On Semi ThermalTrak

[Adrculda]

I think IRFP240 and IRFP9240 would be a bit of overkill
But thats what im going to use as drivers for the thermal track units

 

[Nelson Pass]

I think IRFP240 and IRFP9240 would be a bit of overkill
But thats what im going to use as drivers for the thermal track units

Since you want to minimize distortion, I will suggest that
the IRFP9240 as rendered by IR is not the best choice as
it suffers from the "IR P channel issue", documented by
me and others elsewhere in the forums. You could look
at some of the Fairchild and Harris parts, and I think
probably the best will be the Toshiba equivalents (a little
more expensive)

 

[Adrculda]

Since you want to minimize distortion, I will suggest that
the IRFP9240 as rendered by IR is not the best choice as
it suffers from the "IR P channel issue", documented by
me and others elsewhere in the forums. You could look
at some of the Fairchild and Harris parts, and I think
probably the best will be the Toshiba equivalents (a little
more expensive)

Thank you for your input Mr. Pass
I have found a quicly searched digikey for a replacement from fairchild and came across FDH15N50 as a replacement.
But even better...
What would you suggest to replace those IRFP's with to drive 4 pairs of thermal tracks ??

 

[AndrewT]

.......as lines of transistors on heatsinks have inner and outer devices where inner ones might be warmer than outer ones.

not if you follow the heatsink manufacturers advice on locating the devices for best heat dissipation.
All the contact interfaces run at the same temperature ensuring the same thermal resistance and thus the same device temperature.

 

[jan.didden]

not if you follow the heatsink manufacturers advice on locating the devices for best heat dissipation.
All the contact interfaces run at the same temperature ensuring the same thermal resistance and thus the same device temperature.

Andrew, if you have several pairs of output devices on a common heatsink you can't avoid some temp differences. The thermal impedance from each device to the heatsink may be equal, but from each point on the heatsink to ambient air will be different. Question is how much, and you can minimize it by intelligent positioning as you noted.

jd

 

[AndrewT]

Hi,
according to the manufacturer's papers I have read, it is very possible to arrange the devices so that each gets cooled equally.

This came up recently in another thread and I tried to find the AAvid paper, but failed. Wakefield also published similar advice.

There is a free software package that allows one to place the devices on one's custom sink and by trial and error home in on the ideal layout for most even device interface temperature. This also gives the optimum heatsink dissipation if the device height is chosen appropriately. Maybe I should state that the other way around; optimising the dissipation gives the most even device temperature.

 

[jan.didden]

Hi,
according to the manufacturer's papers I have read, it is very possible to arrange the devices so that each gets cooled equally.

This came up recently in another thread and I tried to find the AAvid paper, but failed. Wakefield also published similar advice.

There is a free software package that allows one to place the devices on one's custom sink and by trial and error home in on the ideal layout for most even device interface temperature. This also gives the optimum heatsink dissipation if the device height is chosen appropriately. Maybe I should state that the other way around; optimising the dissipation gives the most even device temperature.

Well I would be very interested in that, I don't think it is possible. Minimize yes, depending on heatsink size/topology, but I don't think you can actually make it equal.
But I've been wrong before ...

jd

 

[Nelson Pass]

What would you suggest to replace those IRFP's with to drive 4 pairs of thermal tracks ??

Perhaps Toshiba 2SJ313 / 2SK2013

 

[Bob Cordell]

That may be the case in some instances, such as low bias
designs, but I don't think it's true as a generalization.

Hi Nelson,

I agree completely. There are always limitations with each technology, for both BJTs and MOSFETs; pick your poison wisely. I have long disagreed strongly with Doug Self's generalizations about MOSFETs being way more nonlinear than BJTs.

MOSFETs like more bias current for sure. But I LIKE running the output stage at higher bias. MOSFETs virtually never suffer from higher distortion by being biased hotter; there is no so-called gm doubling as there is with BJTs.

That having been said, I also very much like the ThemalTraks. They have mitigated many of the shortcomings that BJTs suffer from, including thermal bias stability.

Cheers,
Bob

 

[mlloyd1]

good luck finding those.

hmmm, then again, i should have realized Nelson probably have a stash of those somewhere ...

mlloyd1

Perhaps Toshiba 2SJ313 / 2SK2013

 

[Adrculda]

good luck finding those.
mlloyd1

Yes, that what i was thinking when i tossed it into both Digikey and Mouser and came up as discontinued
Closest i found was the 2SJ201 / 2SK1530 ....

 

[jacco vermeulen]

The duo can be bought in small/large numbers from a couple of HK sellers at the E-place.
There are a few North-American semi vendors that sell the J313/K2013, of which B&D in Pennsylvania is the most familiar at the web.(and priciest)

 

[Adrculda]

The duo can be bought in small/large numbers from a couple of HK sellers at the E-place.
There are a few North-American semi vendors that sell the J313/K2013, of which B&D in Pennsylvania is the most familiar at the web.(and priciest)

but why not substitute with a more recent MOSFET that's still in production and is widely available rather than something that has its "niche" market and cost hasnt been marked up 1000% because they are obsolete and sought after items.

 

[Nelson Pass]

I wasn't aware that they were discontinued. Surely you
can come up with a rough equivalent. I mention Toshiba
as the last big vendor who doesn't have an some example
of a device with the "P channel issue".

 

[Adrculda]

After a quick chat with a "engineer" from Natianal Semiconductor he told me that my choice of the LME49830 is poor as it was designed for MOSFET's and should use LME49811 instead as it was meant to drive BJT's.
Seems like this will be a lot easier as i dont have to cascade devices now

 

[jan.didden]

Hi

My quick answer would be that the LME49830 is a MOSFET driver, and it is not meant to drive BJT ops.
[snip]
Cheers

That engineer must have read Stinius' post

jd

 

[Adrculda]

All i got left now is to set up biasing with the thermal track's

 

[Adrculda]

I been going at it most of the day and came up with nothing !!!
i mean the biasing is done internally, so thermal tracking will have to b implemented some other way to keep things in check

 

[Adrculda]

I been going at it most of the day and came up with nothing !!!
i mean the biasing is done internally, so thermal tracking will have to b implemented some other way to keep things in check

Never mind, brain fart....
Its been a long day ( 2 sick kids and nad 3 hrs of sleep on top of that :S )

 

[stinius]

Hi Adrculda

It seems that you have moved back to start.
You will have to answer the questions that Janneman asked you about in his post:

Hi,

You asked a very general question "what do you guys think". Just to get a better understanding of what it is we can help you with, are you comfortable with configuring the bias with the TT diodes? How many diodes are needed, additional resistive elements to get at the target bias voltage? Are you comfortable with calculating the current through the TT diodes and how to relate the TT diode tempco to the output device tempco and how to match those to get good tracking?

jd

Maybe we can help you.

BTW: The biasing is not done internally, but you will have to use some extra transistors to make the circuit working.

Cheers
stinius