rcs + Rsa))

P = ((12.6V*sqr(2)) - 12.6V) * 2A = 10.44W
Tj = 30°C + (10.44W*((3.0°C/W) + (0.0045°C/W) + 2.60°C/W)) = 88.51°C

* Tj = junction termperature (LM1085)
* Ta = ambient temperature
* P = power to be dissipated
* Rjc = thermal resistance of junction to device case (LM1085)
* Rcs = thermal resistance of device case to heatsink (Thermal Paste I use)
* Rsa = thermal resistance of heatsink to ambient ( 530002b02500g)

This is still in the recommended temperature range of the LM1085 by: 125°C-88.51°C = 36.49°C and well under 100°C (even If Rcs would be as high as 1°C/W we get 99°C)

EDIT: Also the "8W @ 30°C" is based on air circulation near the heatsink or the air velocity, as is stated on the aavidthermalloy page
So in conclusion:

 

[Horio]

Mideel, thanks for the info/calc.

One issue I see, is the unit is going to be enclosed in a chassis with two o22. Though there are some vents on the top panel, I could potentially see the internal temperature rising above 30C. If you do the calculation with say 40C, you bump right up to the 100C recommended threshold.

So maybe if I went with a 12VAC toroid xformer, and I re-run the calc I get about 89C. Maybe that would work...


Quote:

Originally Posted by mideel /img/forum/go_quote.gif You might be a little off on your calculations, using the guide from AMB I get: P = (Vin - Vout) * I Tj = Ta + (P * (Rjc + Rcs + Rsa)) P = ((12.6V*sqr(2)) - 12.6V) * 2A = 10.44W Tj = 30°C + (10.44W*((3.0°C/W) + (0.0045°C/W) + 2.60°C/W)) = 88.51°C * Tj = junction termperature (LM1085) * Ta = ambient temperature * P = power to be dissipated * Rjc = thermal resistance of junction to device case (LM1085) * Rcs = thermal resistance of device case to heatsink (Thermal Paste I use) * Rsa = thermal resistance of heatsink to ambient ( 530002b02500g) This is still in the recommended temperature range of the LM1085 by: 125°C-88.51°C = 36.49°C and well under 100°C (even If Rcs would be as high as 1°C/W we get 99°C)

 

[Lifthanger]

Quote:

Originally Posted by Horio /img/forum/go_quote.gif Thanks Lifthanger. So the revised calc: Vdrop => (17.8V-0.6V)-12.6V = 4.6V P to dissipate => 4.6V x 2A = 9.2W Looks like that puts me over, so I'd need another heatsink option or perhaps two supplies.

no problem, I'll run into the same problems as soon as I start working on my 4 EHHA boards

 

[johnwmclean]

Horio,

For a 12.6V output wouldn’t the transformers secondaries need to be a few volts higher (a little bit of headroom needed), your more likely looking at a 15VAC.

 

[Horio]

Quote:

Originally Posted by johnwmclean /img/forum/go_quote.gif Horio, For a 12.6V output wouldn’t the transformers secondaries need to be a few volts higher (a little bit of headroom needed), your more likely looking at a 15VAC.

Someone please correct me if my understanding is wrong, but the rectified voltage of 12VAC is:

12V x 1.414 = 16.98V

LDO regulators (like the LM1085) only drop the voltage about 0.6V so:

16.98V - 0.6V = 16.4V

So this means the regulator has to shed another 3.8V to get down to 12.6V. Am I thinking of this correctly?

 

[tomb]

Quote:

Originally Posted by Horio /img/forum/go_quote.gif Thanks Lifthanger. So the revised calc: Vdrop => (17.8V-0.6V)-12.6V = 4.6V P to dissipate => 4.6V x 2A = 9.2W Looks like that puts me over, so I'd need another heatsink option or perhaps two supplies.

Actually, you've got drop in the rectifiers as well - 1.4V in a bridge. Plus, you're only going to hit ~0.9 of the 1.414 times the AC voltage when fully rectified. So, you'd have (17.8V x 0.9) - 1.4 = 14.62V going into the regulator. The voltage drop that would have to be dissipated would then be (14.62VDC) - 12.6VDC = 2.02V. Power is then 2.02 x 2A = 4.04W. Say, 4W.*

Of course, this also depends on the current rating of the transformer. If significantly over-rated, it may supply up to a volt or more extra voltage. I've not found too many transformers that supply more than about a volt over rating in this range when under almost any kind of load, though, so use your best judgment. At any rate, it's probably a lot less power to dissipate than first thought.

 

[Horio]

Quote:

Originally Posted by tomb /img/forum/go_quote.gif Actually, you've got drop in the rectifiers as well - 1.4V in a bridge. Plus, you're only going to hit ~0.9 of the 1.414 times the AC voltage when fully rectified. So, you'd have (17.8V x 0.9) - 1.4 = 14.62V going into the regulator. The voltage drop that would have to be dissipated would then be (14.62VDC - 0.6VDC) - 12.6VDC = 1.42V. Power is then 1.42 x 2A = 2.84W. Say, 3W. Of course, this also depends on the current rating of the transformer. If significantly over-rated, it may supply up to a volt or more extra voltage. I've not found too many transformers that supply more than about a volt over rating in this range when under almost any kind of load, though, so use your best judgment. At any rate, it's probably a lot less power to dissipate than first thought.

Thanks Tomb. This is all very helpful and informative. So, say I went with a 12VAC toroid xformer:

Rectified Voltage => (12V x 1.414 x 0.9) - 1.4V = 13.9V
Power to Sink => (13.9V - 12.6V) x 2A = 2.6W

So per this calculation, the 2.5" heatsink has to only dissipate 2.6W. Even if the toroid xformer was putting out an extra volt, that still only bumps it up to 4.6W, which shouldn't be too much of a problem.

 

[Horio]

So the 6GM8 tubes are pretty expensive/hard to come by, but the 6922's are much easier to find. How do the 6922 tubes stack up against the 6GM8's or 6H30's? I'm looking for a decent tube for setting up the amp and to get it up and running.

 

[digger945]

Quote:

Originally Posted by Horio /img/forum/go_quote.gif So the 6GM8 tubes are pretty expensive/hard to come by, but the 6922's are much easier to find. How do the 6922 tubes stack up against the 6GM8's or 6H30's? I'm looking for a decent tube for setting up the amp and to get it up and running.

I've also used 6DJ8, 7308, and 7DJ8/PCC88.
I only have one pair of Sylvania 6922s. They are ok, to me. I like the 7308s better. Each one is different.

EDIT: I'de loan ya a pair of Sovtek 6H30s but I loaned them to a HF'er over a year ago and haven't seen them since.

 

[sachu]

Quote:

Originally Posted by digger945 /img/forum/go_quote.gif I've also used 6DJ8, 7308, and 7DJ8/PCC88.

Yeah, but I didn't find the 6DJ8 to sound as good as the 6GM8 in the EHHA.