Guys, Per-Anders has pointed out that the HV MOSFET symbol in the published schematics might be incorrect.
So to be clear: the IXTP08N100D2 is an N-channel depletion mode MOSFET.
The data sheet uses the attached symbol.
At any rate, the PCB is correct for the given device, of course.
Edit: attachment doesn't work for some reason. But here's the data sheet:
https://www.littelfuse.com/~/media/...l_depletion_mode_ixt_08n100_datasheet.pdf.pdf
Jan
So to be clear: the IXTP08N100D2 is an N-channel depletion mode MOSFET.
The data sheet uses the attached symbol.
At any rate, the PCB is correct for the given device, of course.
Edit: attachment doesn't work for some reason. But here's the data sheet:
https://www.littelfuse.com/~/media/...l_depletion_mode_ixt_08n100_datasheet.pdf.pdf
Jan
Last edited:
Update: I have been informed that the PCB production with my PCB manufacturer in China has been resumed on 13 Feb. Approx. 40% of their employees are still not able to work due to safety measures for the Corona virus.
They restart production of small orders (< 50) on a first-come first-serve basis.
My last order is from end of January so I hope I am at or close to the top of the pile.
Jan
They restart production of small orders (< 50) on a first-come first-serve basis.
My last order is from end of January so I hope I am at or close to the top of the pile.
Jan
Hello
Wondering if boards are available and if the group buy is proceeding.
If yes, I would like to add my name (wct) to the list for 1 board.
I bet in looking through the thread I missed the formal way to join the group
buy, my apologies if that is so. No doubt there must be ship fees and perhaps
paypal fees.
Thanks for doing the group buy.
curt
(central coast California)
Wondering if boards are available and if the group buy is proceeding.
If yes, I would like to add my name (wct) to the list for 1 board.
I bet in looking through the thread I missed the formal way to join the group
buy, my apologies if that is so. No doubt there must be ship fees and perhaps
paypal fees.
Thanks for doing the group buy.
curt
(central coast California)
Power is voltage across the pass device x current through the device.
I have several boards with parts ready to ship. I am not sure I have correctly kept track of who wanted what over the last few months (Alberto I've got you covered, Curt you got PM).
I know some of you have already paid. So please remind me.
Jan
I have several boards with parts ready to ship. I am not sure I have correctly kept track of who wanted what over the last few months (Alberto I've got you covered, Curt you got PM).
I know some of you have already paid. So please remind me.
Jan
Oh, I see, a minimum load is required, and the pushbutton is an alternate way to reset the regulator, apart from removing the load. But the regulator without load always starts at full output, so the load has to be connected before the input is powered. What an interesting approach to avoid the usage of a couple bias resistors or current sources and a normal enhancement MOSFET.
Oh, another thing I do not understand is why one would use a 80Mhz op-amp to end up with 1uF output capacitor.
Oh, another thing I do not understand is why one would use a 80Mhz op-amp to end up with 1uF output capacitor.
Last edited:
Yes there is a minimum load of about 3 mA. If you worry about that a single load resistor that draws 3mA will do.
The opamp is not selected because of its bandwidth, but because of its ability to work at low supply voltage and low supply current. Other such opamps might also work.
That 1uF output cap comes from the fact that 600V electrolytics that are reasonable in size are not that common.
There are many design choices that are not obvious from a schematic diagram.
One goal of this was to provide a supply where the output can be set with a single resistor. My experience with the SuperRegulator was that even with the very simple way to set Vout by the ratio of two resistors, many users find that complicated and I get many, many emails on how to calculate those resistors. That gets boring very fast ...
Sometimes your design ego must take second place after ease of use by the user. ;-)
Jan
The opamp is not selected because of its bandwidth, but because of its ability to work at low supply voltage and low supply current. Other such opamps might also work.
That 1uF output cap comes from the fact that 600V electrolytics that are reasonable in size are not that common.
There are many design choices that are not obvious from a schematic diagram.
One goal of this was to provide a supply where the output can be set with a single resistor. My experience with the SuperRegulator was that even with the very simple way to set Vout by the ratio of two resistors, many users find that complicated and I get many, many emails on how to calculate those resistors. That gets boring very fast ...
Sometimes your design ego must take second place after ease of use by the user. ;-)
Jan
Last edited:
Oh, voltage adjustment, fine signal circuitry.
I mean that with an op-amp that is good to 80Mhz and a pair of MOSFET that are also good to 80Mhz the bandwidth of the circuit could surely be improved. A simple calculation reveals an output zero (from 1uF, 2.2r) of 1/(2*3.1415*1e-6*2.2)~=72Khz. That could be 100nF and 720Khz, or even 7Mhz, with a current source for minimum load and control of stability/bounce. Olympic design.
EDIT: Obviously the PCB is not good up to 80Mhz.
I mean that with an op-amp that is good to 80Mhz and a pair of MOSFET that are also good to 80Mhz the bandwidth of the circuit could surely be improved. A simple calculation reveals an output zero (from 1uF, 2.2r) of 1/(2*3.1415*1e-6*2.2)~=72Khz. That could be 100nF and 720Khz, or even 7Mhz, with a current source for minimum load and control of stability/bounce. Olympic design.
EDIT: Obviously the PCB is not good up to 80Mhz.
Last edited:
I have updated the circuitry for the last batch of boards, so that you need to select only one resistor value to set the current limit, instead of two values.
The schematic is here: https://linearaudio.nl/sites/linearaudio.net/files/TS-4-b schematic.pdf .
The writeup is here: Access denied | Linear Audio .
Jan
The schematic is here: https://linearaudio.nl/sites/linearaudio.net/files/TS-4-b schematic.pdf .
The writeup is here: Access denied | Linear Audio .
Jan
Shipping update:
Belgium Post send out the following message:
Shipping of letters and small packages: shipping within Europe as normal but with possible delays, especially France, Italy, Malta and Russia.
Shipping outside Europe/oversees at this point halted due to lack of flights. Shipments will remain in Belgium until such time that flights become available again.
I recently send out several packages outside Europe, so if you are waiting for it, it may be (much) longer than expected. I apologize but cannot change it at this time.
From now on I will not accept orders for delivery outside Europe unless you specifically agree to wait for delivery whatever the delay.
Sorry about all this, it is just something we need to get through.
Jan
Belgium Post send out the following message:
Shipping of letters and small packages: shipping within Europe as normal but with possible delays, especially France, Italy, Malta and Russia.
Shipping outside Europe/oversees at this point halted due to lack of flights. Shipments will remain in Belgium until such time that flights become available again.
I recently send out several packages outside Europe, so if you are waiting for it, it may be (much) longer than expected. I apologize but cannot change it at this time.
From now on I will not accept orders for delivery outside Europe unless you specifically agree to wait for delivery whatever the delay.
Sorry about all this, it is just something we need to get through.
Jan
Jan,
I have some trouble with the Voltage Regs.
Version T-Reg TS-3
I have a row supply with about DC 275 V
Reg is configured to deliver 170 V
R 11 is 330 K I tested 270 K as well
R6 is 6,8R and R8 is 2,2 R
LED is on, but the voltage doesn't drop to 170 V.
Vin and Vout are mostly similar.
Debugging I have done so far:
checked the position of all parts
Measured all Rs and Cs, as well Diodes.
Exchanged all transistors (4xBC and FDP as well as IXTP)
Any idea what could be the reason? Or any hint whatelse I can try?
I have some trouble with the Voltage Regs.
Version T-Reg TS-3
I have a row supply with about DC 275 V
Reg is configured to deliver 170 V
R 11 is 330 K I tested 270 K as well
R6 is 6,8R and R8 is 2,2 R
LED is on, but the voltage doesn't drop to 170 V.
Vin and Vout are mostly similar.
Debugging I have done so far:
checked the position of all parts
Measured all Rs and Cs, as well Diodes.
Exchanged all transistors (4xBC and FDP as well as IXTP)
Any idea what could be the reason? Or any hint whatelse I can try?
Hi resets,
two pointers come to mind:
- The regulator needs a minimum load to work, maybe even a minimum load is necessary that it doesn't break. I think Jan mentioned at least 3 mA. I did my tests with a load of about 10 mA - just put a 22k power resistor or similar across the outputs.
- if you operated without load and had some other things coming together, maybe the AD8031 is broken. Try changing that as well. No voltage difference (or a very small one) points to the AD8031 as a possible culprit, as a shorted AD8031 provides a path from input to output.
I built two regulators, and actually managed to break the first one in the beginning. In my case, I had two things going on:
- a spurious short of the FDP to the heatsink. I actually measured for shorts, but with the meter set for 200 ohms range. When I later discovered the connection between FDP and heatsink, it measured something like 350R
.
- when I first tried the regulator, with a 22k resistor load, all worked fine. But when I connected the ground to the chassis, and at the same time didn't have a load connected on that test, but 80 uF PP capacitance on the output, the regulator failed when I turned the variac to more than about a hundred volts.
I deduced that when the FDP shorted to the heatsink / ground, the input voltage dropped to below the output voltage, and the capacitors at the output reverse discharged through the circuit, taking the AD8031 with it. This was actually with the 2.7V Zener mod installed ...
It took me a while of head-scratching - I first suspected a broken FDP because of no voltage differential between drain and source in circuit ... but it tested fine out-of-circuit. After testing various segments of the circuit in isolation, it showed that really only the AD8031 was broken ... a new AD8031 fixed the regulator. I actually changed C2 as well, to be on the safe side, because it likely had gotten reverse voltage, but that tested and looked fine in the end (I had a higher voltage rated cap there in the beginning).
Otherwise, the two regulators are working great now - one supplies my EL84 PP amp, the other my Aikido preamp.
Thanks Jan, for the design and the boards !
Best regards,
Claas
two pointers come to mind:
- The regulator needs a minimum load to work, maybe even a minimum load is necessary that it doesn't break. I think Jan mentioned at least 3 mA. I did my tests with a load of about 10 mA - just put a 22k power resistor or similar across the outputs.
- if you operated without load and had some other things coming together, maybe the AD8031 is broken. Try changing that as well. No voltage difference (or a very small one) points to the AD8031 as a possible culprit, as a shorted AD8031 provides a path from input to output.
I built two regulators, and actually managed to break the first one in the beginning. In my case, I had two things going on:
- a spurious short of the FDP to the heatsink. I actually measured for shorts, but with the meter set for 200 ohms range. When I later discovered the connection between FDP and heatsink, it measured something like 350R
.- when I first tried the regulator, with a 22k resistor load, all worked fine. But when I connected the ground to the chassis, and at the same time didn't have a load connected on that test, but 80 uF PP capacitance on the output, the regulator failed when I turned the variac to more than about a hundred volts.
I deduced that when the FDP shorted to the heatsink / ground, the input voltage dropped to below the output voltage, and the capacitors at the output reverse discharged through the circuit, taking the AD8031 with it. This was actually with the 2.7V Zener mod installed ...
It took me a while of head-scratching - I first suspected a broken FDP because of no voltage differential between drain and source in circuit ... but it tested fine out-of-circuit. After testing various segments of the circuit in isolation, it showed that really only the AD8031 was broken ... a new AD8031 fixed the regulator. I actually changed C2 as well, to be on the safe side, because it likely had gotten reverse voltage, but that tested and looked fine in the end (I had a higher voltage rated cap there in the beginning).
Otherwise, the two regulators are working great now - one supplies my EL84 PP amp, the other my Aikido preamp.
Thanks Jan, for the design and the boards !
Best regards,
Claas
Claas,
Jan,
I tested it with a load at the output. It is a complete preamp behind the T-Reg which works with a "Langsregler IRF 840" pretty well.
The diference voltage between in and out of the T-Reg is around 6 V. Can't adjust it to more.
The crazy thing is, after I had changed all 4 BC Transistors it worked at the first startup. and I could adjust it to 170 V. Later I tried it again, but it failed.
Jan,
I tested it with a load at the output. It is a complete preamp behind the T-Reg which works with a "Langsregler IRF 840" pretty well.
The diference voltage between in and out of the T-Reg is around 6 V. Can't adjust it to more.
The crazy thing is, after I had changed all 4 BC Transistors it worked at the first startup. and I could adjust it to 170 V. Later I tried it again, but it failed.
Last edited: