Cyrus 1 output transistor question. push-pull?

[Mooly]

Lol

I find solder braid the most useful for desoldering parts... I've grown up using it and just got used to it I suppose. You should be able to cleanly desolder a part without causing any damage whatsoever but it takes practice. A hot iron with a bit like a poker can often be best as it holds more heat.

This shows the idea but I would say the guy in this video is using far to thin a tip on the iron and he is taking far to long holding the hot iron on the component itself. It should take typically 1 to 2 seconds per joint.

YouTube

 

[randombeliefs]

Lol

I find solder braid the most useful for desoldering parts... I've grown up using it and just got used to it I suppose. You should be able to cleanly desolder a part without causing any damage whatsoever but it takes practice. A hot iron with a bit like a poker can often be best as it holds more heat.

This shows the idea but I would say the guy in this video is using far to thin a tip on the iron and he is taking far to long holding the hot iron on the component itself. It should take typically 1 to 2 seconds per joint.

YouTube

A neighbour lent me his solder sucker when i did the cyrus 2 cos getting the alps selector switch off the board was tricky. I couldn't get the hang of it tbh as i seemed to need to remove the iron from the joint to apply the sucker by which time the solder was no longer suckable ! so i tried copper braid and this worked a treat, the solder just wicked away.

neighbour has a bulb limiter too so I'll bob round see if i can borrow that til i can find an incandescent bulb in the uk ...

 

[Mooly]

I've suckers in the past but prefer braid tbh. You have to keep the solder molten at all times with the iron while applying the Teflon nozzle of the sucker.

 

[randombeliefs]

driver possibilities ?

http://uk.farnell.com/on-semiconductor/bd13910stu/transistor-bipol-npn-80v-to-126/dp/2453822

and

http://uk.farnell.com/on-semiconductor/bd14010stu/transistor-bipol-pnp-80v-to-225aa/dp/2453824

? what do you think ?

I have heard of the MJE3055T being used for replacement output transistors on the Cyrus. Cheap and readily available (CPC). BD139/BD140 for the drivers.

0.22 ohm/3watt resistors,
http://cpc.farnell.com/multicomp/mcknp03sj022ka19/resistor-wirewound-0r22-5-3ws/dp/RE07407

You need to check the value of all the resistors marked with an exclamation mark in the faulty channel. These are low value resistors and can be checked in circuit initially. If you don't see the reading expected then try reversing the meter leads. If it still looks wrong then lift one end and check. The shorted outputs/drivers may skew the result of some of these so remove the faulty transistors first.

The Cyrus appears to have non adjustable bias and this will be a possible issue using different non original semiconductors.

You may have to change the value of R63, a 680 ohm (hard to make the ref number out), perhaps even replacing it with a preset to give continuously variable adjustment.

Also you must use a bulb tester when working on this as it will save damage should anything go amiss.

 

[randombeliefs]

and for the outputs

http://uk.farnell.com/stmicroelectronics/mje3055t/transistor-npn-to-220/dp/9802460

 

[Mooly]

Those look fine.

Also worth getting 5 or 10 each of these in case any of the small signal devices need replacing. These are a good fit for most such uses and cost peanuts.

http://uk.farnell.com/on-semiconductor/bc556bta/transistor-bipol-pnp-65v-to-92/dp/2453801

http://uk.farnell.com/on-semiconductor/bc546bta/transistor-bipol-npn-65v-to-92/dp/2453320

 

[randombeliefs]

thanks !

just ordering with cpc now

just to confuse me there seems to be cpc , element14 and farnell, and they're all the same company with almost identical websites .... went with the darker shade of blue cpc cos with next day delivery it was pretty much same price as element14 with standard.

only thing is, I can't register with cpc because I registered with element14, cpc say the username is gone (can only be me !)

 

[randombeliefs]

I'll get a good handful of those then

 

[Mooly]

Hmmm... I'm registered with both but with slightly different user names. Just add an initial or something. I never gave it any thought at the time and there were years apart between the two registrations.

The small transistors come in useful should any be needed.

I was also looking at other alternatives to the MJE3055T which is the one I saw as being used for replacements in the Cyrus. Although the latest and greatest fastest parts might seem best, that doesn't always follow.

I actually wondered about the old TIP41C which although only 6 amp rated (but 10A peak) could be a suitable candidate. It has the benefit of a 100 volt breakdown voltage. In any case the Cyrus is only rated at 25 watts into 8 ohms.

 

[AndrewT]

7.5Apk for a transient into an 8ohms load is probably enough for a severe reactance speaker driven with a 25W amplifier.

That is comfortably below the 10Apk rating provided the amplifier is never allowed to run hot.

 

[Mooly]

That's kind of what I was thinking. The Cyrus runs on -/+30 volts which seems to be marginal (for secondary breakdown at high output swing) when using 60 volt parts although in practice you would probably never drive it that hard in a domestic setting.

 

[randombeliefs]

That's kind of what I was thinking. The Cyrus runs on -/+30 volts which seems to be marginal (for secondary breakdown at high output swing) when using 60 volt parts although in practice you would probably never drive it that hard in a domestic setting.

so you think something that can take 100v then ? I need to order some resistors and whatnot yet so I'll add the outputs you suggest. The dual resistors i couldn't source so I'll just get individual 0.22k and put in series with the common pins where the central pin of the old duals are now

 

[randombeliefs]

sorry, this is such a basic question.

any tips on choosing suitable replacement resistors for the ones on the circuit mark with an [!] ?

I thought I may as well get those ordered in case they're needed I can't imagine they're expensive. Much of a muchness or are there specs I should look for ?

I had thought of changing out the rectifier diodes too, but maybe I'm getting carried away ...

 

[randombeliefs]

failing that I'll just try to match what the parts list has to say about them

 

[Mooly]

They are 0.22 ohm, not 0.22k (which would be 220 ohm).

A couple of possibilities:
http://uk.farnell.com/welwyn/w31-r22-ji/resistor-3w-5-0r22/dp/9497242


http://uk.farnell.com/multicomp/mcknp03sj022ka19/resistor-wirewound-0r22-5-3ws/dp/1903854

The ones marked with an exclamation mark are officially 'safety components' and are actually fusible meaning they fail open circuit (and most importantly for customers) they don't smell or smoke if they are overloaded to the point of failure. They tend to be hard to find and expensive, however........

the first thing to do is simply measure the ones in the amp. They have to be checked simply because it is low value resistors that fail under certain fault conditions. So we have to know. If they read as being correct value then there is absolutely zero point in replacing them.

So lets measure them first.

If any do need replacing then you can use a standard part of 0.25 watt rating to initially fix the amp. The amplifier is all low voltage circuitry and so assuming the amp has a metal top and so on then there is no real safety risk. If it was made of cardboard or some flammable material then it would be a different story.

For the outputs... it would not be unusual to see parts used close to their ratings in commercial equipment, its just a fact of life. In normal use they will not see 60 volts across them, just 30 volts under idle conditions and 30 'plus' whatever the output voltage delivered to the speaker is under use. Given that most listening is done under a couple of watts output it is not as big an issue as might be first thought. If we were designing an amp we might use different parts today but for fixing an old amp we have to look at similar type devices as were first used. Given that the TIP41C costs buttons it is worth getting a pair.

Leave the rectifiers for now. Concentrate of fixing the original fault first.

 

[randombeliefs]

as for the 0.22k business, it was cos of the markings on the original white ceramic dual resistors, sure enough the parts list says R22 , thanks !

 

[Mooly]

The K is actually the tolerance on these (confusing isn't it ). K = -/+10% and J = -/+5%

 

[randombeliefs]

learning learning learning

 

[AndrewT]

so you think something that can take 100v then ? I need to order some resistors and whatnot yet so I'll add the outputs you suggest. The dual resistors i couldn't source so I'll just get individual 0.22k and put in series with the common pins where the central pin of the old duals are now

Not 0.22k.
the source resistors are 0r22
The two in series should measure 0r44

If you pass 100mAdc through that you will see a voltage drop of 44mVdc. Check the one that is not damaged to see what drop you get across each half.

You can use this method to check your source resistors and match them rather than just accept the standard 5%, or 10% tolerance.
Alternatively you can use 5 off 1r0 1% 600mW in parallel to get 0r2 3W
or 4off 1r0 in parallel with 1r8 to get almost exactly 0r22, but only 2.5W

 

[randombeliefs]

Not 0.22k.
the source resistors are 0r22
The two in series should measure 0r44

If you pass 100mAdc through that you will see a voltage drop of 44mVdc. Check the one that is not damaged to see what drop you get across each half.

You can use this method to check your source resistors and match them rather than just accept the standard 5%, or 10% tolerance.
Alternatively you can use 5 off 1r0 1% 600mW in parallel to get 0r2 3W
or 4off 1r0 in parallel with 1r8 to get almost exactly 0r22, but only 2.5W

I'm on that now, thanks !