Forums

strange use of 78** regulators or not ?

Started by M Philbrook November 2, 2018

I do some repair work for my friends now and then and have a Furman Ar-
1220 AC volt line regulator.

 Original issue was the Triac and opto being shorted for the main 
primary side so it wouldn't shut the outputs down on extrame voltages.
 
 Anyways, while poking around and generating a reversed schematic for it 
I found it strange that it has a small low current control transformer 
supplying the volt sensing circuit which also supplies the 7805 reg, 
this same supply also supports a 7815 and 7915 to supplies a TLO84 chip 
to be used as a comparator for over volts from what I can see..

  What is strange is there is only 10-12DC at the bridge and yes the 
caps are ok. this isn't enough to property operate the the regs. 

 After looking up the part number of the transformer it appears that it 
has a 230V primary but the unit is slated for 120V service? So since 
this transformer looks like it was made to order I thought maybe someone
put the wrong one in and its been runing this way? I took a temp 
transformer that would produce the required output for the 7815 and 7915 
post regs. 

 The strange part of this was that, although the regs were then ok, the 
unit went into a EVS mode (extreme shut down mode) which it does for a 
short time when it normally powers up, but this time never came out of 
it.
 
 It looks like it was designed to actually operate below the post regs 
ratings in normal input voltages and using the regs as a limit not a 
fixed output.
 
  Is this propery use of these regs ?
On 2018-11-02 07:42, M Philbrook wrote:
> > > I do some repair work for my friends now and then and have a Furman Ar- > 1220 AC volt line regulator. > > Original issue was the Triac and opto being shorted for the main > primary side so it wouldn't shut the outputs down on extrame voltages. > > Anyways, while poking around and generating a reversed schematic for it > I found it strange that it has a small low current control transformer > supplying the volt sensing circuit which also supplies the 7805 reg, > this same supply also supports a 7815 and 7915 to supplies a TLO84 chip > to be used as a comparator for over volts from what I can see.. > > What is strange is there is only 10-12DC at the bridge and yes the > caps are ok. this isn't enough to property operate the the regs. > > After looking up the part number of the transformer it appears that it > has a 230V primary but the unit is slated for 120V service? So since > this transformer looks like it was made to order I thought maybe someone > put the wrong one in and its been runing this way? I took a temp > transformer that would produce the required output for the 7815 and 7915 > post regs. >
Maybe the design didn't work right and someone found that by using a 230V transformer it kind of worked?
> The strange part of this was that, although the regs were then ok, the > unit went into a EVS mode (extreme shut down mode) which it does for a > short time when it normally powers up, but this time never came out of > it. > > It looks like it was designed to actually operate below the post regs > ratings in normal input voltages and using the regs as a limit not a > fixed output. > > Is this propery use of these regs ? >
You can but that's IMHO a pretty hokey design. The drop-out voltage isn't well defined and a lot of 120Hz ripple would be passed onto the rails. The TL084 is designed for operation on up to +/-15V supplies. Using it as a comparator may not have been a very wise decision. Then there is the matter of potential phase reversal. https://www.analog.com/media/en/technical-documentation/data-sheets/10578fd.pdf Quote page 9 "Most industry standard JFET input single, dual and quad op amps (e.g., LF156, LF351, LF353, LF411, LF412, OP-15, OP-16, OP-215, TL084) exhibit phase reversal at the output when the negative common mode limit at the input is exceeded". It might be worth trying stable +/-15V supplies and a real comparator in there, unless the TL084 was an overvoltage limiter and not a trip-off circuit, in which case the use of an opamp would be required. If unsure sketch up the schematic and post it here. A photo of a reasonable hand-sketch is fine. -- Regards, Joerg http://www.analogconsultants.com/
Joerg <news@analogconsultants.com> wrote:

> The TL084 is designed for operation on up to +/-15V supplies. Using it > as a comparator may not have been a very wise decision. Then there is > the matter of potential phase reversal.
> https://www.analog.com/media/en/technical-documentation/data-sheets/10578 > fd.pdf > > Quote page 9 "Most industry standard JFET input single, dual and quad op > amps (e.g., LF156, LF351, LF353, LF411, LF412, > OP-15, OP-16, OP-215, TL084) exhibit phase reversal at > the output when the negative common mode limit at the > input is exceeded".
Amazing catch. Thanks. it looks like the LT1057/LT1058 goes into oscillation at the negative peak. I'm not so sure that is a good idea.
On Friday, November 2, 2018 at 9:39:03 AM UTC-4, M Philbrook wrote:
> I do some repair work for my friends now and then and have a Furman Ar- > 1220 AC volt line regulator. > > Original issue was the Triac and opto being shorted for the main > primary side so it wouldn't shut the outputs down on extrame voltages. > > Anyways, while poking around and generating a reversed schematic for it > I found it strange that it has a small low current control transformer > supplying the volt sensing circuit which also supplies the 7805 reg, > this same supply also supports a 7815 and 7915 to supplies a TLO84 chip > to be used as a comparator for over volts from what I can see.. > > What is strange is there is only 10-12DC at the bridge and yes the > caps are ok. this isn't enough to property operate the the regs. > > After looking up the part number of the transformer it appears that it > has a 230V primary but the unit is slated for 120V service? So since > this transformer looks like it was made to order I thought maybe someone > put the wrong one in and its been runing this way? I took a temp > transformer that would produce the required output for the 7815 and 7915 > post regs.
They use a 230VAC transformer at the 120 VAC operating voltage to eliminate linearity and possibly saturation problems at line fault overvoltages. Put it back. Last time I checked, the 7800/7900 maintain their fault protection features at anything above 6.3V input. The power supply for the TL opamp does not require regulation, there is such a thing as PSRR, so no problem with the regulators not regulating so well.
> > The strange part of this was that, although the regs were then ok, the > unit went into a EVS mode (extreme shut down mode) which it does for a > short time when it normally powers up, but this time never came out of > it.
There you go.
> > It looks like it was designed to actually operate below the post regs > ratings in normal input voltages and using the regs as a limit not a > fixed output. > > Is this propery use of these regs ?
Yes.
On Friday, 2 November 2018 14:32:52 UTC, Joerg  wrote:
> On 2018-11-02 07:42, M Philbrook wrote: > > > > > > I do some repair work for my friends now and then and have a Furman Ar- > > 1220 AC volt line regulator. > > > > Original issue was the Triac and opto being shorted for the main > > primary side so it wouldn't shut the outputs down on extrame voltages. > > > > Anyways, while poking around and generating a reversed schematic for it > > I found it strange that it has a small low current control transformer > > supplying the volt sensing circuit which also supplies the 7805 reg, > > this same supply also supports a 7815 and 7915 to supplies a TLO84 chip > > to be used as a comparator for over volts from what I can see.. > > > > What is strange is there is only 10-12DC at the bridge and yes the > > caps are ok. this isn't enough to property operate the the regs. > > > > After looking up the part number of the transformer it appears that it > > has a 230V primary but the unit is slated for 120V service? So since > > this transformer looks like it was made to order I thought maybe someone > > put the wrong one in and its been runing this way? I took a temp > > transformer that would produce the required output for the 7815 and 7915 > > post regs. > > > > Maybe the design didn't work right and someone found that by using a > 230V transformer it kind of worked? > > > > The strange part of this was that, although the regs were then ok, the > > unit went into a EVS mode (extreme shut down mode) which it does for a > > short time when it normally powers up, but this time never came out of > > it. > > > > It looks like it was designed to actually operate below the post regs > > ratings in normal input voltages and using the regs as a limit not a > > fixed output. > > > > Is this propery use of these regs ? > > > > You can but that's IMHO a pretty hokey design. The drop-out voltage > isn't well defined and a lot of 120Hz ripple would be passed onto the rails.
the datasheet defines 7812 dropout voltage as 2.5V max. Opamps tend to not care much about mains derived ripple.
> The TL084 is designed for operation on up to +/-15V supplies. Using it > as a comparator may not have been a very wise decision. Then there is > the matter of potential phase reversal. > > https://www.analog.com/media/en/technical-documentation/data-sheets/10578fd.pdf > > Quote page 9 "Most industry standard JFET input single, dual and quad op > amps (e.g., LF156, LF351, LF353, LF411, LF412, > OP-15, OP-16, OP-215, TL084) exhibit phase reversal at > the output when the negative common mode limit at the > input is exceeded".
They're rated for anything upto 3 or 4v within psu rails. Many apps just won't take them into 'inversion.'
> It might be worth trying stable +/-15V supplies and a real comparator in > there, unless the TL084 was an overvoltage limiter and not a trip-off > circuit, in which case the use of an opamp would be required. If unsure > sketch up the schematic and post it here. A photo of a reasonable > hand-sketch is fine.
I don't think rejigging the circuit is a good idea until one understands what's being done already. I have encountered commercial products that were just not designed to work properly, but not often. NT
On 11/2/2018 10:30 AM, bloggs.fredbloggs.fred@gmail.com wrote:
> On Friday, November 2, 2018 at 9:39:03 AM UTC-4, M Philbrook wrote: >> I do some repair work for my friends now and then and have a Furman Ar- >> 1220 AC volt line regulator. >> >> Original issue was the Triac and opto being shorted for the main >> primary side so it wouldn't shut the outputs down on extrame voltages. >> >> Anyways, while poking around and generating a reversed schematic for it >> I found it strange that it has a small low current control transformer >> supplying the volt sensing circuit which also supplies the 7805 reg, >> this same supply also supports a 7815 and 7915 to supplies a TLO84 chip >> to be used as a comparator for over volts from what I can see.. >> >> What is strange is there is only 10-12DC at the bridge and yes the >> caps are ok. this isn't enough to property operate the the regs. >> >> After looking up the part number of the transformer it appears that it >> has a 230V primary but the unit is slated for 120V service? So since >> this transformer looks like it was made to order I thought maybe someone >> put the wrong one in and its been runing this way? I took a temp >> transformer that would produce the required output for the 7815 and 7915 >> post regs. > > They use a 230VAC transformer at the 120 VAC operating voltage to eliminate linearity and possibly saturation problems at line fault overvoltages. Put it back. Last time I checked, the 7800/7900 maintain their fault protection features at anything above 6.3V input. The power supply for the TL opamp does not require regulation, there is such a thing as PSRR, so no problem with the regulators not regulating so well.
You must mean the 7805. He posted "same supply also supports a 7815 and 7915" so I don't think they will regulate at 6.3V.
>> >> The strange part of this was that, although the regs were then ok, the >> unit went into a EVS mode (extreme shut down mode) which it does for a >> short time when it normally powers up, but this time never came out of >> it. > > There you go. > >> >> It looks like it was designed to actually operate below the post regs >> ratings in normal input voltages and using the regs as a limit not a >> fixed output. >> >> Is this propery use of these regs ? > > Yes. >
On 2018-11-02 10:53, tabbypurr@gmail.com wrote:
> On Friday, 2 November 2018 14:32:52 UTC, Joerg wrote: >> On 2018-11-02 07:42, M Philbrook wrote:
[...]
>>> The strange part of this was that, although the regs were then >>> ok, the unit went into a EVS mode (extreme shut down mode) which >>> it does for a short time when it normally powers up, but this >>> time never came out of it. >>> >>> It looks like it was designed to actually operate below the post >>> regs ratings in normal input voltages and using the regs as a >>> limit not a fixed output. >>> >>> Is this propery use of these regs ? >>> >> >> You can but that's IMHO a pretty hokey design. The drop-out >> voltage isn't well defined and a lot of 120Hz ripple would be >> passed onto the rails. > > the datasheet defines 7812 dropout voltage as 2.5V max. Opamps tend > to not care much about mains derived ripple. >
Until there is "fuzz" on that from nearby dimmers and such, that made it past the caps because of their ESR. Then they often also use the rails to bias something and that's where it really becomes ugly. If an overvoltage curb was needed the proper way to do it would be, for example, a depletion mode FET.
> >> The TL084 is designed for operation on up to +/-15V supplies. Using >> it as a comparator may not have been a very wise decision. Then >> there is the matter of potential phase reversal. >> >> https://www.analog.com/media/en/technical-documentation/data-sheets/10578fd.pdf >> >> >>
Quote page 9 "Most industry standard JFET input single, dual and quad op
>> amps (e.g., LF156, LF351, LF353, LF411, LF412, OP-15, OP-16, >> OP-215, TL084) exhibit phase reversal at the output when the >> negative common mode limit at the input is exceeded". > > They're rated for anything upto 3 or 4v within psu rails. Many apps > just won't take them into 'inversion.' > > >> It might be worth trying stable +/-15V supplies and a real >> comparator in there, unless the TL084 was an overvoltage limiter >> and not a trip-off circuit, in which case the use of an opamp would >> be required. If unsure sketch up the schematic and post it here. A >> photo of a reasonable hand-sketch is fine. > > I don't think rejigging the circuit is a good idea until one > understands what's being done already. I have encountered commercial > products that were just not designed to work properly, but not > often. >
I have seen really gross ones. The topper was a strip chart printer where the reset function consisted on shorting out the 5V rail. This was nowhere to be found in the datasheet. We planned to install them in large ultrasound machines where the 5V rail featured a sustained 100 amps ... *KABLAMMO* ... internal meltdown. That manufacturer landed on our blacklist, for gross incompetence. -- Regards, Joerg http://www.analogconsultants.com/
On 2018-11-03 15:33, Joerg wrote:
> On 2018-11-02 10:53, tabbypurr@gmail.com wrote: >> On Friday, 2 November 2018 14:32:52 UTC, Joerg wrote: >>> On 2018-11-02 07:42, M Philbrook wrote: > > [...] > > >>>> The strange part of this was that, although the regs were then >>>> ok, the unit went into a EVS mode (extreme shut down mode) >>>> which it does for a short time when it normally powers up, but >>>> this time never came out of it. >>>> >>>> It looks like it was designed to actually operate below the >>>> post regs ratings in normal input voltages and using the regs >>>> as a limit not a fixed output. >>>> >>>> Is this propery use of these regs ? >>>> >>> >>> You can but that's IMHO a pretty hokey design. The drop-out >>> voltage isn't well defined and a lot of 120Hz ripple would be >>> passed onto the rails. >> >> the datasheet defines 7812 dropout voltage as 2.5V max. Opamps tend >> to not care much about mains derived ripple. >> > > Until there is "fuzz" on that from nearby dimmers and such, that > made it past the caps because of their ESR. Then they often also use > the rails to bias something and that's where it really becomes ugly. > > If an overvoltage curb was needed the proper way to do it would be, > for example, a depletion mode FET. > >> >>> The TL084 is designed for operation on up to +/-15V supplies. >>> Using it as a comparator may not have been a very wise decision. >>> Then there is the matter of potential phase reversal. >>> >>> https://www.analog.com/media/en/technical-documentation/data-sheets/10578fd.pdf >>> > Quote page 9 "Most industry standard JFET input single, dual and > quad op >>> amps (e.g., LF156, LF351, LF353, LF411, LF412, OP-15, OP-16, >>> OP-215, TL084) exhibit phase reversal at the output when the >>> negative common mode limit at the input is exceeded". >> >> They're rated for anything upto 3 or 4v within psu rails. Many apps >> just won't take them into 'inversion.' >> >> >>> It might be worth trying stable +/-15V supplies and a real >>> comparator in there, unless the TL084 was an overvoltage limiter >>> and not a trip-off circuit, in which case the use of an opamp >>> would be required. If unsure sketch up the schematic and post it >>> here. A photo of a reasonable hand-sketch is fine. >> >> I don't think rejigging the circuit is a good idea until one >> understands what's being done already. I have encountered >> commercial products that were just not designed to work properly, >> but not often. >> > > I have seen really gross ones. The topper was a strip chart printer > where the reset function consisted on shorting out the 5V rail. This > was nowhere to be found in the datasheet. We planned to install them > in large ultrasound machines where the 5V rail featured a sustained > 100 amps ... *KABLAMMO* ... internal meltdown. > > That manufacturer landed on our blacklist, for gross incompetence.
Gross incompetence knows no boundaries. Here's what I've seen in the power circuit of a set of desktop speakers. Note how the switch sits between 2 capacitors: T ---- SW <---)|(---|~ +|---*---o/ o---*---> + )|( | | | | AC )|( | | = C1 C2 = DC )|( | | | | <---)|(---|~ -|---*----------*---> - ---- The caps C1 and C2 were on the order of 1000 uF (at 16 V or so) each, and were of identical type (pretty much maximizing the spark energy). The switch had a sliding foil-like piece of brass. It did not last. -- Dimitrij
Dimitrij Klingbeil wrote:

>> > > That manufacturer landed on our blacklist, for gross incompetence. > > > Gross incompetence knows no boundaries. Here's what I've seen in the > power circuit of a set of desktop speakers. Note how the switch sits > between 2 capacitors: > > > T ---- SW > <---)|(---|~ +|---*---o/ o---*---> + > )|( | | | | > AC )|( | | = C1 C2 = DC > )|( | | | | > <---)|(---|~ -|---*----------*---> - > ---- > > The caps C1 and C2 were on the order of 1000 uF (at 16 V or so) each, > and were of identical type (pretty much maximizing the spark energy). > > The switch had a sliding foil-like piece of brass. It did not last. >
** The energy stored in a cap is given by: 1/2 CV squared In your example, ( 1000uF & 16V ) about 0.1 Joules. A famous brand tube amp I saw had its standby toggle switch wired so that it paralleled connected a charged 100uF electro cap with an uncharged one when operated. The voltage level was over 450V DC. Equates to 10 Joules. The switch made a loud bang, every time this was done. ..... Phil
On Saturday, November 3, 2018 at 2:17:00 AM UTC-4, John S wrote:
> On 11/2/2018 10:30 AM, bloggs.fredbloggs.fred@gmail.com wrote: > > On Friday, November 2, 2018 at 9:39:03 AM UTC-4, M Philbrook wrote: > >> I do some repair work for my friends now and then and have a Furman Ar- > >> 1220 AC volt line regulator. > >> > >> Original issue was the Triac and opto being shorted for the main > >> primary side so it wouldn't shut the outputs down on extrame voltages. > >> > >> Anyways, while poking around and generating a reversed schematic for it > >> I found it strange that it has a small low current control transformer > >> supplying the volt sensing circuit which also supplies the 7805 reg, > >> this same supply also supports a 7815 and 7915 to supplies a TLO84 chip > >> to be used as a comparator for over volts from what I can see.. > >> > >> What is strange is there is only 10-12DC at the bridge and yes the > >> caps are ok. this isn't enough to property operate the the regs. > >> > >> After looking up the part number of the transformer it appears that it > >> has a 230V primary but the unit is slated for 120V service? So since > >> this transformer looks like it was made to order I thought maybe someone > >> put the wrong one in and its been runing this way? I took a temp > >> transformer that would produce the required output for the 7815 and 7915 > >> post regs. > > > > They use a 230VAC transformer at the 120 VAC operating voltage to eliminate linearity and possibly saturation problems at line fault overvoltages. Put it back. Last time I checked, the 7800/7900 maintain their fault protection features at anything above 6.3V input. The power supply for the TL opamp does not require regulation, there is such a thing as PSRR, so no problem with the regulators not regulating so well. > > You must mean the 7805. He posted "same supply also supports a 7815 and > 7915" so I don't think they will regulate at 6.3V.
"...maintain their fault protection features at anything above 6.3V input..", that means stuff like current limit and overtemperature foldback. You can't seriously think anyone would claim the 7815 maintains 15V regulation at 6.3V input???
> > >> > >> The strange part of this was that, although the regs were then ok, the > >> unit went into a EVS mode (extreme shut down mode) which it does for a > >> short time when it normally powers up, but this time never came out of > >> it. > > > > There you go. > > > >> > >> It looks like it was designed to actually operate below the post regs > >> ratings in normal input voltages and using the regs as a limit not a > >> fixed output. > >> > >> Is this propery use of these regs ? > > > > Yes. > >