Roender's FC-100 prototype and builder's thread

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
BJT Matching circuit troubles

All,
I've set up and measured Hfe for my NJL0281 transistors per the circuit in the build guide. I used a bread board, everything went smoothly and the numbers seem sensible to me.

However, when I set up the circuit to measure Hfe for the NJL0302, it doesn't seem to work. Here are the relevant voltages (all with respect to ground)

Power supply voltage: 15V
Vemitter of DUT (NJL0302) 15V (per the schematic)
Vbase of DUT 15V (i.e. no current flowing into base, thru 2k resistor)
Vcollector of DUT 0.3V
Vbase of MJE15030 1.03V
Vcollector of MJE15030 0.3V
Vemitter of MJE15030 0.3V

I have measured the hfe of the MJE device and it checks out (150)
If I pull the MJE from the circuit, then I get simply a series LED and 470R resistor, the LED lights up as it should.

Any ideas as to what I am doing wrong? It's not complicated. Again, it's on a bread board and I successfully measured the 0281 devices. I did rebuild the circuit (R6 in different position) per the builders guide for the the 0302 devices.

Thanks in advance.
Ryan
 
Andrew,
I wired it per the construction manual which doesn't have -15v on the supply. It does change the position of the bjt so the relative emitter and collector voltages changes from PNP to NPN.

I retreated and did the hfe testing as described here:

Transistor matching

I. e. with a passive network of resistors and all the measurements came out sensible i.e. Hfe in the 110 to 125 range. I can only guess that I had some bad connections on the breadboard that I was too impatient to find.

Thanks
Ryan
 
Finally i bought a infrared thermometer and i was able to measure my FC-100 temperatures.
At 26-27 celsius degrees heatsinks measure around 50 deg on outside.
Inside highest temp i measure on shunts for frontend a bit higher than 60 deg celsius.
Is not a bit too much ?
 
Running away

Hi I have built my fc-100 this time I have taken great care testing after fitting each part ,both boards look very neat.:)
Except i have a problem i can get 4 thermaltraks to work at 36 mv but when i try 5 with the pot at zero i power on it shoots to 40 + and would keep on going up,this is for both boards.:(
On reading back Rudi had this problem he used a bat42 to fix it ,However if i changed a pair of thermaltraks would this help ?
regards john
 

Attachments

  • DSCF0005.JPG
    DSCF0005.JPG
    130.6 KB · Views: 441
Last edited:
I have disabled each thermaltrak in turn I cannot get 5 to work they all act the same,so I fix in a bat42 as per the manual and it works ,its stable I am happy :).
However I have a question I cannot find in the manual what do I set the pot to,i have so far set it to 21mv but should it be 36/38mv thanks john
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.