Github was updated with latest KiCad source files.
These include fine design modifications that make Q17 build even simpler.
I know some of you are struggling to source BS250 for Q17 and this is unavailable at all major distributors.
BS250 it may be replaced for better performance with ZVP3306A and this is still available at https://uk.farnell.com/diodes-inc/zvp3306a/mosfet-p-e-line/dp/9524851?st=zvp3306
Also I want to remind that a group buy is open for Q17 PCB's. Boards are in stock and Saligny Standard active rectifiers as well.
Brgds
These include fine design modifications that make Q17 build even simpler.
I know some of you are struggling to source BS250 for Q17 and this is unavailable at all major distributors.
BS250 it may be replaced for better performance with ZVP3306A and this is still available at https://uk.farnell.com/diodes-inc/zvp3306a/mosfet-p-e-line/dp/9524851?st=zvp3306
Also I want to remind that a group buy is open for Q17 PCB's. Boards are in stock and Saligny Standard active rectifiers as well.
Brgds
Hi Q17 adepts!
I finished my second channel of the Q 17 (otiginal Tibi's), with two output devices,
Glorious music is flowing out of the machine.
What is different comparing with the first one, playing for a few months already?
The heatsinks are about five degrees Celcius warmer than the first build. 35 degrees C in rest. It is built with exactly the same components though.
After hours of examining I apear not to be able to track the reason down.
I checked the power across :
R11=3.55V
R12=3.14V (?! should be slightly higher than R11)
R 3= 325mV (?! should be about the same as R13)
R13= 654mV
Does not look good but please point me where to look at/for?
Thanks in advance...
Ed
I finished my second channel of the Q 17 (otiginal Tibi's), with two output devices,
Glorious music is flowing out of the machine.
What is different comparing with the first one, playing for a few months already?
The heatsinks are about five degrees Celcius warmer than the first build. 35 degrees C in rest. It is built with exactly the same components though.
After hours of examining I apear not to be able to track the reason down.
I checked the power across :
R11=3.55V
R12=3.14V (?! should be slightly higher than R11)
R 3= 325mV (?! should be about the same as R13)
R13= 654mV
Does not look good but please point me where to look at/for?
Thanks in advance...
Ed
Last edited:
Hi Ed,
Glad to hear you love Q17 sound !
It is normal to have different voltage across R11 and R12.
This voltage imbalance is dictated by threshold voltage by N and P MOSFET in output stage. These mosfets are not pair and have slightly different Vgs voltages.
MOSFET threshold voltage is the minimum gate to source voltage (Vgs) required to turn the MOSFET 'ON' (which allows current to start flowing between source and drain). Even MOSFET's are from same source, each will have different threshold voltages, hence your slightly warmer output stage.
R3 voltage is too low. It must be around 0.6V. Check Q9 and Q11, position, type, etc.
Brgds
Glad to hear you love Q17 sound !
It is normal to have different voltage across R11 and R12.
This voltage imbalance is dictated by threshold voltage by N and P MOSFET in output stage. These mosfets are not pair and have slightly different Vgs voltages.
MOSFET threshold voltage is the minimum gate to source voltage (Vgs) required to turn the MOSFET 'ON' (which allows current to start flowing between source and drain). Even MOSFET's are from same source, each will have different threshold voltages, hence your slightly warmer output stage.
R3 voltage is too low. It must be around 0.6V. Check Q9 and Q11, position, type, etc.
Brgds
Hello Tibi!
Thanks for your tips.
I could not find the cause in Q9 and 11 but changed them anyhow. I replaced the IRF610 as well. Problem stays. I ordered some more 910 and 9610 and Q2 and Q10 (2SA970).
I have to wait for a few days since I ordered them today.
A weird thing i have to tell is: as soon as I put the probes on R13 's legs, the fuses blow! Not everytime but it happend a few times. Maybe I indused an instant buzz or something with the touching, and the peak killed the fuse?...Just a laymans guess.
Thank you so far,
Regards,
Ed
Hello Ed,
I don't expect you to believe what I say, as this is about electronic circuits and not about beliefs.
I am sure from my experiments and those of Stef that it is harmonics in the MHz range that produce both the heat and the "wrong" voltages at the resistor. This thesis also explains quite simply the effect that your fuses blow when a very small charge disturbs the system. I have been struggling with this issue from the beginning and have worked out solutions that make the circuit more complex but also more stable.
Therefore: If you have the possibility to measure with an oscilloscope the quiescent oscillation of the amplifier before the output filter, i.e. the voltage between the power MosFET (before the coil) against ground, then this would either prove or disprove my remote diagnosis.
Regards Tim
I don't expect you to believe what I say, as this is about electronic circuits and not about beliefs.
I am sure from my experiments and those of Stef that it is harmonics in the MHz range that produce both the heat and the "wrong" voltages at the resistor. This thesis also explains quite simply the effect that your fuses blow when a very small charge disturbs the system. I have been struggling with this issue from the beginning and have worked out solutions that make the circuit more complex but also more stable.
Therefore: If you have the possibility to measure with an oscilloscope the quiescent oscillation of the amplifier before the output filter, i.e. the voltage between the power MosFET (before the coil) against ground, then this would either prove or disprove my remote diagnosis.
Regards Tim
I noticed something strange on my Power Supply PCB's : On both still unpopulated boards I measure a full short between the plus and minus connections of C2 and C3 as well as the output of the associated bridge rectifier between plus and minus....... I have the Gold plated PCB's
I paid invoice number 0089 on November 27 and you are correct I was not in the GB list. This is a project where a friend of mine will build the Amp with my assistance in the background.
absolutely correct comment, because this other amplifier does not need a Quad 405 analog bridge, because the output stage is not current controlled. In the Quad 405 bridge, the 47 ohm resistor (or 22 + 22 in Q17) is a negative feedback resistor at frequencies near the unity gain frequency, because the inductor helps to reverse this phase. in this case, the output power stage at high frequencies is the current source controlled by the current VAS. When the output power stage is Mosfet, then controlling its current is nonsense.
it is necessary to correctly connect the correction circuits and not to hide the usual technical nonsense behind the concept of audiophile.
Last batch of Active Power supply boards is bad and those who have purchased will get a replacement.
Please check if you have a short at positive rail, between V+ and GND. If so, then you have a board from latest batch.
Issue may be corrected by cutting the short as indicated in attached picture at C3 negative pad.
Apologise for trouble !
Brgds
Hello Tibi/Tim!
Here a litlle follow-up of my adventure...
First I replaced the 992(Q10) for a 970, just to be sure since my boards are stacked with 2sc2240 ones. maybe a better match?
Than I gave the connection of the strap of R32 a better look. I used here a rather thick wire, which did not fit the holes on the board, so I widend the hole a bit, but so much that te connections winth the ground plane was ruined! Silly me! I made some new spots by removing the lacker arround a bit, asuring a proper connection now. No big diffence in mesurements anymore. They look like this:
11 Left= 3.05V 11Right=3.29V
12 Left= 3,43V 12Right=3,11V
3 Left= 768mV 3 Right=655mV
13 Left=630mV 13 Right=661mV
This looks at least a lot better, though the difference between 3 and 13 Left seems a litlle big?
About the difference in sound of the first set Q17, one output pair and this one, two output pairs.
I am missing the easyness of throwing music at you in the double one, untill now.
The one-pair had a really deep, deep (yes two times!) base. Not rumbling, but in a dry, natural and pleasent way. And just a litlle more sparkling in the top end. Mids just in tune...
Cimbals were bronze cimbals. And most of all, very easy and exiting to listen to. Even on lower power. Everything there in balance,
While the two output paired ones sound leaner in the bottom, until now, more shouty in the mids. I need more listening though.
One important crazy thing I have to admit is, when taring down the first amp, I found out that I used a set of IRF610/9610 in stead of FQP3P20/FQP3N30!
I don't know how I came to do so anymore, but I did.
However, it is the amp I listened to for about a year, and I liked it!
Since this is diy, there are endless possibillities to tune and change parts to make everything right for ones ears.
I am sure about that.
Thanks for the help,
I put another nickel in the machine now and drown myself in some glorious tunes!
Thanks for reading...
Measurements are OK.
Thank you very much for your feedback, Ed !
Give him a little more burn-in. It will shine after 300h.
Brgds
Thank you very much for your feedback, Ed !
Give him a little more burn-in. It will shine after 300h.
Brgds
Thanks Tibi, I know I should be more patient
ly!
I have one question
In my right channel R29/30 went really hot, not touchable by a finger. I had chosen 100 oHm.
After changing to 1K, the right amp did not start anymore, no leds burning, Left channel did work.
After parallelling with 10oHm everything went smooth again and the resister does not get hot anymore,,,, Strange but it works.
Something to look at?
Regards,
Ed
R29 and R30 should not be hot at all. Change Q13 or Q14, or look at C13 or C14 polarity ...
Brgds
Brgds
Hi Stef
I know there is 1uf in the schematic. I used 2.2uf as I don't have a 1mf capacitor. I haven't had any problems with the sound.
I like purple. 🙂
Both 15V zeners had a short. Can I replace them wth a 0,4 Watt one? Special that he made music....?