I've been trying to build a Pass Zen v8 in a 80-ies Yamaha A-700 chassis as my first attempt at building an amplifier from basic components. When putting the plug in the AC socket (which was admittedly a bit scary), there was no magic smoke, no loud bang and the positive rail indicated +48.5V DC. Pffew!
Per instructions I then slowly increased the resistance at trimmer RV1 (called P1 in figure 10 of https://www.passdiy.com/project/amplifiers/zen-variations-8), and low and behold: I arrived at +25V at point J1. This is also good and made the lamps glow and a current of 2.01A was flowing through the lamps into the drain of Q3 (the IRFP 150N Mosfet), making the heatsink hot, but not scorchingly hot. So far, so good....
The problem I have is with the 9v indicated at J2 (which is the gate voltage for the Q3 Mosfet): I cannot get anywhere near the 9V I should get....
When I have 25V at J1, I only get 4.7V at J2 instead of 9V.
At point J3 (source of Q3 and drain of the LU1014D power FET) I only get 0.5V instead of the 4.3V I should get.
Increasing the resistance at RV1 will decrease the value at J1 and raise J2 a little, but even if I turn RV1 all the way up, I won't even get to 6V on J2.
I tried using only one lamp which would give about 23,4Ω resistance and 1A of current through Q3 instead of 2.01A, I've changed R1 into 47K and 20K, but nothing seemed to help: I never seem to get J2 at even 6V, let alone 9V.
Maybe I made a stupid mistake that I keep overlooking... Anyone any ideas?
NB: I bought the LU1014D power FETs on Ebay and I don't know if they are genuine and if that may be causing trouble. All other components are from respectable vendors.
Per instructions I then slowly increased the resistance at trimmer RV1 (called P1 in figure 10 of https://www.passdiy.com/project/amplifiers/zen-variations-8), and low and behold: I arrived at +25V at point J1. This is also good and made the lamps glow and a current of 2.01A was flowing through the lamps into the drain of Q3 (the IRFP 150N Mosfet), making the heatsink hot, but not scorchingly hot. So far, so good....
The problem I have is with the 9v indicated at J2 (which is the gate voltage for the Q3 Mosfet): I cannot get anywhere near the 9V I should get....
When I have 25V at J1, I only get 4.7V at J2 instead of 9V.
At point J3 (source of Q3 and drain of the LU1014D power FET) I only get 0.5V instead of the 4.3V I should get.
Increasing the resistance at RV1 will decrease the value at J1 and raise J2 a little, but even if I turn RV1 all the way up, I won't even get to 6V on J2.
I tried using only one lamp which would give about 23,4Ω resistance and 1A of current through Q3 instead of 2.01A, I've changed R1 into 47K and 20K, but nothing seemed to help: I never seem to get J2 at even 6V, let alone 9V.
Maybe I made a stupid mistake that I keep overlooking... Anyone any ideas?
NB: I bought the LU1014D power FETs on Ebay and I don't know if they are genuine and if that may be causing trouble. All other components are from respectable vendors.
Last edited:
exact value of mosfet gate voltage is irrelevant, except as controlling figure in case of troubleshooting
what you need to chase is voltage across JFet itself
though, not in that case - both these heatsinks together are not enough for 50W of heat, even if leaving case without cover
if you want A Class amp, you must invest in proper PSU and - if not chunky heatsink, then moderate heatsink with forced air thorugh - so fan(s)
but - first thing first - if you go to FW site, you'll find F3 article and there is explained how to match LU ........ which is perfect way of knowing do you have genuine ones
what you need to chase is voltage across JFet itself
though, not in that case - both these heatsinks together are not enough for 50W of heat, even if leaving case without cover
if you want A Class amp, you must invest in proper PSU and - if not chunky heatsink, then moderate heatsink with forced air thorugh - so fan(s)
but - first thing first - if you go to FW site, you'll find F3 article and there is explained how to match LU ........ which is perfect way of knowing do you have genuine ones
The PSU toroidal transformer is rated 800VA so should be good, but the heatsinks may indeed need to grow, however first I would like to get the circuit working.
The voltage across the JFET is just 0.02V as the voltage at the drain is 0.48V and 0.46V at source.
If I could get the voltage at J2 up, it will probably also go up on the drain of the JFET and the voltage across the JFET will rise accordingly. But as stated, I don't know how to sufficiently raise the voltage at J2 (or across the JFET)....
In the mean time I will try to find out if the jfets are genuine, or at least n-channel depletion mode jfets (I found this article: jfet-matching)
The voltage across the JFET is just 0.02V as the voltage at the drain is 0.48V and 0.46V at source.
If I could get the voltage at J2 up, it will probably also go up on the drain of the JFET and the voltage across the JFET will rise accordingly. But as stated, I don't know how to sufficiently raise the voltage at J2 (or across the JFET)....
In the mean time I will try to find out if the jfets are genuine, or at least n-channel depletion mode jfets (I found this article: jfet-matching)
Last edited:
My (cheap) transistor tester says: N-JFET, I=2.1mA, Vgs=1.4V, which seems very high considering the distribution Nelson posted in the thread here (highest value being 1.15V in his sample).
When I measure using the schematic on the same thread, I initially measure only 0.75V (increasing as the temp on the Jfet rises) which seems very low.
Below pictures of my (fake?) LU1014D
BTW: Even without the LU1014D I cannot get anywhere near 9V on the gate, try for yourself at https://www.circuitlab.com/circuit/w78qexd53wvq/test-2/
When I measure using the schematic on the same thread, I initially measure only 0.75V (increasing as the temp on the Jfet rises) which seems very low.
Below pictures of my (fake?) LU1014D
BTW: Even without the LU1014D I cannot get anywhere near 9V on the gate, try for yourself at https://www.circuitlab.com/circuit/w78qexd53wvq/test-2/
Last edited:
If you have 25V across the voltage divider (33.1k + 20k ohms), then the current is 25/53.1mA = 0.47mA. That should make the voltage to the MOSFET gate = 20k x 0.47 = 9.4V.
If you can only get 5V, then some of the current going through the 33.1k ohm resistor ((25V - 4.7V) / 33.1k) is not going through the 20k ohm trimmer (4.7V/20k), and is going somewhere else.
If you can only get 5V, then some of the current going through the 33.1k ohm resistor ((25V - 4.7V) / 33.1k) is not going through the 20k ohm trimmer (4.7V/20k), and is going somewhere else.
The problem is that when the potentiometer is turned up further towards max. (20k) value, the voltage at the MOSFET gate rises --> the current through the lamps and MOSFET is becoming higher --> the voltage drop across the lamps increase --> there will no longer be 25V at the start of the voltage divider, but a lower value --> the voltage divider will also have a (considerable) lower value than 9.4V in this case.
As soon as I increase the value at the potentiometer (RV1) the voltage at the MOSFET drain drops further and with that also the voltage at the MOSFET gate.🤔
The PCBs are designed according to the schematic, so if the schematic is ok, the pcb should be ok as well. Following the connections on the pcb, I cannot find a discrepancy with the schematic.
Maybe I need to look again tomorrow, with fresh eyes.
As soon as I increase the value at the potentiometer (RV1) the voltage at the MOSFET drain drops further and with that also the voltage at the MOSFET gate.🤔
The PCBs are designed according to the schematic, so if the schematic is ok, the pcb should be ok as well. Following the connections on the pcb, I cannot find a discrepancy with the schematic.
Maybe I need to look again tomorrow, with fresh eyes.
you need bulbs having proper reactance at intended Iq
that would be approx. 25V voltage sag and current of 2A, so 12R5 as apparent resistance
when not having exact lamp from origin schematic, that could be hen chasing a bob game, needing some luck
and then, any DC source of 24Vdc capable of 2A at least will help you in fast test
only when you find proper bulbs (combo) you can proceed with amp build
that would be approx. 25V voltage sag and current of 2A, so 12R5 as apparent resistance
when not having exact lamp from origin schematic, that could be hen chasing a bob game, needing some luck
and then, any DC source of 24Vdc capable of 2A at least will help you in fast test
only when you find proper bulbs (combo) you can proceed with amp build
You could increase the JFET source resistance. That would increase Vgs, lower the current, and increase Vds.
Hi @Zen Mod, I measured the current through the lamps at intended voltage and R was 11.7 Ohm, so that should be ballpark OK (original design mentions 11 Ohm across the lamps).
@pinholer, that's a great suggestion to increase source resistance. In a simulation that seems to solve the problem, so I ordered some extra resistors. As soon as I have them, I will experiment with values between .5 and 2.5 Ohm and report on the progress!
@pinholer, that's a great suggestion to increase source resistance. In a simulation that seems to solve the problem, so I ordered some extra resistors. As soon as I have them, I will experiment with values between .5 and 2.5 Ohm and report on the progress!
Hmmm, I raised the jfet source voltage from .5 to 1.2 Ω and now I get 2.44V at the drain and 2.42V at the source (so about 2V more than before), but it (still) seems there is practically no current going through the jfet (since there's no voltage drop across it). As the Mostfet does generate heat, and it can only run its current via the Jfet to ground, I am not sure what is happening...
If I know more I'll post an update.
If I know more I'll post an update.
if you test properly your JFets and it shows that you don't have proper part, you can't expect to make functional amp with non- functional parts
ask around or look directly in Swap Meet, those are not so scarce
when you get genuine parts, go further with your build
ask around or look directly in Swap Meet, those are not so scarce
when you get genuine parts, go further with your build
If you get 2.4V at 1.2 ohm resistor to ground, then you have 2.4/1.2 = 2 amps through the resistor and also the JFET. It seems like the JFET is shorted or fully turned on.
Did you check voltage at the JFET gate? Just to cover all possibilities, measure Vgs.
I'm away for a few days, but I have a few spare curve traced LD1014Ds and I can send you two when I get back.
Did you check voltage at the JFET gate? Just to cover all possibilities, measure Vgs.
I'm away for a few days, but I have a few spare curve traced LD1014Ds and I can send you two when I get back.
Gate voltage is 0V, so Vgs is 2.4V.
A shorted Jfet could indeed be the cause! That would explain the (almost) exact same voltage on drain and source AND the fact that the mosfet is conducting.
Thanks for the kind offer, pinholer but earlier today I already ordered some Lu1014D's from member mbrennwa.
Thanks again for the help and I will continue investigating after Christmas!
A shorted Jfet could indeed be the cause! That would explain the (almost) exact same voltage on drain and source AND the fact that the mosfet is conducting.
Thanks for the kind offer, pinholer but earlier today I already ordered some Lu1014D's from member mbrennwa.
Thanks again for the help and I will continue investigating after Christmas!