# Zener voltage limiter without voltage drop to the load

Started by July 9, 2012
```Schematic at link below.

Input voltage varies from 8-18V. The voltage to relay shouldn't drop as long as it is less than 12V and remain at 12V when the input voltage raises above 12V.  Will the above circuit do the job?

All the Zener voltage limiters I see have current limiting resistor in between the input and load before the zener. I cannot have that topology as limiting resistor between input and load drops the voltage to load.

What should be the resistor value and zener wattage?

bhav
```
```On Jul 9, 12:36=A0pm, bhavanire...@gmail.com wrote:
>
>
> Input voltage varies from 8-18V. The voltage to relay shouldn't drop as l=
ong as it is less than 12V and remain at 12V when the input voltage raises =
above 12V. =A0Will the above circuit do the job?

Nope.  The 12 V output should come from the top of the zener.  But the
the 275 ohm coil load is too much for a zener reference. (At least the
ones I know of.)

How about a three terminal voltage regulator (LM7812.. or cousin.)
(Hmmm, maybe not so good at the low voltage end.)
George H.
>
> All the Zener voltage limiters I see have current limiting resistor in be=
tween the input and load before the zener. I cannot have that topology as l=
>
> What should be the resistor value and zener wattage?
>
> bhav

```
```On 7/9/2012 9:36 AM, bhavanireddy@gmail.com wrote:
>
>
> Input voltage varies from 8-18V. The voltage to relay shouldn't drop as long as it is less than 12V and remain at 12V when the input voltage raises above 12V.  Will the above circuit do the job?
>
> All the Zener voltage limiters I see have current limiting resistor in between the input and load before the zener. I cannot have that topology as limiting resistor between input and load drops the voltage to load.
>
> What should be the resistor value and zener wattage?
>
> bhav

How many of these are you gonna build?
Stated another way...is this something you can tweak once and be done?
Or is it a production device that has to work over a wide range of relay
vendors
and temperatures?

You can't reliably run a 12V relay on 8V.  That's why they call it a
12V relay.  Yes, you can often make ONE work.

Use a 5V relay and drive it with a current source.  LM317 and a resistor?
```
```On Monday, July 9, 2012 12:36:22 PM UTC-4, (unknown) wrote:
>=20
>=20
> Input voltage varies from 8-18V. The voltage to relay shouldn't drop as l=
ong as it is less than 12V and remain at 12V when the input voltage raises =
above 12V.  Will the above circuit do the job?=20
>=20
> All the Zener voltage limiters I see have current limiting resistor in be=
tween the input and load before the zener. I cannot have that topology as l=
>=20
> What should be the resistor value and zener wattage?
>=20
> bhav

You would need something like this, not pretty:
Please view in a fixed-width font such as Courier.

.
.
.
.
.          8V-18V  ---------+----------+---------.
.                           |          |         |
.                           |          |         |
.                           | +        |         | +
.                           |          |         ) 12V
.                          ---/      [2.7K]      ) 275R
.                 1N4742A // \ 12V     |         ) relay
.                          --- zener   |         | coil
.                           |  1W      '---------+ -
.                           | -                  |
.                           |                    |
.                           |            1N4002  |
.                           |          .--|>|----+
.                           |          |         |
.                           |          |         |
.                           |          |       |<
.                           +----------+-------|    TIP42C
.                           |                  |\
.                           |                    |
.                           |                    |
.                         [390]                  |
.                           |                   ---
.                           |                   ///
.                           |
.                           |
.                           |
.                           D
.           ON/OFF  ---- G    NUD3124
.                           S
.                           |
.                           |
.                           |
.                          ---
.                          ///
.
.
.

I am assuming this is automotive application and the 8V is due to cranking =
cycle. In that case, the 8V will hold the relay in provided it was pulled i=
n while BATT was at 12V. There are integrated constant current relay driver=
s, mostly working on the PWM step down principle. See http://www.ti.com/lit=
http://www.supertex.com/pdf/datasheets/HV9901.pdf
maybe others- you look them up.
```
```bhavanireddy@gmail.com wrote:
>
>
> Input voltage varies from 8-18V. The voltage to relay shouldn't drop
> as long as it is less than 12V and remain at 12V when the input
> voltage raises above 12V.  Will the above circuit do the job?
>
> All the Zener voltage limiters I see have current limiting resistor
> in between the input and load before the zener. I cannot have that
> topology as limiting resistor between input and load drops the
>
> What should be the resistor value and zener wattage?
>

Using a depletion mode FET should be better, almost no voltage drop
until it reaches 12V, then starts to hold. Just as an idea to get you
started:

8-18V   --------o-------'
|       |
|       |
\       |
/       '
\       |
/     |-+
|      <- Depletion mode N-channel
o----||-+
|       |
.---------'       |
|         |       |                 \
|         |       '-----o--------_ o o
|         z       '     |         )|
|         A             o         )|
||-+         | 12V         -        _)|
Logic Low: On    ||<-         | Zener       ^       |
Logic High: Off -||-+         |             |       |
|         |             |       |
GND       GND           GND     GND
(created by AACircuit v1.28.6 beta 04/19/05 www.tech-chat.de)

Maybe the BSP149 works and it's cheap:

http://www.infineon.com/dgdl/BSP149_Rev1.2.pdf?folderId=db3a304412b407950112b408e8c90004&fileId=db3a304412b407950112b42ef8fa4acf

The zener voltage has to be adapted to your needs, this circuit will let
the relay coil voltage go slightly high than the zener voltage.

The other FET can be a cheap 7002 or BSS123. Or if you have ULN relay
drivers then you don't need it. Just note that the logic is reversed,
low means the relay coil is powered.

--
Regards, Joerg

http://www.analogconsultants.com/
```
```On Mon, 9 Jul 2012 09:36:22 -0700 (PDT), bhavanireddy@gmail.com wrote:

>
>
>Input voltage varies from 8-18V. The voltage to relay shouldn't drop as long as it is less than 12V and remain at 12V when the input voltage raises above 12V.  Will the above circuit do the job?

---
No.

If Vcc drops to 8V there is no mechanism in the circuit which can
raise the coil voltage to 12V, and if Vcc is a stiff voltage source
the voltage on pin 1 of the coil will always be Vcc.
---

>All the Zener voltage limiters I see have current limiting resistor in between the input and load before the zener. I cannot have that topology as limiting resistor between input and load drops the voltage to load.
>
>What should be the resistor value and zener wattage?

---
Since the circuit can't work if the relay needs to be energized when
Vcc is 8V, the question is moot.

As mike posted earlier, you need to select a relay which will pick up
when Vcc is at 8V, and then use a series pass element to limit the
coil current (or voltage) when Vcc rises above 8V.

I see you're using a relay with a 500 milliwatt coil so, assuming you
can get the same relay with a 5V coil, something like this would work.

(View using a fixed-pitch font.)

-                2N2222
Vcc>-----+-------C   E---+--------+
|         B     |        |
[270R]       |     |        |
|         |     |        |
+---------+     |        |K
|K        | [5V COIL] [DIODE]
[Z6V2]       |     |        |
|         |     |        |
__       |         C     |        |
ON/OFF>--|-[2K2]-B 2N2222|        |
|         E     |        |
|         |     |        |
GND>-----+---------+-----+--------+

--
JF
```
```John Fields wrote:
> On Mon, 9 Jul 2012 09:36:22 -0700 (PDT), bhavanireddy@gmail.com wrote:
>
>>
>>
>> Input voltage varies from 8-18V. The voltage to relay shouldn't drop as long as it is less than 12V and remain at 12V when the input voltage raises above 12V.  Will the above circuit do the job?
>
> ---
> No.
>
> If Vcc drops to 8V there is no mechanism in the circuit which can
> raise the coil voltage to 12V, and if Vcc is a stiff voltage source
> the voltage on pin 1 of the coil will always be Vcc.
> ---
>
>> All the Zener voltage limiters I see have current limiting resistor in between the input and load before the zener. I cannot have that topology as limiting resistor between input and load drops the voltage to load.
>>
>> What should be the resistor value and zener wattage?
>
> ---
> Since the circuit can't work if the relay needs to be energized when
> Vcc is 8V, the question is moot.
>

Not necessarily. One might consider a "kicker circuit" that produces a
spike out of an inductor and then drops back to 8V. For most 12V relays
8V is enough hold voltage.

[...]

--
Regards, Joerg

http://www.analogconsultants.com/
```
```On Monday, July 9, 2012 4:01:42 PM UTC-4, Joerg wrote:
> bhavanireddy@gmail.com wrote:
> &gt; Schematic at link below.
> &gt;=20
> &gt;=20
> &gt; Input voltage varies from 8-18V. The voltage to relay shouldn&#39;t =
drop
> &gt; as long as it is less than 12V and remain at 12V when the input
> &gt; voltage raises above 12V.  Will the above circuit do the job?
> &gt;=20
> &gt; All the Zener voltage limiters I see have current limiting resistor
> &gt; in between the input and load before the zener. I cannot have that
> &gt; topology as limiting resistor between input and load drops the
> &gt;=20
> &gt; What should be the resistor value and zener wattage?
> &gt;=20
> &gt; Thanks in advance. bhav
>=20
>=20
> Using a depletion mode FET should be better, almost no voltage drop
> until it reaches 12V, then starts to hold. Just as an idea to get you
> started:
>=20
>=20
>=20
>                 8-18V   --------o-------&#39;
>                                 |       |
>                                 |       |
>                                 \       |
>                                 /       &#39;
>                                 \       |
>                                 /     |-+
>                                 |      &lt;- Depletion mode N-channel
>                                 o----||-+
>                                 |       |
>                       .---------&#39;       |
>                       |         |       |                 \
>                       |         |       &#39;-----o--------_ o o
>                       |         z       &#39;     |         )|
>                       |         A             o         )|
>                    ||-+         | 12V         -        _)|
>   Logic Low: On    ||&lt;-         | Zener       ^       |
>   Logic High: Off -||-+         |             |       |
>                       |         |             |       |
>                      GND       GND           GND     GND
> (created by AACircuit v1.28.6 beta 04/19/05 www.tech-chat.de)
>=20
>=20
> Maybe the BSP149 works and it&#39;s cheap:
>=20
> http://www.infineon.com/dgdl/BSP149_Rev1.2.pdf?folderId=3Ddb3a304412b4079=
50112b408e8c90004&amp;fileId=3Ddb3a304412b407950112b42ef8fa4acf
>=20
> The zener voltage has to be adapted to your needs, this circuit will let
> the relay coil voltage go slightly high than the zener voltage.
>=20
> The other FET can be a cheap 7002 or BSS123. Or if you have ULN relay
> drivers then you don&#39;t need it. Just note that the logic is reversed,
> low means the relay coil is powered.
>=20
> --=20
> Regards, Joerg
>=20
> http://www.analogconsultants.com/

Maybe not an unqualified success over variation in VGS(TH) and all relays a=
nd environmental conditions. The problem: may not drop out. This hysteria a=
bout not getting enough voltage across the relay is unfounded. Most manufac=
turers comply with the SAE minimums so relay will pull-in at low cranking v=
oltages and across temperature. Here is a detailed characterization:  http:=
//pewa.panasonic.com/assets/pcsd/catalog/jj-m-catalog.pdf
All the OP needs to worry about is definitely down regulating anything over=
12V in such a way that there is not an outrageous overhead drop at lesser =
voltages. This does not require anything extreme, like going to a depletion=
mode or using a 6V relay.
```
```bloggs.fredbloggs.fred@gmail.com wrote:
> On Monday, July 9, 2012 4:01:42 PM UTC-4, Joerg wrote:
>> bhavanireddy@gmail.com wrote: &gt; Schematic at link below. &gt;
>> &gt;
>> &gt; &gt; Input voltage varies from 8-18V. The voltage to relay
>> shouldn&#39;t drop &gt; as long as it is less than 12V and remain
>> at 12V when the input &gt; voltage raises above 12V.  Will the
>> above circuit do the job? &gt; &gt; All the Zener voltage limiters
>> I see have current limiting resistor &gt; in between the input and
>> load before the zener. I cannot have that &gt; topology as limiting
>> resistor between input and load drops the &gt; voltage to load.
>> &gt; &gt; What should be the resistor value and zener wattage? &gt;
>>  &gt; Thanks in advance. bhav
>>
>>
>> Using a depletion mode FET should be better, almost no voltage drop
>>  until it reaches 12V, then starts to hold. Just as an idea to get
>> you started:
>>
>>
>>
>> 8-18V   --------o-------&#39; |       | |       | \       | /
>> &#39; \       | /     |-+ |      &lt;- Depletion mode N-channel
>> o----||-+ |       | .---------&#39;       | |         |       |
>> \ |         |       &#39;-----o--------_ o o |         z
>> &#39;     |         )| |         A             o         )| ||-+
>> | 12V         -        _)| Logic Low: On    ||&lt;-         | Zener
>> ^       | Logic High: Off -||-+         |             |       | |
>> |             |       | GND       GND           GND     GND
>> (created by AACircuit v1.28.6 beta 04/19/05 www.tech-chat.de)
>>
>>
>> Maybe the BSP149 works and it&#39;s cheap:
>>
>> http://www.infineon.com/dgdl/BSP149_Rev1.2.pdf?folderId=db3a304412b407950112b408e8c90004&amp;fileId=db3a304412b407950112b42ef8fa4acf
>>
>>
>> The zener voltage has to be adapted to your needs, this circuit
>> will let the relay coil voltage go slightly high than the zener
>> voltage.
>>
>> The other FET can be a cheap 7002 or BSS123. Or if you have ULN
>> relay drivers then you don&#39;t need it. Just note that the logic
>> is reversed, low means the relay coil is powered.
>>
>> -- Regards, Joerg
>>
>> http://www.analogconsultants.com/
>
> Maybe not an unqualified success over variation in VGS(TH) and all
> relays and environmental conditions. The problem: may not drop out.

You are right, it may not drop off if the relay tends to still "stick"
at 2V or so.

> This hysteria about not getting enough voltage across the relay is
> unfounded. Most manufacturers comply with the SAE minimums so relay
> will pull-in at low cranking voltages and across temperature. Here is
> a detailed characterization:
> http://pewa.panasonic.com/assets/pcsd/catalog/jj-m-catalog.pdf ...

I tested some relays a couple weeks ago for a project and they would not
reliably pull in unless I went to about 65% of nominal coil voltage.

>                                                          ... All
> the OP needs to worry about is definitely down regulating anything
> over 12V in such a way that there is not an outrageous overhead drop
> at lesser voltages. This does not require anything extreme, like
> going to a depletion mode or using a 6V relay.

Or just use a regulated PWM driver from iC-Haus or similar.

--
Regards, Joerg

http://www.analogconsultants.com/
```
```On 7/9/2012 5:57 PM, bloggs.fredbloggs.fred@gmail.com wrote:
> On Monday, July 9, 2012 4:01:42 PM UTC-4, Joerg wrote:
>> bhavanireddy@gmail.com wrote:
>> &gt; Schematic at link below.
>> &gt;
>> &gt;
>> &gt; Input voltage varies from 8-18V. The voltage to relay shouldn&#39;t drop
>> &gt; as long as it is less than 12V and remain at 12V when the input
>> &gt; voltage raises above 12V.  Will the above circuit do the job?
>> &gt;
>> &gt; All the Zener voltage limiters I see have current limiting resistor
>> &gt; in between the input and load before the zener. I cannot have that
>> &gt; topology as limiting resistor between input and load drops the
>> &gt;
>> &gt; What should be the resistor value and zener wattage?
>> &gt;
>> &gt; Thanks in advance. bhav
>>
>>
>> Using a depletion mode FET should be better, almost no voltage drop
>> until it reaches 12V, then starts to hold. Just as an idea to get you
>> started:
>>
>>
>>
>>                  8-18V   --------o-------&#39;
>>                                  |       |
>>                                  |       |
>>                                  \       |
>>                                  /&#39;
>>                                  \       |
>>                                  /     |-+
>>                                  |&lt;- Depletion mode N-channel
>>                                  o----||-+
>>                                  |       |
>>                        .---------&#39;       |
>>                        |         |       |                 \
>>                        |         |&#39;-----o--------_ o o
>>                        |         z&#39;     |         )|
>>                        |         A             o         )|
>>                     ||-+         | 12V         -        _)|
>>    Logic Low: On    ||&lt;-         | Zener       ^       |
>>    Logic High: Off -||-+         |             |       |
>>                        |         |             |       |
>>                       GND       GND           GND     GND
>> (created by AACircuit v1.28.6 beta 04/19/05 www.tech-chat.de)
>>
>>
>> Maybe the BSP149 works and it&#39;s cheap:
>>
>> http://www.infineon.com/dgdl/BSP149_Rev1.2.pdf?folderId=db3a304412b407950112b408e8c90004&amp;fileId=db3a304412b407950112b42ef8fa4acf
>>
>> The zener voltage has to be adapted to your needs, this circuit will let
>> the relay coil voltage go slightly high than the zener voltage.
>>
>> The other FET can be a cheap 7002 or BSS123. Or if you have ULN relay
>> drivers then you don&#39;t need it. Just note that the logic is reversed,
>> low means the relay coil is powered.
>>
>> --
>> Regards, Joerg
>>
>> http://www.analogconsultants.com/
>
> Maybe not an unqualified success over variation in VGS(TH) and all relays and environmental conditions. The problem: may not drop out. This hysteria about not getting enough voltage across the relay is unfounded. Most manufacturers comply with the SAE minimums so relay will pull-in at low cranking voltages and across temperature. Here is a detailed characterization:  http://pewa.panasonic.com/assets/pcsd/catalog/jj-m-catalog.pdf
> All the OP needs to worry about is definitely down regulating anything over 12V in such a way that there is not an outrageous overhead drop at lesser voltages. This does not require anything extreme, like going to a depletion mode or using a 6V relay.

That's the second time today, in this thread, that I've seen the word "most"
as justification for a circuit topology.
If you'd said that in a design review, I'd have sent you to your room
without supper.

The only justification for using a 12V relay in this design is if the relay
is in some subassembly over which you have no control.  In that case,
a topology that relies on a parameter of an uncontrolled component is
ludicrous.  If you have control, design a circuit that works by design,
not by accident.
Don't confuse clever design with stupid design.  The difference is that
clever designs work.

If the op wants to build just one and is satisfied that it works most
of the time, I'd say, "go for it!"
```