[apa]

The -18 volt side dies in the power supply for Trident 65 board I'm
using. The regulator chip appears to be fried. Looking over the parts
list in the manual it lists 2 18 volt regulators 7918CKC (18 volt
1.5A) for the negative rails and 7918CT (18 volt 3A) for the positive
side. I expected to find a 78XX series listed for the positive side
and a 79XX on the negative both with the same current rating. Each of
the rails is fused at 2 amps. Is this likely just a typo or am I
missing something critical?
Thanks, Andy

[Richard Crowley]

"apa" wrote ...
The -18 volt side dies in the power supply for Trident 65 board I'm
using. The regulator chip appears to be fried. Looking over the parts
list in the manual it lists 2 18 volt regulators 7918CKC (18 volt
1.5A) for the negative rails and 7918CT (18 volt 3A) for the positive
side. I expected to find a 78XX series listed for the positive side
and a 79XX on the negative both with the same current rating. Each of
the rails is fused at 2 amps. Is this likely just a typo or am I
missing something critical?
Thanks, Andy

Sounds like a typo to me.
Unless they're doing something screwy with the circuit. Its not
impossible, merely improbable. Do you have the schematic?
Should be easy to tell from that.

[Scott Dorsey]

apa wrote:
The -18 volt side dies in the power supply for Trident 65 board I'm
using. The regulator chip appears to be fried. Looking over the parts
list in the manual it lists 2 18 volt regulators 7918CKC (18 volt
1.5A) for the negative rails and 7918CT (18 volt 3A) for the positive
side. I expected to find a 78XX series listed for the positive side
and a 79XX on the negative both with the same current rating. Each of
the rails is fused at 2 amps. Is this likely just a typo or am I
missing something critical?

Is the center pin tied to ground? If the center pin is tied to ground
and there's no other goofy stuff around it, it's almost certainly a typo.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[gareth magennis]

"apa" wrote in message
...
The -18 volt side dies in the power supply for Trident 65 board I'm
using. The regulator chip appears to be fried. Looking over the parts
list in the manual it lists 2 18 volt regulators 7918CKC (18 volt
1.5A) for the negative rails and 7918CT (18 volt 3A) for the positive
side. I expected to find a 78XX series listed for the positive side
and a 79XX on the negative both with the same current rating. Each of
the rails is fused at 2 amps. Is this likely just a typo or am I
missing something critical?
Thanks, Andy


Well it could be that the positive rail is also supplying other circuits
that the negative rail is not. Meters etc.

If I were you I would fit the highest rated regulators you can find in that
package. Period. There will only be pennies difference in price.


Gareth.

[apa]

On Aug 25, 6:46 pm, (Scott Dorsey) wrote:
apa wrote:
The -18 volt side dies in the power supply for Trident 65 board I'm
using. The regulator chip appears to be fried. Looking over the parts
list in the manual it lists 2 18 volt regulators 7918CKC (18 volt
1.5A) for the negative rails and 7918CT (18 volt 3A) for the positive
side. I expected to find a 78XX series listed for the positive side
and a 79XX on the negative both with the same current rating. Each of
the rails is fused at 2 amps. Is this likely just a typo or am I
missing something critical?

Is the center pin tied to ground? If the center pin is tied to ground
and there's no other goofy stuff around it, it's almost certainly a typo.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Yes, they are tied to ground and everything looks straightforward. The
only reason I even looked at the manual was because I'm having trouble
tracking down an 18v regulator at 3 amps or more. Mouser and Digikey
don't seem to stock anything (or at least I can't pull anything up on
their websites). Is there somewhere else that's good to look?

[Eeyore]

apa wrote:

The -18 volt side dies in the power supply for Trident 65 board I'm
using. The regulator chip appears to be fried. Looking over the parts
list in the manual it lists 2 18 volt regulators 7918CKC (18 volt
1.5A) for the negative rails and 7918CT (18 volt 3A) for the positive
side. I expected to find a 78XX series listed for the positive side
and a 79XX on the negative both with the same current rating. Each of
the rails is fused at 2 amps. Is this likely just a typo or am I
missing something critical?

Probably a typo but look inside. Trident did some VERY odd things.

Graham

[Scott Dorsey]

apa wrote:
On Aug 25, 6:46 pm, (Scott Dorsey) wrote:
apa wrote:
The -18 volt side dies in the power supply for Trident 65 board I'm
using. The regulator chip appears to be fried. Looking over the parts
list in the manual it lists 2 18 volt regulators 7918CKC (18 volt
1.5A) for the negative rails and 7918CT (18 volt 3A) for the positive
side. I expected to find a 78XX series listed for the positive side
and a 79XX on the negative both with the same current rating. Each of
the rails is fused at 2 amps. Is this likely just a typo or am I
missing something critical?

Is the center pin tied to ground? If the center pin is tied to ground
and there's no other goofy stuff around it, it's almost certainly a typo.

Yes, they are tied to ground and everything looks straightforward. The
only reason I even looked at the manual was because I'm having trouble
tracking down an 18v regulator at 3 amps or more. Mouser and Digikey
don't seem to stock anything (or at least I can't pull anything up on
their websites). Is there somewhere else that's good to look?

There is no such thing, not in a TO-220 package anyway. Even calling it
a two-amp package is kind of optimistic. You used to be able to get 7805s
in TO-3 packages that could put out a good five amps... the chip inside
was the same, but the improved package dissipation changed everything.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Eeyore]

apa wrote:

On Aug 25, 6:46 pm, (Scott Dorsey) wrote:
apa wrote:
The -18 volt side dies in the power supply for Trident 65 board I'm
using. The regulator chip appears to be fried. Looking over the parts
list in the manual it lists 2 18 volt regulators 7918CKC (18 volt
1.5A) for the negative rails and 7918CT (18 volt 3A) for the positive
side. I expected to find a 78XX series listed for the positive side
and a 79XX on the negative both with the same current rating. Each of
the rails is fused at 2 amps. Is this likely just a typo or am I
missing something critical?

Is the center pin tied to ground? If the center pin is tied to ground
and there's no other goofy stuff around it, it's almost certainly a typo.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Yes, they are tied to ground and everything looks straightforward. The
only reason I even looked at the manual was because I'm having trouble
tracking down an 18v regulator at 3 amps or more. Mouser and Digikey
don't seem to stock anything (or at least I can't pull anything up on
their websites). Is there somewhere else that's good to look?

Look on the semi manufacturers' sites.

Plus take the part number and stuff it into Google.

And mount it better thermally next time e.g. AlO2 washer in place of mica. And
'silver' thermapath instead of aluminium as they supply with CPUs these days if
you can find a little tube.

Graham

[apa]

On Aug 25, 11:44 pm, "Tim Perry"
wrote:
Yes, they are tied to ground and everything looks straightforward. The
only reason I even looked at the manual was because I'm having trouble
tracking down an 18v regulator at 3 amps or more. Mouser and Digikey
don't seem to stock anything (or at least I can't pull anything up on
their websites). Is there somewhere else that's good to look?

I don't know if the following stock this part in particular but you can look
or ask.

Newark electronics

Allied electronics

Jameco electronics

Parts Express

Mendelson (a surplus sales place) Ohio

Fair radio sales (surplus) Ohio

Kiesub electronic supply - Las Vegas

data sheethttp://datasheet.digchip.com/477/477-73425-0-UA7918CKC.pdf

I'm sad to discover that Purchase Radio is no morehttp://jim.rees.org/aa/purchase.html

here is one that has a uA7912 for 45 centshttp://www.goldmine-elec-products.com/products.asp?dept=1170&pagenumb...
If you know how to bootstrap with a 6V zener or something your in business.

Likely you can get this part from Crown at inflated prices.

The closest I found was a 2.1 amp one at Newark which appears to be a
discontinued part.

I assume bootstrapping with a zener would just be inserting the zener
between the reg chip and ground - any more to it than that?

[Mark]

On Aug 26, 7:47*am, apa wrote:
On Aug 25, 11:44 pm, "Tim Perry"
wrote:





Yes, they are tied to ground and everything looks straightforward. The
only reason I even looked at the manual was because I'm having trouble
tracking down an 18v regulator at 3 amps or more. Mouser and Digikey
don't seem to stock anything (or at least I can't pull anything up on
their websites). Is there somewhere else that's good to look?

I don't know if the following stock this part in particular but you can look
or ask.

Newark electronics

Allied electronics

Jameco electronics

Parts Express

Mendelson * (a surplus sales place) Ohio

Fair radio sales (surplus) Ohio

Kiesub electronic supply - Las Vegas

data sheethttp://datasheet.digchip.com/477/477-73425-0-UA7918CKC.pdf

I'm sad to discover that Purchase Radio is no morehttp://jim.rees.org/aa/purchase.html

here is one that has a uA7912 for 45 centshttp://www.goldmine-elec-products.com/products.asp?dept=1170&pagenumb...
If you know how to bootstrap with a 6V zener or something *your in business.

Likely you can get this part from Crown at inflated prices.

The closest I found was a 2.1 amp one at Newark which appears to be a
discontinued part.

I assume bootstrapping with a zener would just be inserting the zener
between the reg chip and ground - any more to it than that?- Hide quoted text -

- Show quoted text -

the difficulty with bootstraping is that you must also insulate the
case from ground which makes the thermal resistance worse...

You should be able to find an 18 Volt regulator... if not get one of
the adjustable regulators at the required current and connect the
appropriate resistors to set it for 18 Volts... unfortunatly with
these you will also need to insulate the case...

Mark

[Scott Dorsey]

apa wrote:

The closest I found was a 2.1 amp one at Newark which appears to be a
discontinued part.

I suspect that is an optimistic rating too. Go to the actual manufacturer's
website and see what the maker of the chip rates the one that you have in
there at. I bet it's not 3 amps.

I assume bootstrapping with a zener would just be inserting the zener
between the reg chip and ground - any more to it than that?

Add a capacitor across the zener to reduce the noise on it.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Richard Crowley]

"Mark" wrote ...
the difficulty with bootstraping is that you must also insulate the
case from ground which makes the thermal resistance worse...

But you can use a package with good thermals (like TO-3)
and mount it to a good heatsink without (electrical or thermal)
insulation, then just isolate the heatsink.

[Paul Stamler]

The LT1033 is an adjustable 3A negative regulator. You'll have to add the
bias and adjustment resistors, and a bypass cap for the adjustment resistor,
plus adapt to a different pinout. A small satellite board would probably be
the easiest way to do all this -- and would let you use a bigger heatsink to
boot.

Peace,
Paul

[Eeyore]

Scott Dorsey wrote:

There is no such thing, not in a TO-220 package anyway. Even calling it
a two-amp package is kind of optimistic. You used to be able to get 7805s
in TO-3 packages that could put out a good five amps... the chip inside
was the same, but the improved package dissipation changed everything.

TO-220 78/79xx's max out @ 1.5A these days. It used to be only 1 A.

However, as I mentioned before, nothing Trident did would surprise me.

Graham

[Eeyore]

Scott Dorsey wrote:

apa wrote:

The closest I found was a 2.1 amp one at Newark which appears to be a
discontinued part.

I suspect that is an optimistic rating too. Go to the actual manufacturer's
website and see what the maker of the chip rates the one that you have in
there at. I bet it's not 3 amps.

I assume bootstrapping with a zener would just be inserting the zener
between the reg chip and ground - any more to it than that?

Add a capacitor across the zener to reduce the noise on it.

Are we talking about the problem of not getting the right voltage regulator here
?

I can't see that being a problem.

Here's a cute method that works. You 'bypass' the regulator in and out terminals
with a fixed value power resistor that supplies say 1A of current with typical
Vin-Vout values. The regulator then only has to regulate the remainder.

Works a treat. Damn I'm cheap !

Graham

[Scott Dorsey]

Eeyore wrote:
Scott Dorsey wrote:

There is no such thing, not in a TO-220 package anyway. Even calling it
a two-amp package is kind of optimistic. You used to be able to get 7805s
in TO-3 packages that could put out a good five amps... the chip inside
was the same, but the improved package dissipation changed everything.

TO-220 78/79xx's max out @ 1.5A these days. It used to be only 1 A.

However, as I mentioned before, nothing Trident did would surprise me.

Well, the actual current you can run through 'em has a lot to do with
the input voltage... if you're putting 20V into it, you're going to be
able to get a lot more current through than if you are putting 34V
because there is so much less excess power to dissipate.

But I wouldn't recommend using 'em for a new design that way.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[apa]

On Aug 26, 4:14 pm, Eeyore
wrote:
Scott Dorsey wrote:
apa wrote:

The closest I found was a 2.1 amp one at Newark which appears to be a
discontinued part.

I suspect that is an optimistic rating too. Go to the actual manufacturer's
website and see what the maker of the chip rates the one that you have in
there at. I bet it's not 3 amps.

I assume bootstrapping with a zener would just be inserting the zener
between the reg chip and ground - any more to it than that?

Add a capacitor across the zener to reduce the noise on it.

Are we talking about the problem of not getting the right voltage regulator here
?

I can't see that being a problem.

Here's a cute method that works. You 'bypass' the regulator in and out terminals
with a fixed value power resistor that supplies say 1A of current with typical
Vin-Vout values. The regulator then only has to regulate the remainder.

Works a treat. Damn I'm cheap !

Graham

I can find 18 volt ones - nothing that can handle the current though.

Thanks for the tip. I may do that to repair this one. When I opened it
up I found the circuit board had been well burned and repaired with a
jumper wires sometime before it's tenure with me. So it's not worth a
lot of effort or expense in the repair.

-Andy

[Richard Crowley]

"Eeyore" wrote ...
Here's a cute method that works. You 'bypass' the regulator in and out
terminals
with a fixed value power resistor that supplies say 1A of current with
typical
Vin-Vout values. The regulator then only has to regulate the remainder.

Brilliant! :-)
(As long as you have some minimum load, I suppose.)

[Mike Rivers]

apa wrote:

I can find 18 volt ones - nothing that can handle the current though.

Why not contact John Oram at Trident and ask him to send you the correct
replacement part?

--
If you e-mail me and it bounces, use your secret decoder ring and reach
me he
double-m-eleven-double-zero at yahoo -- I'm really Mike Rivers
)

[Paul Stamler]

"Eeyore" wrote in message
...


Scott Dorsey wrote:

There is no such thing, not in a TO-220 package anyway. Even calling it
a two-amp package is kind of optimistic. You used to be able to get
7805s
in TO-3 packages that could put out a good five amps... the chip inside
was the same, but the improved package dissipation changed everything.

TO-220 78/79xx's max out @ 1.5A these days. It used to be only 1 A.

The chips haven't changed; they're just interpreting the specs differently.
If you look at the max-current graph in the specs (even old specs) there
were plenty of circumstances in which you could get 1A out of them -- IF you
kept the chip temperature down with good heat-sinking.

Peace,
Paul

[Mike Rivers]

Paul Stamler wrote:

The chips haven't changed; they're just interpreting the specs differently.
If you look at the max-current graph in the specs (even old specs) there
were plenty of circumstances in which you could get 1A out of them -- IF you
kept the chip temperature down with good heat-sinking.

There was an article in QST magazine back in the late 1950s when I was
active in ham radio that described a 100 watt transmitter built with a
pair of 6L6s turned upside down, submerged in a bucket of oil. I
wouldn't recommend that approach for a console power supply though. Too
messy!


--
If you e-mail me and it bounces, use your secret decoder ring and reach
me he
double-m-eleven-double-zero at yahoo -- I'm really Mike Rivers
)

[Scott Dorsey]

Mike Rivers wrote:
Paul Stamler wrote:

The chips haven't changed; they're just interpreting the specs differently.
If you look at the max-current graph in the specs (even old specs) there
were plenty of circumstances in which you could get 1A out of them -- IF you
kept the chip temperature down with good heat-sinking.

There was an article in QST magazine back in the late 1950s when I was
active in ham radio that described a 100 watt transmitter built with a
pair of 6L6s turned upside down, submerged in a bucket of oil. I
wouldn't recommend that approach for a console power supply though. Too
messy!

It was actually pretty common. My first transmitter used a single oil-cooled
6L6. I could get 50W on 80M easily, maybe 20W on 40M... by the time it got up
to 10M I was getting about 2W out and the plates were glowing bright red.
Needless to say the Miller capacitance of the 6L6 is an issue when using it
at RF applications, which is why the 807 came out. But the 807 was expensive
and replacement 6L6s were $1.25 at the corner drugstore.

I think I got the design from one of the Tab project books, with all the
typos in them...
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Eeyore]

Scott Dorsey wrote:

Eeyore wrote:
Scott Dorsey wrote:

There is no such thing, not in a TO-220 package anyway. Even calling it
a two-amp package is kind of optimistic. You used to be able to get 7805s
in TO-3 packages that could put out a good five amps... the chip inside
was the same, but the improved package dissipation changed everything.

TO-220 78/79xx's max out @ 1.5A these days. It used to be only 1 A.

However, as I mentioned before, nothing Trident did would surprise me.

Well, the actual current you can run through 'em has a lot to do with
the input voltage... if you're putting 20V into it, you're going to be
able to get a lot more current through than if you are putting 34V
because there is so much less excess power to dissipate.

Sure, that's the dissipation limitation but I'm sure they current lint @ around
just over 1.5A typically these days.

It's not a hybrid regulator plus transistor design is it ?

Graham

[Eeyore]

apa wrote:

When I opened it
up I found the circuit board had been well burned and repaired with a
jumper wires sometime before it's tenure with me. So it's not worth a
lot of effort or expense in the repair.

Typical Trident by the sounds of it. I heard once that all Trident mix bus resistors
(and as many other resistors of practically near enough value) were 22k because if
they bought them by the bucketload they got them cheaper. Sounds like a demented
purchasing dept to me or maybe it was Toft's influence himself ?

Graham

[Eeyore]

Mike Rivers wrote:

apa wrote:

I can find 18 volt ones - nothing that can handle the current though.

Why not contact John Oram at Trident and ask him to send you the correct
replacement part?

Isn't he supposed to be some mega pompous self-promoting ass ? I doubt he'd
even know what the part is.

Graham

[Eeyore]

Richard Crowley wrote:

"Eeyore" wrote ...
Here's a cute method that works. You 'bypass' the regulator in and out
terminals with a fixed value power resistor that supplies say 1A of
current with
typical Vin-Vout values. The regulator then only has to regulate the
remainder.

Brilliant! :-)
(As long as you have some minimum load, I suppose.)

ICs are pretty good at that ! Yes, it really is a little winner and helps
keep costs down.

Another little trick I've used is to turn voltage regularors into constant
current sources/sinks, so that digital loads with massive switching noise
don't radiate noise down the power leads. The difference is astonishing.

Graham

[Scott Dorsey]

Eeyore wrote:
Kludge wrote:
Well, the actual current you can run through 'em has a lot to do with
the input voltage... if you're putting 20V into it, you're going to be
able to get a lot more current through than if you are putting 34V
because there is so much less excess power to dissipate.

Sure, that's the dissipation limitation but I'm sure they current lint @ around
just over 1.5A typically these days.

There's no real limiting... you can put any load you want on it and it'll
keep putting out current until it burns up. If you throw a load on fast
enough, you can even make them explode.

It's not a hybrid regulator plus transistor design is it ?

It's a big transistor, a little transistor, and a zener diode, all
fabricated on one substrate. That's it, nothing fancy.
--scott


--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Mike Rivers]

Eeyore wrote:

Why not contact John Oram

Isn't he supposed to be some mega pompous self-promoting ass ?

Yeah, but he's actually a pretty nice guy when you get to know him, and
you aren't fighting him for the name of your product.

I doubt he'd even know what the part is.

He certainly knows what a voltage regulator is. He's an electronics
engineer. If the original poster contacted John (or just the general
support e-mail address) he'd get some help obtaining the right part.


--
If you e-mail me and it bounces, use your secret decoder ring and reach
me he
double-m-eleven-double-zero at yahoo -- I'm really Mike Rivers
)