Topics

1502 TDR project

Mark Pilant
 

I've had a 1502 project for a while, and I'm finally getting some time
to work on it. I've already done the cap/resistor battery "mod" as
suggested on the TekWiki, and I have 14.4 volts on the "BATT" test point
on the power supply board.

However, after that, the power supply doesn't appear to be working so
well. All the various test points do have a voltage, but they are way
off. For instance, the +25V test point measures about 3 volts.

I'm trying to wrap my head around how the "DC VOLTAGE CONVERTER" portion
of the supply is supposed to work, but I'm not having much luck. (Maybe
working with software too long :-)

I did check the voltage at Q6345 (it is the same as the "BATT" voltage,
so it would seem the primary of T6535 is OK.

Looking for some guidance.

- Mark N1VQW

Tom Gardner
 

On 09/12/19 15:18, Mark Pilant wrote:
I've had a 1502 project for a while, and I'm finally getting some time
to work on it.  I've already done the cap/resistor battery "mod" as
suggested on the TekWiki, and I have 14.4 volts on the "BATT" test point
on the power supply board.

However, after that, the power supply doesn't appear to be working so
well.  All the various test points do have a voltage, but they are way
off.  For instance, the +25V test point measures about 3 volts.

I'm trying to wrap my head around how the "DC VOLTAGE CONVERTER" portion
of the supply is supposed to work, but I'm not having much luck. (Maybe
working with software too long :-)

I did check the voltage at Q6345 (it is the same as the "BATT" voltage,
so it would seem the primary of T6535 is OK.

Looking for some guidance.

- Mark  N1VQW
That PSU is, um, subtle. I've had successes and failures with cap/resistor substitutes, and don't trust the technique. Unless the PSU considers the values are /just right/, transient voltages/currents can trip the PSU.

A reliable technique is to put a couple of 4mm banana sockets on the ends of wires attached to a bench PSU set to 12V. ISTR the transient current can be 600mA, but the operating current is much lower. Attach the banana sockets to the connector in the case, ensuring the polarity is correct. That is easier when the case has been removed.

Once you have such a good "battery", you can can ignore the battery charger part of the PSU and proceed with faultfinding the main part of the equipment.

There are fuses on the board; if they have blown then unsurprisingly you will need to find the reason.

Apart from that, faults would normally be faulty electrolytics. I've had a couple on the 25V line (c6341, c6246) spew acid and corrode PCB tracks. Don't look at the prices of direct replacements, unless you have had a stiff drink and are sitting down :) Do look at the diameter of replacements :)

Mark Pilant
 

Hi Tom.

I haven't tried supplying power directly to the banana plugs, but the
parallel cap/resistor battery substitute (ala TekWiki) seems to work
fine, as the battery meter on the front panel indicates the battery is
"good" and I do see 14.4VDC supplying the DC converter.

So it would seem the battery portion of the power supply is working and
I need to determine why the rest of the power supply isn't working.

My general sense is the DC converter is an oscillator driving the T6535
primary to then generate all the secondary voltages needed. I suspect
the oscillator isn't oscillating (well?), but I'm not sure of the best
way to investigate. (Not sure how many ways there are to shut it down;
lack of a battery is one.)

I'm a little wary just probing around with my scope (7904 with the usual
vertical / horizontal plugins), because of potential ground issues.
Although I do have a Tek 224 I can use that shouldn't have any ground
issues.

I just checked, and the power supply for the 1503 is the same as the one
for the 1502. I may be able to open up my 1503 and use its power supply
to troubleshoot the 1502 power supply.

- Mark N1VQW

Tom Gardner
 

On 09/12/19 16:38, Mark Pilant wrote:
Hi Tom.

I haven't tried supplying power directly to the banana plugs, but the
parallel cap/resistor battery substitute (ala TekWiki) seems to work
fine, as the battery meter on the front panel indicates the battery is
"good" and I do see 14.4VDC supplying the DC converter.
That's insufficient.

You realise that PSU trips if there is a transient overvoltage. At that point the voltmeter reads "excessively good" :)

I presume you have read the circuit operation section of the manual?


So it would seem the battery portion of the power supply is working and
I need to determine why the rest of the power supply isn't working.

My general sense is the DC converter is an oscillator driving the T6535
primary to then generate all the secondary voltages needed.  I suspect
the oscillator isn't oscillating (well?), but I'm not sure of the best
way to investigate.  (Not sure how many ways there are to shut it down;
lack of a battery is one.)
One way is the overvoltage detection, via q6547/8/9.


I'm a little wary just probing around with my scope (7904 with the usual
vertical / horizontal plugins), because of potential ground issues.
Although I do have a Tek 224 I can use that shouldn't have any ground
issues.
If you power off a bench supply, then clipping the scope shield to the chassis is fine.

The 1502's transformer is fully isolated, so**clipping the scope shield to the chassis is fine.


I just checked, and the power supply for the 1503 is the same as the one
for the 1502.  I may be able to open up my 1503 and use its power supply
to troubleshoot the 1502 power supply.
Yes.

Mark Pilant
 

That's insufficient.
Good to know.

You realise that PSU trips if there is a transient overvoltage. At that
point the voltmeter reads "excessively good" :)
I did not. I just checked, and when powered on, the battery meter reads
about 1/16" past/above the upper portion of the green area. I'm not sure
if this matches what you mean by "excessively good". :-)

I presume you have read the circuit operation section of the manual?
Yes, although I didn't see anything about overvoltage tripping.

- Mark N1VQW

Tom Gardner
 

On 09/12/19 17:09, Mark Pilant wrote:
That's insufficient.
Good to know.

You realise that PSU trips if there is a transient overvoltage. At that
point the voltmeter reads "excessively good" :)
I did not.  I just checked, and when powered on, the battery meter reads
about 1/16" past/above the upper portion of the green area.  I'm not sure
if this matches what you mean by "excessively good". :-)
Probably :)


I presume you have read the circuit operation section of the manual?
Yes, although I didn't see anything about overvoltage tripping.
Last paragraph of the "Battery Charger" section indicates VR6134 provides protection in case the battery pack is removed and the line cord is plugged into a line voltage source.

And my foulup: q6547/8/9 are the undervoltage protection.

Mark Pilant
 

Good to know all this information about the batter charging circuit. One
thing I do have going for me is the same modified "battery pack" I'm using
to troubleshoot my 1502 works fine in my 1503. (I probably should have
mentioned this at the beginning :-)

Anyway, back to the DC converter.

One thing that has be scratching my head is the 25V/9.7V generated from the
secondary controlling the oscillator of the primary. Maybe I just haven't
puzzled it long enough, but I can't seem to wrap my head around it.

Really seems like I need to break out my scope and see what is going on with
Q6349 and Q6435.

- Mark N1VQW

Dave Casey
 

One thing that has be scratching my head is the 25V/9.7V generated from
the
secondary controlling the oscillator of the primary. Maybe I just haven't
puzzled it long enough, but I can't seem to wrap my head around it.
That's a feedback path for the switcher so it can regulate. The three axial
caps on the back of the board are often bad. Being open or leaky could
interfere with the switcher operating properly; I'd check these first. I
have a similarly broken 1502 that's waiting for some new caps so I can
proceed.

Dave Casey

Mark Pilant
 

Hi Dave.

I checked the three caps, and C6415 is open. The other two read good
on my cap/ESR meter (about 400uF and ESR both < .5ohm). I'll definitely
replace the open one, and may do the other two as well.

Looking at the schematic, it looks as though I may be able to test the
board outside the TDR, with a couple of appropriate connections. This
would certainly make things easier.

- Mark N1VQW

fiftythreebuick
 

Mark, as Tom said, it can be a transient condition that trips the power supply.

When developing a battery emulator for my units, at one point mine would fail when the unit was first powered up after sitting idle for a long time, but if you just turned the power off and right back on, it would work every time. The inrush current to all the capacitors in the 1502 was causing a transient low-voltage condition that would trip the power supply. I just increased the size of the capacitor in the emulator and she runs first time, every time now.

As he said, gotta pay close attention to the "subtleties" of this one! :-)

Tom

On Mon, Dec 9, 2019 at 08:53 AM, Tom Gardner wrote:


On 09/12/19 16:38, Mark Pilant wrote:
Hi Tom.

I haven't tried supplying power directly to the banana plugs, but the
parallel cap/resistor battery substitute (ala TekWiki) seems to work
fine, as the battery meter on the front panel indicates the battery is
"good" and I do see 14.4VDC supplying the DC converter.
That's insufficient.

You realise that PSU trips if there is a transient overvoltage. At that point
the voltmeter reads "excessively good" :)

I presume you have read the circuit operation section of the manual?


So it would seem the battery portion of the power supply is working and
I need to determine why the rest of the power supply isn't working.

My general sense is the DC converter is an oscillator driving the T6535
primary to then generate all the secondary voltages needed.  I suspect
the oscillator isn't oscillating (well?), but I'm not sure of the best
way to investigate.  (Not sure how many ways there are to shut it down;
lack of a battery is one.)
One way is the overvoltage detection, via q6547/8/9.


I'm a little wary just probing around with my scope (7904 with the usual
vertical / horizontal plugins), because of potential ground issues.
Although I do have a Tek 224 I can use that shouldn't have any ground
issues.
If you power off a bench supply, then clipping the scope shield to the chassis

is fine.

The 1502's transformer is fully isolated, so**clipping the scope shield to the

chassis is fine.


I just checked, and the power supply for the 1503 is the same as the one
for the 1502.  I may be able to open up my 1503 and use its power supply
to troubleshoot the 1502 power supply.
Yes.

Mark Pilant
 

Hi Tom.

I just increased the size of the capacitor in the emulator and she runs
first time, every time now.
I have a 2200uF cap, per the TekWiki. Works fine for the 1503, so I expect
there is something else going on in the 1502.

As he said, gotta pay close attention to the "subtleties" of this one! :-)
That's why I'm still scratching my head. I really need to find some (of that
rare substance) time to probe with my scope. That way I see what is happening.
(At least I hope :-)

- Mark N1VQW

Tom Gardner
 

I've had a "battery eliminator" that worked in one 1502 but not
another. That's one reason I don't like them. Another is that tabbed C
NiCd cells are easily available, so the battery can be reconstructed.
Battery operation is pleasing; I used one of my 1502s last Saturday in
the middle of a field, to diagnose the cable between a glider's
cockpit and antenna.

You may well be right about the source of a (possibly the) problem,
but using a bench PSU is quick, easy, and eliminates one significant
variable.

On 10/12/2019, Mark Pilant <mark@...> wrote:
Hi Tom.

> I just increased the size of the capacitor in the emulator and she runs
> first time, every time now.

I have a 2200uF cap, per the TekWiki. Works fine for the 1503, so I expect
there is something else going on in the 1502.

> As he said, gotta pay close attention to the "subtleties" of this one!
:-)

That's why I'm still scratching my head. I really need to find some (of
that
rare substance) time to probe with my scope. That way I see what is
happening.
(At least I hope :-)

Mark Pilant
 

tabbed C NiCd cells are easily available, so the battery can be reconstructed.
Very true.

However, my own problem with NiCd batteries is I use the equipment so infrequently
I would be constantly replacing batteries. I understand this is less of an issue
with NiMH and Lithium batteries, but then you can get into making changes to the
charging circuitry to get the correct charge rates.

I may try the bench power supply just to see if it turns up additional information.

- Mark N1VQW

Tom Gardner
 

NiCd cells have a longer shelf life than lithium cells - I believe.

The 1502/3 has a nasty characteristic - even when turned off there is
a current drain of ~1mA through the chart recorder. If left
discharging then eventually the weakest cell will discharge first and
then be reverse charged - and that's bad news, It is behind the
statement that the battery should be fully charged once a month,

My solution is simply to remove the battery pack when not in use. Even
if each cell self-discharges, it won't be reverse charged.

But you should treat that with caution; I'm not a battery expert.

On 10/12/2019, Mark Pilant <mark@...> wrote:
> tabbed C NiCd cells are easily available, so the battery can be
reconstructed.

Very true.

However, my own problem with NiCd batteries is I use the equipment so
infrequently
I would be constantly replacing batteries. I understand this is less of an
issue
with NiMH and Lithium batteries, but then you can get into making changes to
the
charging circuitry to get the correct charge rates.

I may try the bench power supply just to see if it turns up additional
information.

- Mark N1VQW




Majdi S. Abbas
 

On Tue, Dec 10, 2019 at 04:21:02PM -0500, Mark Pilant wrote:
However, my own problem with NiCd batteries is I use the equipment so infrequently
I would be constantly replacing batteries. I understand this is less of an issue
with NiMH and Lithium batteries, but then you can get into making changes to the
charging circuitry to get the correct charge rates.
NiCds are more forgiving than the newer chemestries -- they can
be cold stored fully discharged. Or allowed to self discharge to
nothing, and when needed, you can still trickle charge them back up.

I find them to be optimal for sporadic use applications where
NiMh or Lithium chemistry batteries would be damaged by excessive
discharge.

--msa

Chuck Harris
 

Everything you say is absolutely true, as long as
you never, ever, ever allow a cell to reverse charge.

If you do, it will grow a nickel dendrite through the
inter plate insulation, and become a dead short.

-Chuck Harris

Majdi S. Abbas wrote:
...

NiCds are more forgiving than the newer chemestries -- they can
be cold stored fully discharged. Or allowed to self discharge to
nothing, and when needed, you can still trickle charge them back up.

I find them to be optimal for sporadic use applications where
NiMh or Lithium chemistry batteries would be damaged by excessive
discharge.

--msa

Tom Gardner
 

Precisely.

That's why I remove my 1502 battery packs when I'm not using them. If I don't then sooner or later the 1mA trickle /dis/charge will reverse charge the weakest cell.

On 11/12/19 14:59, Chuck Harris wrote:
Everything you say is absolutely true, as long as
you never, ever, ever allow a cell to reverse charge.

If you do, it will grow a nickel dendrite through the
inter plate insulation, and become a dead short.

-Chuck Harris

Majdi S. Abbas wrote:
...

NiCds are more forgiving than the newer chemestries -- they can
be cold stored fully discharged. Or allowed to self discharge to
nothing, and when needed, you can still trickle charge them back up.

I find them to be optimal for sporadic use applications where
NiMh or Lithium chemistry batteries would be damaged by excessive
discharge.

--msa

John Griessen
 

On 12/11/19 8:59 AM, Chuck Harris wrote:
Everything you say is absolutely true, as long as
you never, ever, ever allow a cell to reverse charge.
So, just stopping the pack discharge is not enough.
The cells need to be managed separately as in today's
lithium ion battery manager IC's and no cells in series...

Tom Gardner
 

On 11/12/19 15:38, John Griessen wrote:
On 12/11/19 8:59 AM, Chuck Harris wrote:
Everything you say is absolutely true, as long as
you never, ever, ever allow a cell to reverse charge.
So, just stopping the pack discharge is not enough.
The cells need to be managed separately as in today's
lithium ion battery manager IC's and no cells in series...
A cell cannot reverse charge itself. The weakest cell in a battery can be reverse charged as the other cells continue to discharge themselves through the load.

If there is no complete circuit (e.g. battery pack not connected to a 1502), then a cell can only self-discharge.

A NiCd cell should be capable of being continuously trickle charged at the C /10 rate or lower, independently of whether or not it is in a battery. If you want to "trickle" charge a NiCd battery at >C/10, then it would be prudent to manage each cell individually.

It is not entirely clear to me whether the C/10 rate is peak or mean. In the 1502 the peak current each 20ms (or 16ms in the US) is noticeably more than the mean current. In the 1502, R6131 sets the mean current to be around C/10.

Other chemistries, especially Li, cannot tolerate continuous trickle charge and each cell has to be managed individually.

John Griessen
 

On 12/11/19 9:50 AM, Tom Gardner wrote:
So, just stopping the pack discharge is not enough.
The cells need to be managed separately as in today's
lithium ion battery manager IC's and no cells in series...
A cell cannot reverse charge itself. The weakest cell in a battery can be reverse charged as the other cells continue to discharge themselves through the load.
So, self discharge can never pull charge in from other cells in series?