The MONGREL (supersym II)

[keantoken]

I modified the 1381/3503 models because they caused simulator errors with a certain circuit. I don't remember what I did. They are only Andy_C's if they have the _x suffix. I should have been more clear.

ARGH, I included the 15033/34, instead of the 34/35. Yes!!! Here is the proper include file uploaded.

- keantoken

[sng001]

Hi OS,

I finally received my parts from Farnell today. I found one problem with the 0.68uF ceramic capacitor that I ordered for C8 & C10.

Manufacturer: KEMET
Order Code: 1168237
Manufacturer Part No: C330C684M5U5TA

Description CAPACITOR CERAMIC 0.68UF, 50V, Z5U, RADIAL
Dielectric Characteristic:Z5U
Capacitance:0.68µF
Capacitance Tolerance:± 20%
Voltage Rating:50V
Temperature Coefficient:+22%, -56%
Series:Gold Max

=-=-=

The .68uF Z5U cap is smaller than 0.01uF COG and it look like a monolithic ceramic cap.

My question are:
(1) What is the part number for your 0.68uF ceramic cap?
(2) Is it critical to use multilayer ceramic in this position?
(3) Is the value critical for C8 & C10?

I found a few 0.47uF 100V mylar caps in my parts bin, I may give them a try if they are suitable.

Attached is a picture showing the relative size of 0.68uF & 0.01uF.


Cheers, Stanley

[ostripper]

Hi, sng. I used the .68uf's as this dropped the HF distortion of the GX from .01% to .002% as the bypass of the hawksford's emitter resistors. Their value is not truly critical as the effect (lower distortion) "kicks in" after .1uF. Your .47's should do nicely. I used this mouser part... DME1P68K-F Cornell Dubilier Polyester Film Capacitors .

Voltage wise you could even use a 25v device as there is only a volt max across the Re of the hawksford.

I still plan further GX development (swapping parts or changing values) mainly playing with the "magic resistor" values (R20/24) maybe even jumpering them to see what the audible effects are. When I have 2 low power PB60's in the "lab" , I can go crazy with the mods. This will be better as I will not be experimenting on a 250w unit (PB250- scary power and voltages) , I feel much more comfortable with 45-0-45VDC and only 1 pair OP'S !!

Don't worry , the gx is as predictable as a simple bootstrap amp , and as reliable. I have successfully tested it with .9 - 1.7ma tail current , 3- 12ma VAS current (using the trimmers). the powerboard Vbias has enough range to accommodate these conditions but always start with the ops underbiased. The absolute best way to fire these up for the first time is to run the driver pair with 2 -33R temporary resistors tacked to the out/NFB point , adjust vbias for + - .5V ... then install OP's.

BTW your .47uF is in the wrong place .. it is where the trimmer should go!!
OS

[ostripper]

SNG , what are those right angle edge connectors you use on the GX's ?? Doesn't have to be a part # , just what are they called ??
OS

[sng001]

Originally Posted by ostripper

SNG , what are those right angle edge connectors you use on the GX's ?? Doesn't have to be a part # , just what are they called ??
OS

Hi OS,

Thanks for the reply. I googled the Kemet part number, it is a multilayer ceramic (Z5U is less stable than C0G).

I salvaged the right-angle edge connector from the JVC amp. It was a 3-pin 0.1inch-pitch, I converted to 2-pin 0.2inch-pitch:-;

The closest replacement is molex 10-06-1025, which is still available on Farnell. MOLEX|10-06-1025|HEADER, R/A, SQUARE PIN, 0.2" | Farnell Australia , but Mouser does not stock this part.

The Molex 90121-0763 (Headers & Wire Housings 2.54MM CGRIDIII HDR 3P R/A SR SEL AU ), Mouser part number 538-90121-0763 may also be modified to fit.

I attached a picture of the JVC amp board, it was labeled as TP703 (Test Point), which fit for our purpose exactly.

Cheers, Stanley

[ostripper]

Thanx , 6 less screws to tighten down and 3 less euro connectors. Very good..
OS

[sng001]

Hi OS,

I found one error in the GX-VB overview. The polarity of LED (D5) is wrong, the anode of D5 should be connected to base of Q10. I found that out after power on the GX board, the D5 did not lit and that did not cause any disaster.

BTW, the green LEDs (D4 & D5) are only dimly lit. I measured the voltage drop across the LEDs:
Both D4 & D5 (Green LED) are 1.863V
D3 (I used a standard RED LED) is 1.860V

I tested the LEDs on a breadboard, the Green LED gives the following forward voltage drop:

1mA 1.864V
4.6mA 1.964V
15.4mA 2.128V

The RED LED gives the following forward voltage drop:

1mA 1.857V
4.6mA 1.960V
15.4mA 2.196V

Can you please measure your voltages with your GX1+PB60 at 45V supply?

Cheers, Stanley

[keantoken]

Using a + instead of "A" for Anode would be less confusing to some of us, who never learned the difference between Anode and Cathode.

- keantoken

[Holger Barske]

Originally Posted by sng001

I salvaged the right-angle edge connector from the JVC amp.

Don't forget to pull those 2SA992/2SC2240 .

[sng001]

Originally Posted by Holger Barske

Don't forget to pull those 2SA992/2SC2240 .

Hi Holger,

I used "recycled" 2SA970 & 2SC2240 transistors on the GX-VB. Four of 2SC2240s are B0 grade & they were Hfe matched at the factory.

I also recycled the output coil & the MPC722 resistors on the PB120, the MPC722 resistors are compact & non-inductive.

I used the money saved to buy some premium resisters;-)

Cheers, Stanley