[Danster.dk]

Hi there.


I only need to match 2sk170 and the 2sj74 before my amp is done! But how do I do it?

It's for the gilmore class A headamp...
Should I place the transistors same as the 2sj109 and 2sk389?

Kind regards

Danster

 

[ericj]

If you have some protoboard and a battery there is a simple circuit you can build to test the IDSS of jfets. It's described on tangent's site, I don't have the url handy.

 

[MisterX]

http://www.diamondstar.de/transistor_matching_jfet.html

 

[Danster.dk]

Quote:

Originally Posted by MisterX /img/forum/go_quote.gif http://www.diamondstar.de/transistor_matching_jfet.html

And then I should find the transistors which give the same output current?
Is this method exact?

 

[Pars]

Quote:

Originally Posted by Danster.dk /img/forum/go_quote.gif And then I should find the transistors which give the same output current? Is this method exact?

How many JFETs do you have? Match them as closely as possible... both pairs of J74s or K170s, then those pairs to the opposite pair, per channel. Nothing in analog (or digital, for that matter) is ever exact...

 

[Danster.dk]

I have:

15 x 2sk170
15 x 2sj74

Im sure if I get it right, lets make an example:

I find 2 x 2sk170 which both have an output of 2mA (just a random number)
When Im matching the 2sj74s, should I then those the two which are closest to 2mA, or just pick the 2sj74s which are closest to each other?

When I have found 2 x 2sj74 and then are going to put them on the board instead of the 2sj109, should I turn one of the 2sj74 around, or?

Kind regards

Danster

 

[amb]

Quote:

Originally Posted by Danster.dk /img/forum/go_quote.gif I find 2 x 2sk170 which both have an output of 2mA (just a random number) When Im matching the 2sj74s, should I then those the two which are closest to 2mA, or just pick the 2sj74s which are closest to each other?

For the Dynalo (and Dynahi) amps, just match the N-channel pair to each other, and the P-channel pair to each other. It is not necessary to match the across the N- and P-channel devices, because they are independently loaded with their own CCS.

Quote:

When I have found 2 x 2sj74 and then are going to put them on the board instead of the 2sj109, should I turn one of the 2sj74 around, or?

Yes. You should have a look at the datasheets for each of these devices and be able to answer this question for yourself.

 

[Danster.dk]

Thanks for all the replays.
Im done with the channels now, but when I hook one channel to the PSU, the output voltage DC is almost 300mV and I have matched all the components :S
Have some one made a guide, which explains what the voltage over different components should be?
I have searched the forum, but I couldnt find it.

Thanks guys!

 

[Pars]

Take a look at http://www.djgardner.com. There is a construction manual which has a voltage chart in it. You need to start a the CCS stage and work towards the outputs. 300mV isn't actually that bad for a start.

 

[Danster.dk]

Maybe Im blind but I cant find the voltage chart?

I will just give you guys a little brief:

When I hook it up, the DC output, slowly raises, its starts at 150mV DC, and like every 2 second the DC gets 1mV higher.
And after a few minuts its 300mV (still encreases).
The strange thing is, when Im holding my fingers on the 2sk170s, the DC voltage slowly drops, and it even gets in - DC. Do you know what can be wrong? I cant seem to find any short circuit.
Have I turned the transistors wrong?
When I meassured the LEDS, the voltage across them, where about 1,596V, but when the board is hooked up, the drop is only 1,56XV
Here's a pic:

Thanks!

 

[Pars]

Try http://www.djgardner.com/headphone/g...g%20Points.pdf. A head-fi thread is referenced on Dan's site (there should be several threads here and at headwize regarding the Dynalo, both home etched or pt-pt construction, as well as on the Rev. B and Rev. C boards from the group buy). Dan never did put the more or less final edit of the construction document up on his site... all the ones listed were preliminary edits.

Also, do all of your preliminary adjustments without the servo opamps in. Get the offset as low and as stable as possible; then put the servos in.

The transistors will change with heat, so your observations WRT putting your finger on one and seeing readings change is normal. It is also normal for the readings to drift upwards and downwards as the parts heat up. They should get more or less stable at some point, and it is the servo's job to keep them that way.