Hello all,
i have my amp for 8 years ,its a class A 150w which uses 5x2sk
1058 and 5x2sj162 in the current gain stage.
My questions are:
1) Is there a need to change the mosfets with new ones?
2) If i do so,do i have to match them ?
I read in another thread that there isnt any need to match the specific mosfets .Is it true?
regards
Nikos😕
i have my amp for 8 years ,its a class A 150w which uses 5x2sk
1058 and 5x2sj162 in the current gain stage.
My questions are:
1) Is there a need to change the mosfets with new ones?
2) If i do so,do i have to match them ?
I read in another thread that there isnt any need to match the specific mosfets .Is it true?
regards
Nikos😕
>1) Is there a need to change the mosfets with new ones?
never change a working device,
but clean your relais-contacts sometimes.
best regards
Kay
never change a working device,
but clean your relais-contacts sometimes.
best regards
Kay
You risk damaging the amplifier by changing the output device. Also pc board generally don't like to be soldered more than once(depend on board but still..)
I agree with KayKay said:
Transistors work without problem "forever"
Almost Only if they break down they need change.
There is also no new really replacements
for SK1058 and SJ162
that would give you substancial better performance.
You could instead get into the problem
of having to alter other things in the circuit
to fit the new transistors.
Clean it up, and change old Electrolytic Capacitor
and Carbon Potentionmeters and bad wornout Switches.
Those are component that gets bad with age.
Transistors live forever, if not shortcut.
/halo - does not live forever - at least on this planet 😉
nikosgr said:Hello all,
i have my amp for 8 years ,its a class A 150w which uses 5x2sk
1058 and 5x2sj162 in the current gain stage.
My questions are:
1) Is there a need to change the mosfets with new ones?
2) If i do so,do i have to match them ?
I read in another thread that there isnt any need to match the specific mosfets .Is it true?
regards
Nikos😕
Hi,
Unlike some components (like valves, for example) mosfets will not (usually!) deteriorate over time, towards the end of their useful life.
They usually either work OK or they don't work at all. So, as the others have said, there should not only be no need for such a change, but there is no point in doing this, either.
Regards,🙂
nikosgr said:
8 years is no age for a amp in home environment. 15-20 years may be "old".
As a conclusion: If it ain't broken, don't fix it.
Your question about matching (output transistors), it can be wise to do some matching only if you have more than one pair of transistors.
8 years for a class-A amplifier (or AB I guess). Don’t you think that there is a need to re-adjust the DC offset, Halojoy?
Nikos, the Hitachi are in complementary pairs, what’s more, in your amp they are configured in parallel. Of course they need to be matched.
Nikos, the Hitachi are in complementary pairs, what’s more, in your amp they are configured in parallel. Of course they need to be matched.
OK guys thanks for the replies
please let me give you some more details about my amp and what improvements i have done since i bought it.
concerning psu
2x200va toroids (2x40000mf) and separate regulated psu for each channel on the voltage gain stage.
2x1500va toroids (2x100000mf) separate unregulated psu for each channel on the current gain stage.
amplified circuits
all electrolytic/poly. capasitors and resistors have been changed and been upgrated(starget/solen/holco)
wiring inside the amp
i used van der hull
Anyway what i really want to do to the current gain stage which uses three pairs, is to add an extra two pair of mosfets .
The amp comes in two versions:1)150w and 200w same schematic.
The positions of the extra mosfets to be used are on the pcb.
Surely i will have to increase the size of the heatsinks and the size of the box, to get the extra 50watts
Thats why i was asking you about matching.............
Concerning the dc offset is measured from time to time (80mv on each mosfet) 1mv dc at output on one channel and 4mv dc on the other .
Thats why i asked you whether its needed to change the old mosfets with new ones.......because what i really want to do is to
add two more pairs of mosfets to the existing three old ones.
Waiting anxiously for your suggestions
regards
Nikos
please let me give you some more details about my amp and what improvements i have done since i bought it.
concerning psu
2x200va toroids (2x40000mf) and separate regulated psu for each channel on the voltage gain stage.
2x1500va toroids (2x100000mf) separate unregulated psu for each channel on the current gain stage.
amplified circuits
all electrolytic/poly. capasitors and resistors have been changed and been upgrated(starget/solen/holco)
wiring inside the amp
i used van der hull
Anyway what i really want to do to the current gain stage which uses three pairs, is to add an extra two pair of mosfets .
The amp comes in two versions:1)150w and 200w same schematic.
The positions of the extra mosfets to be used are on the pcb.
Surely i will have to increase the size of the heatsinks and the size of the box, to get the extra 50watts
Thats why i was asking you about matching.............
Concerning the dc offset is measured from time to time (80mv on each mosfet) 1mv dc at output on one channel and 4mv dc on the other .
Thats why i asked you whether its needed to change the old mosfets with new ones.......because what i really want to do is to
add two more pairs of mosfets to the existing three old ones.
Waiting anxiously for your suggestions
regards
Nikos
How much voltage do you have over the output transistors?
Normally one pair is the best and you add extra ONLY because the single pair can't handle the power. With 3 pairs I gather that you can take out 400-600 W at 4 ohms in home environment. When you add power think how many decibells it will result in! 1200 W is only 3 dB higher volume!
Normally one pair is the best and you add extra ONLY because the single pair can't handle the power. With 3 pairs I gather that you can take out 400-600 W at 4 ohms in home environment. When you add power think how many decibells it will result in! 1200 W is only 3 dB higher volume!
Hi,nikosgr said:
This is, of course, a quite different query from the first one, because this is now about adding to, and not replacing, mosfets.
As you do not seem to be very experienced in this matter, I would be very cautious about adding mosfets in the way you suggested, as several problems can possibly ensue, including dissipating the extra heat which you have mentioned.
Mosfets in parallel, as is the case here, need to share the current equally between them, and this is not always very straightforward to achieve. Also, the quiescent current will need to be changed in accordance with the requirements of the added transistors, and any added devices will most likely need their own gate stopper resistors.
Also, any stabilising circuitry or frequency compensation, may be in need of revision, too.
There are likely to be more severe 'mismatches' here than usual (i.e. If you buy all devices at the same time) as you already have some existing devices which you wish to add to.
In this case it might be easier to start again and replace all the mosfets to get a better initial match, but there is no guarantee of this, and is it worth all this extra cost and trouble?
In *any* multiple mosfet situation like this, you do need to ideally match the devices and/or use some low value source resistors to ensure this eqaul current sharing, otherwise at high outputs (and this, of course, is the reason for increasing the number of mosfets in the first place) unequal current sharing between the mosfets will quite likely cause some of them to fail.
I don't wish to discourage you in trying this, but do be aware that it is not as simple as connecting them up and away you go.
I would strongly reccommend that you study some of the circuits and writings of designers who have done this job many times before, like Nelson Pass, John Linsley Hood, and Erno Borbely etc., before you get too heavily into this modification.
I hope this helps. 🙂
Regards,
double trafo and power = +3dB
Double the trafo and add heatsinks COSTS money.
Put the same money on a Driver/woofer with +3dB SPL
and I think you have easier problems to deal with.
add 8 Watt to a 8 watt amplifier,
is the same as adding 600Watt to an 600 Watt amplifier.
The cost is somewhat different!
/halo - wants to get the most out of spent money (minimum waste) 😉
peranders points out something interesting:peranders said:
Double the trafo and add heatsinks COSTS money.
Put the same money on a Driver/woofer with +3dB SPL
and I think you have easier problems to deal with.
add 8 Watt to a 8 watt amplifier,
is the same as adding 600Watt to an 600 Watt amplifier.
The cost is somewhat different!
/halo - wants to get the most out of spent money (minimum waste) 😉
Bob is totally right but you forget to point out the hardest part if you have minor experience: Oscillations!
This is the main and the huge problem! Adding transistor pair in a amp is very hard because you must make in compact and keep all wires and traces short and descent equal for all transistors. My statement is true is you want speed from the output stage. You can slow it down but then you also make the amp worse. (= no meaning with the upgrade)
This is the main and the huge problem! Adding transistor pair in a amp is very hard because you must make in compact and keep all wires and traces short and descent equal for all transistors. My statement is true is you want speed from the output stage. You can slow it down but then you also make the amp worse. (= no meaning with the upgrade)
Hi,
Thanks Peranders, for remedying my oversight here.🙂
Whilst I thought I had adequately covered the matter when I said
"Also, any stabilising circuitry or frequency compensation, may be in need of revision, too"
I did not, of course, explain the consequences which can be as Peranders said, unwanted oscillations, which (without a 'scope to check at the MHz regions) can go unnoticed.
What I am a little unsure of here, is why you think increasing the power by this amount will be of much benefit to you.
From what you tell us you have already done, it should have improved the 'sonic' performance of the amp, anyway.
Somehow, unless you have incredibly inefficient speakers, I doubt that you would even notice the difference in power between 150 and 200 Watts.
Bearing in mind that in my own quite large listening room, with speakers of about 88dB@1W @1M sensitivity (not very efficient), I rarely use over 3 or 4 Watts for most ordinary listening (and have never seen peaks exceeding say 50 Watts), so 150 Watts/ channel is already quite a lot of power for a normal domestic system.
Regards,
Thanks Peranders, for remedying my oversight here.🙂
Whilst I thought I had adequately covered the matter when I said
"Also, any stabilising circuitry or frequency compensation, may be in need of revision, too"
I did not, of course, explain the consequences which can be as Peranders said, unwanted oscillations, which (without a 'scope to check at the MHz regions) can go unnoticed.
What I am a little unsure of here, is why you think increasing the power by this amount will be of much benefit to you.
From what you tell us you have already done, it should have improved the 'sonic' performance of the amp, anyway.
Somehow, unless you have incredibly inefficient speakers, I doubt that you would even notice the difference in power between 150 and 200 Watts.
Bearing in mind that in my own quite large listening room, with speakers of about 88dB@1W @1M sensitivity (not very efficient), I rarely use over 3 or 4 Watts for most ordinary listening (and have never seen peaks exceeding say 50 Watts), so 150 Watts/ channel is already quite a lot of power for a normal domestic system.
Regards,
Those oscillations are very often quite audiable, can sound awful, like a broken (overloaded and damaged voice coil) speaker. If you don't hear it I think your TV/Radio will!Bobken said:
ok guys thanks for the replies
the voltage over the output transistors is +/- 58v
my aim is not the extra 50 watts bat to have more safisient control(more carent output) over the speakers(300w 92db 1w/1m)
can the existing three mos. pairs deliver to the speakers (wen is needed) the carent off the psu (1500VA 100000MF/63V)
regards
nikos
the voltage over the output transistors is +/- 58v
my aim is not the extra 50 watts bat to have more safisient control(more carent output) over the speakers(300w 92db 1w/1m)
can the existing three mos. pairs deliver to the speakers (wen is needed) the carent off the psu (1500VA 100000MF/63V)
regards
nikos
58 V can give you theoretically 210 W and with 3 pairs you will loose 8-13 W only. Adding fet's will give you nothing in terms of power as you see. This is true if your driver can feed the fets with at least 14 volts over 58 V.
Max current is theoretically 58/5 = 10 A for a 8 ohms speaker. DC resistance 5-6 ohms I guess.
The fets can give you in short term 50 A or more if you don't have any special current limitation.
92 dB at 1 W => 200 W gives 115 dB, 400 W 118 dB, 800 W 121 dB.
Party level = 100-105 dB =>8-13 dB = 6.3-20 W!!
Max current is theoretically 58/5 = 10 A for a 8 ohms speaker. DC resistance 5-6 ohms I guess.
The fets can give you in short term 50 A or more if you don't have any special current limitation.
92 dB at 1 W => 200 W gives 115 dB, 400 W 118 dB, 800 W 121 dB.
Party level = 100-105 dB =>8-13 dB = 6.3-20 W!!
- Status
- Not open for further replies.