The mains fuse blew immediately :-( I removed the output transistors, and the N-channel BUZ900 seems to have the internal diode reversed! Don`t know if this really is a P-channel or something else grrr...
The P-channel BUZ905 is shorted between D and S.
The P-channel BUZ905 is shorted between D and S.
I removed and checked all transistors before installing the new output transistors. All (except output transistors..) checked good with a multimeter on diode test. Also checked all electrolytic caps. The drivers seem to have survived this time also..
Last edited:
No😱
Not sure it would have helped, but at least I should have checked the reverse diode. I downloaded the datasheets and checked the pinout, and it was the same as on the circuit board. But it doesn`t help if the transistors have`wrong markings..It sounds very strange, but I have heard of similar problems before.
AndrewT will be referring to the strong likelihood that Chinese sourced Lateral Mosfets will be fakes. There are only one or two known manufacturers of the chips that are suitable for audio applications (Renesas) and a few independent assembly brands like Exicon, Alfet, Semelab in the UK. Many people seem to buy the fakes simply because they are much cheaper and it's easier to buy online, without realising they are actually normal Hexfets, not Lateral Mosfets.
When powered up in the circuit, hexfets and other industrial mosfet types usually short the output stage because the pinout is different to other Mosfet types. That's how dumb parts fakers are but they still attract buyers.
If you use a mains bulb tester (or rather a mains bulb limiter) you can spare yourself the agony of more failures and wasted money. A bulb tester by the way, is usually interpreted by search engines as something that either tests the bulb itself, or where the bulb is used as a power/no-power indicator like these examples below. I don't need to say that they are not suitable.
When powered up in the circuit, hexfets and other industrial mosfet types usually short the output stage because the pinout is different to other Mosfet types. That's how dumb parts fakers are but they still attract buyers.
If you use a mains bulb tester (or rather a mains bulb limiter) you can spare yourself the agony of more failures and wasted money. A bulb tester by the way, is usually interpreted by search engines as something that either tests the bulb itself, or where the bulb is used as a power/no-power indicator like these examples below. I don't need to say that they are not suitable.
Attachments
Last edited:
Thanks!
Yes, it could be a fake. I can`t see anything suspicious with the marking, but as I wrote in a previous post, the internal reverse diode is mounted with the anode to the drain terminal. This transistor is marked BUZ900, and should be a N-channel. The other transistor was a MLP120, for which I can`t find any information at all. I replaced it with a BUZ905, assuming that the MLP120 was a P-channel..
Yes, it could be a fake. I can`t see anything suspicious with the marking, but as I wrote in a previous post, the internal reverse diode is mounted with the anode to the drain terminal. This transistor is marked BUZ900, and should be a N-channel. The other transistor was a MLP120, for which I can`t find any information at all. I replaced it with a BUZ905, assuming that the MLP120 was a P-channel..
A Vertical mosFET has the drain in the middle.
A Lateral mosFET has the source in the middle.
The diode between source and drain runs between middle pin and outside pin.
If the Drain and Source are reversed, then the Diode is also reversed.
Could this explain why you measured the diode as being reversed?
A Lateral mosFET has the source in the middle.
The diode between source and drain runs between middle pin and outside pin.
If the Drain and Source are reversed, then the Diode is also reversed.
Could this explain why you measured the diode as being reversed?
Yes, it could be that the drain and source are reversed. How can I tell? A simple test circuit maybe?
Got new transistors yesterday. This time from UK. 10N20 and 10P20. Startet up gently with a DC bench supply with current limiter, connected to the rails. No smoke! Then I removed the DC supply, and connected an auto transformer to the AC input. Increased the voltage a little at a time, and everything seems to work! Connected a signal from my phone, and measured with an oscope at the output. Seems ok! Didn`t try with a real load so far.
- Status
- Not open for further replies.