Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

output mosfet short when transfomer is connected,voltage drops to low wit little load

Status
Not open for further replies.
M

mossses

Member level 1
Joined
Mar 17, 2012
Messages
38
Helped
1
Reputation
2
Reaction score
1
Trophy points
1,288
Location
9ja
Activity points
1,481
i a novice in electronic and i'll appreciate if anyone can help me with the problem am facing with my diy inverter i built.my output mosfet short at times when i connect my transformer then i get 240volts at d transformer output at 13 volts but when i plug my t.v (79watts) d volt drops to 201volt with a bulb of 60watts it drops again to 180.am using 4 mosfet(irfp250) 2 in parallel.here is d diagram .i replaced d power transistor with mosfet irfp250,1ohm with 220 ohm and 100ohm with 2.2 k.
 

here is d diagram
 

Attachments

  • inversor_4013.pdf
    217 KB · Views: 145

Did you measure the battery voltage drop and AC voltage drop across the primary winding? If it's considerably lower than the secondary voltage drop, I would mainly expect a problem of transformer series resistance.
 

When d battery is at 13.2volts,d secondary is at 14volts then primary 243volts without load.am using a 88ah automobile battery
 

Hola Mossses:

You changed two TIP3055 to 4 IRFP250 and some resistors. I hope you have access to an oscilloscope (2 channel or more). ¿Can you post an oscilloscope graph from both gate voltages (both traces on one screen)? This is to see how they are related in time.

It may be that both mosfets do conduct for a short time during switching. I can't find a provision in your cicuit to avoid this. If this is the case, this will lead to strong current peaks through the mosfets and the transformer.

You may also check the drain waveforms to make sure the mosfets are fully saturated. ¿Are you using a DIY transformer or an existing one?
 

I don't av access to oscilloscope.i buy bought a ready made transformer not a diy one,all i av is a digital muiltmeter with frequency ability
 

When d battery is at 13.2volts,d secondary is at 14volts then primary 243volts without load.am using a 88ah automobile battery
Click to expand...
Yes, the question was, how do these values changes when loading the output?
It may be that both mosfets do conduct for a short time during switching. I can't find a provision in your cicuit to avoid this. If this is the case, this will lead to strong current peaks through the mosfets and the transformer.
Click to expand...
I support your demand for oscilloscope measurements. Missing dead-time usually isn't a problem with low frequency push-pull output stages due to the transformer leakage inductance.
 

FvM said:
Yes, the question was, how do these values changes when loading the output?
Click to expand...
at 13 volts but when i plug my t.v (79watts) d volt drops to 201volt with a bulb of 60watts it drops again to 180

I support your demand for oscilloscope measurements. Missing dead-time usually isn't a problem with low frequency push-pull output stages due to the transformer leakage inductance.[/QUOTE] ;like i said earlier am a novice in electronic i don't understand much of dis electronic terminology.but i wanna know cos i love it
 
Last edited:

What is the VA rating of the transformer? As FvM says, wire resistance may be too high.

FvM: Current may not be that high due to leakage inductance, but the energy has to go somewhere (avalanching the mosfet?). It looks to me the drive doesn't provide dead time. The BJTs, together with the mosfet drive (there is no provision to discharge the gates rapidly), may lead to significant overlap. I am sure you know that having dead time makes a kind of "modified sinewave" reducing the stress when switching capacitive loads.

Maybe adding some capacitors across BE of both 2n3904 may introduce some dead time.
 

the man dat sold d transformer said it a 1000va transformer.no label was on it

---------- Post added at 23:19 ---------- Previous post was at 23:18 ----------
WimRFP said:
What is the VA rating of the transformer? As FvM says, wire resistance may be too high.

FvM: Current may not be that high due to leakage inductance, but the energy has to go somewhere (avalanching the mosfet?). It looks to me the drive doesn't provide dead time. The BJTs, together with the mosfet drive (there is no provision to discharge the gates rapidly), may lead to significant overlap. I am sure you know that having dead time makes a kind of "modified sinewave" reducing the stress when switching capacitive loads.

Maybe adding some capacitors across BE of both 2n3904 may introduce some dead time.
Click to expand...
what will be d value of d capacitor
 

My assumption is, that the leakage inductance between both primary windings of a regularly designed 50/60 Hz transformer is high enough to keep the additional current caused by simulatanous conduction of both transistor neglectable. Leakage inductance will generally cause overvoltage when commutating the output current, the energy will be most likely absorbed by controlled MOSFET avalanche. This problem exist with or without dead-time in gate control.

I don't opt against your suggestions of introducing a clear deadtime. But I don't see an indication, that the observed problems are related to it.
 

A 1000VA transformer has good regulation (few percent), so when the mosfets do saturate, and you have solid supply voltage (sufficiently thick wiring from battery to circuit), the output voltage shouldn't drop that much.

¿What is the weight of the transformer, is is a E-core or toroidal core transformer? Does the circuit run at the intended frequency?

---------- Post added at 23:39 ---------- Previous post was at 23:35 ----------

For debudding the circuit I would really recommend you to get an oscilloscope (for a short time). Are you sure you did use the right components, please double check.. For example using wrong resistors in the mosfet drive may result in malfunction.

---------- Post added at 23:47 ---------- Previous post was at 23:39 ----------

FvM: primary to primary leakage inductance can have lower value then primary to secondary leakage inductance. This all depends on how the transformer is wound, so I don't know for his circuit. Energy in primary to secondary leakage inductance is recycled via the body diode of the mosfet, so that is no problem.

I am not sure for this case, but may simultaneous conduction of the mosfets lead to more current unbalance (flux walking)?
 
  • Like
Reactions: FvM

    FvM

    FvM

    Points: 2
    Helpful Answer Positive Rating
FvM: primary to primary leakage inductance can have lower value then primary to secondary leakage inductance. This all depends on how the transformer is wound, so I don't know for his circuit. Energy in primary to secondary leakage inductance is recycled via the body diode of the mosfet, so that is no problem.
Click to expand...
Yes, good point. You preferably would wind both primary windings bifiliarly to achieve minimal leak inductance. In this case, missing dead-time can matter. You remind me to the fact, that we have effectively no information about the transformer design.
 

i don't know the weight but very heavy the bulky type of transformer.
e-core.

---------- Post added at 00:38 ---------- Previous post was at 00:35 ----------
FvM said:
Yes, good point. You preferably would wind both primary windings bifiliarly to achieve minimal leak inductance. In this case, missing dead-time can matter. You remind me to the fact, that we have effectively no information about the transformer design.
Click to expand...
d transformer was removed from an inverter dat had panel problem rated 1000 watts
 

If the weight is many kgs (5 kgs?), wire resistance will not give significant contribution to the voltage drop, so something else must be wrong. Did you check DC wire resistance of both secondary and primary windings that you are actually using for your inverter?

Did you check the DC voltage between the center conductor of the transformer's low voltage (primary) winding and ground of PCB when applying the lamp load (to make sure you don't have voltage drop)? Do the mosfets run hot rapidly when applying load?
 

WimRFP said:
If the weight is many kgs (5 kgs?), wire resistance will not give significant contribution to the voltage drop, so something else must be wrong. Did you check DC wire resistance of both secondary and primary windings that you are actually using for your inverter?

Did you check the DC voltage between the center conductor of the transformer's low voltage (primary) winding and ground of PCB when applying the lamp load (to make sure you don't have voltage drop)? Do the mosfets run hot rapidly when applying load?
Click to expand...
yes i did and it was around 13volts with load when d battery was 12volts.i dont understand was u mean by Do the mosfets run hot rapidly when applying load
 

I expected that the DC average voltage between "tap" (as shown in the inversor's circuit diagram) and ground on PCB is less the battery voltage. Maybe your volt meter behaves strange because of possible voltage spikes. What type (brand, true rms, etc) do you have?


If you have a DC ampere function on your volt/ampere meter, can you measure the current consumption of the inverter at no load. Make sure you use the 10A range (or higher) of your volt/current meter. If your volt/current meter doesn't have a high current range, don't do the measurement as it may damage your instrument.

With this measurement you can determine the dissipation without load (Pdis = U*I). If this is very high, say tens of Watts, you can be sure the mosfets will become hot and there will be something wrong. you may also do this measurement with (for example) a 25W light bulb. Start with a moderate load, so you can see when something goes wrong.

Did you measure the DC resistance of the transformer windings you are actually using? If so, what are the values?

---------- Post added at 23:29 ---------- Previous post was at 23:17 ----------

Regarding capacitors for creating dead time, I was thinking of around 200nF between B and E of both 2N3904 transistors.

Complete other issue with this circuit.
There is a 1n4007 connected to the 820 Ohms base resistors of the TIP125 darlington transistors. This may result in low collector current that causes some voltage across the gate resistors (that you changed from 100 Ohms to 2.2 kOhms. Who told you to do this modification? You may check the behavior of your inverter when you short the 1N4007 between supply voltage and the 2, 820 Ohms resistors.

Maybe others have an opinion on that diode.
 

WimRFP said:
I expected that the DC average voltage between "tap" (as shown in the inversor's circuit diagram) and ground on PCB is less the battery voltage. Maybe your volt meter behaves strange because of possible voltage spikes. What type (brand, true rms, etc) do you have?


If you have a DC ampere function on your volt/ampere meter, can you measure the current consumption of the inverter at no load. Make sure you use the 10A range (or higher) of your volt/current meter. If your volt/current meter doesn't have a high current range, don't do the measurement as it may damage your instrument.

With this measurement you can determine the dissipation without load (Pdis = U*I). If this is very high, say tens of Watts, you can be sure the mosfets will become hot and there will be something wrong. you may also do this measurement with (for example) a 25W light bulb. Start with a moderate load, so you can see when something goes wrong.

Did you measure the DC resistance of the transformer windings you are actually using? If so, what are the values?

---------- Post added at 23:29 ---------- Previous post was at 23:17 ----------

Regarding capacitors for creating dead time, I was thinking of around 200nF between B and E of both 2N3904 transistors.

Complete other issue with this circuit.
There is a 1n4007 connected to the 820 Ohms base resistors of the TIP125 darlington transistors. This may result in low collector current that causes some voltage across the gate resistors (that you changed from 100 Ohms to 2.2 kOhms. Who told you to do this modification? You may check the behavior of your inverter when you short the 1N4007 between supply voltage and the 2, 820 Ohms resistors.

Maybe others have an opinion on that diode.
Click to expand...
i was told to change the 100ohms to 2.2k by a friend since i want to change to mosfet,do u think having the mosfet will turn on fully if i didnt change it
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top