H_A_M_P_T_O_N_E jfet amp modules

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
So, building sixty odd of them, any comments on having to trim the design or other concerns? Did you select your transistors or just go with what you had?

Bear
 
Hi al_p
Do not open the file, just save it.

Rafael
Sorry but I forgot to respond to your question, the electrolyt is Panasonic FC.

Bear
It was very dificult to make 60 identical ones. I selected from 300 2n5457 and 200 ZTX653.
The ZTX should not have to big Hfe [around 100, not more than 200] or it will be too hot. You can solve the problem by changing the base resistor of the ZTX to a higher value but I would not recomend to go too high because it will affect the output.
You should first measure all the fets you have and put them in categories. I neede 60 fets to be as close as possible to each other so I measured all the 300.

chrissugar
 
Hi Jaakko

These are fixed gain modules and they work at aprox 26 to 29dB depending on fet and resistors.
They do not work at unity gain. [ use a 26 dB atenuator at the input or output and it is a unity gain amp :green: :green: :green: ]

chrissugar
 
In another thread Sleeper posted a scan of the original article:
http://members.cox.net/soma/hamptone.jpg

chrissugar
 
Just wondering about that pile of 60 units... did you try other caps as well i.s.o. the Wima-MKS ?

BTW, the Hamptone-JFP-article nicely states polypro or better... since the polypro is already about number one on the quick&dirty-cap-ranking-list what would have been the idea ?! :roll:

I'll be using metallized polyester since for now only those easily available (@ RS: Evox Rifa MKK)

Regards,

Peter
 
Wima polyester caps are very good, and the circuit is not exactly a clean one so i think there is no problem to use them.
Many high end manufacturers use these polyester Wimas so no problem. I like very much the sound of this gain block as it is. Tried the prototype with polyprop but the difference is not significant in this case.

chrissugar
 
I didn't use the 15k value because my fets were biased optimally at a diferent value.

What value did you end up using? (can't tell from the pictures)

I've built a couple of these, and they are "crunchier" than I thought they would be. They are "crunchier" with the 470uf cap in place than without it.

I'm thinking of playing around with the bias on the jfet and the ZTX653, any suggestions?
 
[quote author="E-money"]
I didn't use the 15k value because my fets were biased optimally at a diferent value.

What value did you end up using? (can't tell from the pictures)
[/quote]

the value I used is relevant only for my situation. In your case it will be probably another value. Because my intention was to build 60 modules I hat to find 60 JFETs with similar parameters. I selected the 60 from some hundreds and I found that those 60 JFETs needed a different value to work optimal.

chrissugar
 
[quote author="E-money"]I've built a couple of these, and they are "crunchier" than I thought they would be. They are "crunchier" with the 470uf cap in place than without it.
[/quote]

I supose maybe you didn't bias the fet correctly and that's why it is crunchier sound.
Of course they are very coloured, and yes, if you take out the 470micro cap it will sound clean. If you look at this thread:
http://www.groupdiy.com/index.php?topic=4737
you will see that I included a knob named "harmonics".
It is a 10K pot in series with the 470micro capacitor and it will let you choose the apropriate sound for your application from clean to coloured. :grin:
This is one of my favorite preamps for sound design. :grin:

Don't forget that excluding the 470micro capacitor will also lower the amplification of the module from 26-30dB to 16dB depending on the values of the resistors used.

chrissugar
 
[quote author="Ptownkid"]Pardon my ignorance, but what do these plug into to make an actual mic pre?[/quote]

Into this:
http://members.cox.net/soma/hamptone.jpg

If you read this whole thread you have almost all the info needed. Even the link above is posted at page two in this thread.

chrissugar
 
you will see that I included a knob named "harmonics".
It is a 10K pot in series with the 470micro capacitor and it will let you choose the apropriate sound for your application from clean to coloured.
Interesting addition ! A log I guess ?

BTW, you're using output-TXs ?

Bye,

Peter
 
It can be a log pot but also works a linear.

[quote author="clintrubber"]
BTW, you're using output-TXs ?
[/quote]
Yes and no. You already know that I include in all my projects transformer and transformerless out as a sound design option. :grin:

chrissugar
 
Hey,

I asked Chris is this was correct:

jfp_harmonics.JPG


Yes that is the right implementation.
Although the sound is clean when the pot is at max value (left in your picture) and crunch when pot is zero ohm (right in your picture).
Please can you post this pic in the thread so others will understand the idea.

Thanks again Chris! :thumb:
 
[quote author="chrissugar"][quote author="Ptownkid"]Pardon my ignorance, but what do these plug into to make an actual mic pre?[/quote]

Into this:
http://members.cox.net/soma/hamptone.jpg

If you read this whole thread you have almost all the info needed. Even the link above is posted at page two in this thread.

chrissugar[/quote]

Yeah, i read it all, but a lot of it is greek to me. I just got ino electronics 4-5 months ago really. I'm a really fast learner, but some things take some time to wrap your virgin brain around.

Cheers
 
Because I received a lot of mails and PM about selecting JFETs for this module I will post what I wrote in my last mail to a member:


The easyest way to select fets is to power one Jfet amp module at 24V, and do not connect anything to it, (nothing to the input) just a multimeter and measure the DC between output and ground (output before the 10microfarad condenser, not after). It should be exactly 12volts. Probably you will have a different value so one option is to buy more jfets and find one that will give you exactly 12volts at out.

The second option is to replace the 15K resistor with a 10k in series with a 10k linear pot and adjust till you have 12volts at the output. Then disconect the 10k plus 10k pot and measure the total resistance.
You will put in that place a resistor equvalent with the measured value.

The third option is to do the same thing with the 1k8 resistor, you replace it with a 1k resistor in series with a 2k5 pot and adjust to find the right value for 12volts at output and then replace with the measured value.


The third option it is better than the second because it will not influence the gain of the module, only the DC biasing. If you whant to change amplification you can use the second option, it will influence at the same time the DC and gain.

chrissugar

***later correction:
I forgot that on my breadboard I had a 1M resistor for biasing the fet between the fet gate and ground. That fet is necessary for the measurement process.
Sorry, it was long ago.
 
Yes and no. You already know that I include in all my projects transformer and transformerless out as a sound design option.
Sorry Chris, I didn't think enough before typing :roll:
I mean, only just recently there was some discussion about disconnecting the TX when going out direct etc etc.

Bye,

Peter
 
a few tidbits that are different in the current kit production jfet gain blocks:

1) mpsa14 has been changed to a ztx603
2) on semiconductor brand 2n5457s are used which have a lower average hFE than the fairchild ones
3) the output bias set resistor has been raised from 47k/68k to 127k

ed
 
Back
Top