# H-bridge blowing up in my face...

#### Juhahoo

Joined Jun 3, 2019
249
That almost meets my requirements ... the drawback is that an extra output (intput, in this case) is needed for it to work ... perhaps if an or-gate were added to CW/CCW and its output used as the kill switch?
With a kill switch, you dont even have to worry shoot through, you just need to make sure kill switch is put ON the last, and OFF first... So you set your inputs as needed first, then activate the H bridge with kill switch to make your solenoid do its thing, and then kill it again OFF, then put your inputs to "idle"

#### Tonyr1084

Joined Sep 24, 2015
6,451
I first learned this when, as a 12-year old I got a flying capacitor can to the eye (thankfully no real injury) when the poor little thing got mains voltage applied to it for a while.
As a teen I had a car radio and a 12 volt battery charger. I tried hooking them up to play music in my room. MAN I got a terrible hum. A friend recommended I use a filtering capacitor. Never told me what size to use or where to put it. First, I took an electrolytic cap, don't remember the value, probably 16V. Hooked it to the radio across the power and ground leads. Still had quite the hum. Next I moved it to the secondary output. Hum was even worse. So I moved it to the primary. VERY exciting events took place.
Need to check the back of my shed for that last one ... maybe I'll find an Apollo 11 surplus space suit that I could use.
My 'hoarding' neighbor must have something. This guy paid $100 a month for a storage shed for three years to hold junk. One day I asked him "What's in there that's worth$3600? That afternoon he was emptying that shed.

#### LowQCab

Joined Nov 6, 2012
1,287
What is a "3-Port-Latch-Valve" ?
Sounds rather implausible.
.
.
.

#### MrChips

Joined Oct 2, 2009
24,601
My 'hoarding' neighbor must have something. This guy paid $100 a month for a storage shed for three years to hold junk. One day I asked him "What's in there that's worth$3600? That afternoon he was emptying that shed.
Off topic.
I too suffer from the same predicament. Though often tempted I have yet to rent storage space and have been prevented from doing so by the same logic.

#### cmartinez

Joined Jan 17, 2007
7,510
@crutschow , things just got serious ... I changed the pull up resistors to 750Ω (to be on the safest side possible) ... and the thing worked only for a couple of cycles before it caught fire ... literally ...

I now have to remake the entire PCB using a different circuit ... *sigh* ... back to the drawing board.

#### cmartinez

Joined Jan 17, 2007
7,510
I think I might have found a solution. And by only rearranging the triggering transistors and adding 2 resistors.

Better yet, with this arrangement all the transistors are always off when CW/CCW are down. (the transistors are being triggered by a logic high signal of 3.3V, btw ... not 5V as I previously mentioned). In the previous design, M2 and M4 were on when CW/CCW were low. And that's the main reason for the malfunction.

Anyway, when i.e. CW goes high, only M3 and M4 are turned on, and when CCW goes high, the same happens with M1 and M2. There should never be a short circuit with this logic if CW and CCW are never turned on at the same time. Which is easy to accomplish since those signals are being produced by an MCU.

Here's the sim:

One thing drew my attention, though. What's with the bouncy thingy voltage shown at n001 on the falling edges of CW and CCW?

Any thoughts?

EDIT: Removed redundant R1 and R2

#### Attachments

• 81.7 KB Views: 90
Last edited:

#### MrChips

Joined Oct 2, 2009
24,601
If you are using MCU, check the PWM feature. They usually include turn-on delays for this very same reason.

#### cmartinez

Joined Jan 17, 2007
7,510
If you are using MCU, check the PWM feature. They usually include turn-on delays for this very same reason.
Thanks, I already know how to use the MCU's PWM function. But luckily in this case the solenoid will always be either fully on or fully off for about only 200ms

#### Alec_t

Joined Sep 17, 2013
12,309
What's with the bouncy thingy voltage shown at n001 on the falling edges of CW and CCW?
Inductive kick getting back to n001 at solenoid switch-off. Where is n001 in your circuit?

#### cmartinez

Joined Jan 17, 2007
7,510
Inductive kick getting back to n001 at solenoid switch-off. Where is n001 in your circuit?
It's labeled on top of M6

#### cmartinez

Joined Jan 17, 2007
7,510
I think I might have found a solution. And by only rearranging the triggering transistors and adding 4 resistors.

Better yet, with this arrangement all the transistors are always off when CW/CCW are down. (the transistors are being triggered by a logic high signal of 3.3V, btw ... not 5V as I previously mentioned). In the previous design, M2 and M4 were on when CW/CCW were low. And that's the main reason for the malfunction.

Anyway, when i.e. CW goes high, only M3 and M4 are turned on, and when CCW goes high, the same happens with M1 and M2. There should never be a short circuit with this logic if CW and CCW are never turned on at the same time. Which is easy to accomplish since those signals are being produced by an MCU.

Here's the sim:

One thing drew my attention, though. What's with the bouncy thingy voltage shown at n001 on the falling edges of CW and CCW?

Any thoughts?
I've just figured that R6 and R7 are unnecessary. And not only that, they're actually in parallel with R1 and R2. A big redundant no-no... duh!

#### strantor

Joined Oct 3, 2010
5,694
You know you could do this with a couple of relays.

Parts count:2
Risk of shoot-through: 0
Power draw at idle:0
Inductive load voltage spike tolerance: excellent compared to solid state

If you need PWM (doesn't seem like you would for this), use the relays just to steer the current and throw a single switching transistor in the mix.

Last edited:

#### cmartinez

Joined Jan 17, 2007
7,510
You know you could do this with a couple of relays.

Parts count:2
Risk of shoot-through: 0
Power draw at idle:0
Inductive load voltage spike tolerance: excellent compared to solid state

View attachment 248739

If you need PWM (doesn't seem like you would for this), use the relays just to steer the current and throw a single switching transistor in the mix.

View attachment 248740
Thanks for chiming in, my friend. But I see relays as uncompliant with my requirements of low power draw during activation. It's a device that is battery powered, so the use of solid state components is a must.

#### crutschow

Joined Mar 14, 2008
28,155
so the use of solid state components is a must.
So I would then think your circuit in post #26 (minus the redundant resistors) would fit the bill.

#### strantor

Joined Oct 3, 2010
5,694
Thanks for chiming in, my friend. But I see relays as uncompliant with my requirements of low power draw during activation. It's a device that is battery powered, so the use of solid state components is a must.
Sorry I missed that part. Isn't this a latching solenoid? I would think it requires only a pulse, and a brief relay activation would be negligible like a camera flash (or like the activation of the solenoid itself). Also, there are relays with low power coils.

Last edited:

#### cmartinez

Joined Jan 17, 2007
7,510
Sorry I missed that part. Isn't this a latching solenoid? I would think it requires only a pulse, and a brief relay activation would be negligible like a camera flash (or like the activation of the solenoid itself). Also, there are relays with low power coils.
Yup... it's a latching solenoid, and it only takes a pulse longer than 100ms to make it latch, according to its specs. But I'm feeding it a 200ms pulse just to be on the safe side.

#### strantor

Joined Oct 3, 2010
5,694
Yup... it's a latching solenoid, and it only takes a pulse longer than 100ms to make it latch, according to its specs. But I'm feeding it a 200ms pulse just to be on the safe side.
Here's a selection of relays which might work (probably unnecessarily constrained by criteria i just assumed) which all have a coil current < 40mA, which by my math is only 1% of your solenoid current.

#### eetech00

Joined Jun 8, 2013
2,644
Yup... it's a latching solenoid, and it only takes a pulse longer than 100ms to make it latch, according to its specs. But I'm feeding it a 200ms pulse just to be on the safe side.
How much current for 200ms does the solenoid require?

#### cmartinez

Joined Jan 17, 2007
7,510
How much current for 200ms does the solenoid require?
3.8 amps, at 16VDC

#### Janis59

Joined Aug 21, 2017
1,356
You have highly inductive load yet no any depfer circ to avoid the voltage spikes. Apply the resistor +diode+capacitor and probably will work better.