The Phase Linear 400 Guinea Pig

wattsabundant

Chief Journeyman
Joined
Apr 26, 2011
Messages
943
Location
Central Ohio
#21
At my day job we use thermocouples for semiconductor temps. The infrared guns tend to give misleading data. I understand it's due to the shiny surface of the transistors and milled heatsinks. With a thermocouple you look for the hot spot on the transistor case. You'll likely find the heatsink temps drop quickly as you move away from the heatsinks.
 

laatsch55

Administrator,
Staff member
Joined
Jan 14, 2011
Messages
75,023
Location
Gillette, Wyo.
Tagline
Halfbiass...Electron Herder and Backass Woof
#23
Can't find the thermocouple for my Fluke. I understand about the infrared inaccuracy with shiny stuff. that's why you see blue masking tape on the case.
 

Gibsonian

Chief Journeyman
Joined
Aug 3, 2011
Messages
845
#24
Pretty impressive, 40 deg less at transistor, about 23% reduction in case temp. Difference in temp will likely grow if you do this at say 175 watt continuous. Does anyone know maximum continuous temp that the transistors can run at?
 

ksrigg

Chief Journeyman
Joined
Jul 28, 2011
Messages
1,451
Location
Wise, Virginia
#27
That copper is really beautiful stuff. Now, how are you going to keep it from oxidizing? I have experimented with cymbals, and have found that plain ole Pledge or even "off brand" furniture polish will keep the copper from turning brown or other funky colors. I've tried about everything out there including all the copper cleaners and polishes, and silver and stainless cleaners and polishes...even one which was originall used for cleaning the Royal Families silver...can't remember the name, but nice stuff for silver...just not copper...I advise getting the copper really clean and shiny and then putting a thin coat of the furniture polish on the copper. I think the wax seals the copper from oxidizing...and it shines like new for a long time...

Just make sure the copper is shining when youy apply it, and
 

Gepetto

Administrator
Staff member
Joined
May 15, 2011
Messages
14,098
Location
Sterling, MA
Tagline
Old 'Arn Enthusiast
#28
The 150C is the max junction temperature, not the case. Plus there is no SOA left as you move in on 150C
 

mlucitt

Veteran and General Yakker
Joined
Jun 24, 2011
Messages
3,497
Location
Jacksonville, FL
#29
For us pedestrians, 150C is equal to 302F. I think the S1 thermostat in the PL400 and the S2 and S3 thermostats in the PL700 are rated at 90C or 194F.
 

Skratch

Chief Journeyman
Joined
Dec 27, 2011
Messages
1,169
Location
Huntley Il.
Tagline
Life is short, so take your time
#30
Running a 3.6 ohm load on a 700 will open those thermostats rather quickly when driven moderately. Trust me
 

wattsabundant

Chief Journeyman
Joined
Apr 26, 2011
Messages
943
Location
Central Ohio
#32
When I developed the 400 relay board I wanted to observe output temps at the time of failure. To cause a 400 to fail, I removed two of three outputs in one channel and then loosened the screws on the one output that was left. Note that an amp runs hottest at 1/3 of it's output power with a resistive load.

At about 70 watts, failure was predictable at case temp of 150C. I don't know what the heatsink temp was. TO3 insulators were gray silpads. Device used was mj15024's. I repeated this about 10 times. I'm sure junction temp was much higher, at least 170C.
 

laatsch55

Administrator,
Staff member
Joined
Jan 14, 2011
Messages
75,023
Location
Gillette, Wyo.
Tagline
Halfbiass...Electron Herder and Backass Woof
#33
Well an AB amp does anyway. that's why I was around 100. Dropping to a four ohm load would have helped a bunch too. When they failed did they go short or open??
 

Gibsonian

Chief Journeyman
Joined
Aug 3, 2011
Messages
845
#35
They got function and form (looks) going for them.

So sounds like a good test to see if they can keep the unit from shutting down temp wise is 1/3 power, 4 ohms?

If you added say 0.5" to the length of each fin, any idea on the change to heat dissipation efficiency?
 

laatsch55

Administrator,
Staff member
Joined
Jan 14, 2011
Messages
75,023
Location
Gillette, Wyo.
Tagline
Halfbiass...Electron Herder and Backass Woof
#36
Scott, an excellent idea. No clue on the last question. I do know, just what we've seen so far, the heat gets removed sooner and better, both good things.
 

Gepetto

Administrator
Staff member
Joined
May 15, 2011
Messages
14,098
Location
Sterling, MA
Tagline
Old 'Arn Enthusiast
#37
Area is a first order determinant in power dissipated by radiation, double the area, double the radiation efficiency of a surface

Stefan-Boltzman law describes radiative heat transfer:

Prad = 5.7 * 10^-8 * E * A * [(Ts)^4 - (Ta)^4] in watts
where:

E = emissivity
black anodized aluminum E = 0.9
polished aluminum E = 0.05

A = surface area (m2) through which heat radiation emerges

Ts = surface temperature (K) of component
Ta = ambient temperature (K)"
 

mlucitt

Veteran and General Yakker
Joined
Jun 24, 2011
Messages
3,497
Location
Jacksonville, FL
#39
Let's see, operations inside the brackets first...

This could be fun. If we can compute the power radiated by the outputs, and add it to the power produced by the amplifier, we could come up with a figure of merit or an efficiency rating for the PL amps because we know (or we can measure) the power going in. I wonder if that has ever been done?

Maybe this is a thesis topic...
 

Gepetto

Administrator
Staff member
Joined
May 15, 2011
Messages
14,098
Location
Sterling, MA
Tagline
Old 'Arn Enthusiast
#40
Adding another 1/2 inch will pick up another 8% radiation efficiency compared to the stock heatsink area.
 
Top