Transistor equivalent - UK Vintage Radio Repair and Restoration Discussion Forum

Biggles · 13th Oct 2014, 5:47 pm

Hi all, I am looking for an equivalent for a power mosfet (TO220 package) type ON5194. It is not listed in any of my data books and it is noted as obsolete on the only website I can make sense of. An uprated equivalent would be fine as long as the pinout and package are the same due to limited space. Thanks in advance.
Alan.

Silicon · 13th Oct 2014, 11:37 pm

Are you sure it is a TO220 package? Could it be a SOT226 or I2PAK package?

Have a look at the BUK9E06-55A.

Refugee · 14th Oct 2014, 1:20 am

Is it not a re-branded one of these?
http://alltransistors.com/transistor...ransistor=4616

Maarten · 14th Oct 2014, 3:30 am

ON numbers are custom made or custom selected transistors from Philips Semiconductors (NXP). Equivalents or near equivalents often exist but are hard to find out (I maintain a list at http://www.circuitsonline.net/forum/view/113864 but this one isn't in it - well as of now it is, but without relevant data other than a link to this forum topic).

Could you provide a context, such as brand name and model of the equipment? Sometimes a search by equipment turns up an equivalent. Otherwise some values can be assumed from its position in the circuit.

Herald1360 · 14th Oct 2014, 12:48 pm

Quote:

Originally Posted by Refugee

Is it not a re-branded one of these?
http://alltransistors.com/transistor...ransistor=4616

Not if it's a Mosfet it isn't.

Al (astral highway) · 14th Oct 2014, 6:34 pm

Hey, this has very similar characteristics and is a TO220 device, easily available.

Maarten · 14th Oct 2014, 9:14 pm

Similar characteristics to what? I haven't seen any context on the ON5194 device, so I'm afraid your guess is as good as mine. TO220 MOSFETs range roughly from 50V 100A to 900V 2A.

Biggles · 14th Oct 2014, 10:11 pm

Thanks for your replies. The device is a power switch on a module driving a motor at 12 volts in an automotive application but had failed short circuit causing the motor to go full speed all the time. I measured the current at 16A which I thought was a bit heavy so that may have caused the original to fail. I put an IRF640 in rated at 18A as that is what I had available. This blew after a short while leaving two possibilities; either the motor is drawing too much current or the device I used was under rated, hence the request for info on the original component. I still think the motor current may be the problem but I was going to try say a 25 or 30 amp rated transistor to see if it survived. I don't have much info on power fets but the replacement would have to be in a TO220 package to fit the board correctly. Cheers.
Alan.

Al (astral highway) · 14th Oct 2014, 10:40 pm

Quote:

Originally Posted by Maarten

Similar characteristics to what?

Erm, to the original device, obviously. It was a simple matter of seeing what the manufacturer suggested as a substitute to the obsolete part. If the manufacturer suggests it, that's good enough.

And any case, from the details of the application now given (reply #8) it's quite a generic power device.

Radio Wrangler · 14th Oct 2014, 10:54 pm

16A on an 18A rated device is definitely pushing it too far, but the most popular way to dispatch mosfets to the land of their ancestors is to exceed voltage capabilities, even for brief transients. You need to look carefully at how that transistor is protected from its inductive load.

David

DangerMan · 14th Oct 2014, 10:54 pm

If it's automotive it may be a very low Rdson "Trench" FET.
Quite a specific family of devices.

Tyso_Bl · 14th Oct 2014, 11:18 pm

The motor may be on its way out, arcing causing high back EMF across the FET, the snubber protection network may have failed so leaving any replacement vulnerable.

Biggles · 14th Oct 2014, 11:50 pm

I checked the schottky diode protection (another TO220 package) and that is fine. As RW says, I was probably pushing it using a device rated at 18A, when start up surges will most likely be quite a bit higher than that. At least it proved that the module will work if a suitably rated FET is used.

Maarten · 15th Oct 2014, 2:18 am

Quote:

Originally Posted by astral highway

If the manufacturer suggests it, that's good enough.

I'm quite sure I must have missed something. The only suggestion I saw was BUK9E06-55A but I wasn't under the impression that was the manufacturers suggestion. If it indeed was, I will update my list accordingly.

Anyway, given the application, the FET you suggest as a replacement is almost certainly better than original.

The IRF640 may have failed partially because of the excessive current and partially because it may not have been turned on fully (not a logic level FET) which will have increased the already high RDSon losses.

crackle · 15th Oct 2014, 7:30 am

Too much friction making a DC motor run slow will cause it to draw excessive current. Could this be the cause?

Mike

Silicon · 15th Oct 2014, 10:00 am

Alan, you need to do a bit of maths to find out what is going on.

The heat generated in the transistor can be estimated using the formula Watts = Amps squared times RDS(on). The IRF640 has an RDS(on) of 0.18 Ohms.
16 amps will generate 46 Watts of heat.

Is the transistor firmly attached to a large heatsink with a good flow of air?

If not, you need to find a transistor with a very low RDS(on).
The transistor identified in post 6 had an RDS(on) of 5 milliOhms.
It will dissipate 1.28 Watts when passing 16Amps.

Maarten · 15th Oct 2014, 11:00 am

The RDS(on) of 0.18 ohms is only reached at a gate tension of 10V (at room temperature) so it will be way worse in practice, and faillure can occur even with cooling and when the load is behaving exactly as it should.

Biggles · 15th Oct 2014, 4:54 pm

As the unit is a variable speed controller I suspect that the FET is controlled by a simple variation of the gate voltage. I measured that with a multimeter. I don't think it's complicated enough to use PWM technology i.e. I haven't scoped it. So really if an uprated equivalent is available as previous posts suggest then I will go for that. In the mean time the current drawn by the fan is going to be compared with an identical unit. I am awaiting the results. The circuit is fused at 40A which would point towards a high expected current but of course this is not guaranteed.

Silicon · 15th Oct 2014, 7:02 pm

If it is not using PWM technology and is acting like a big variable resistor it will need a big heatsink and a good flow of air.

Suppose it drops the voltage by 6 Volts and limits the current to 8Amps.
It will dissipate 48Watts of Power as heat.

Biggles · 15th Oct 2014, 7:10 pm

The more I think about it the more it seems likely that the motor is drawing too much current and that killed the original FET. Off the top of my head I would say that a fan motor would draw about 5A?

13th Oct 2014, 5:47 pm	#1
Biggles Rest in Peace Join Date: Feb 2004 Location: Hexham, Northumberland, UK. Posts: 2,234	Transistor equivalent Hi all, I am looking for an equivalent for a power mosfet (TO220 package) type ON5194. It is not listed in any of my data books and it is noted as obsolete on the only website I can make sense of. An uprated equivalent would be fine as long as the pinout and package are the same due to limited space. Thanks in advance. Alan.

13th Oct 2014, 11:37 pm	#2
Silicon Nonode Join Date: Jun 2007 Location: Coulsdon, London, UK. Posts: 2,163	Re: Transistor equivalent Are you sure it is a TO220 package? Could it be a SOT226 or I2PAK package? Have a look at the BUK9E06-55A.

14th Oct 2014, 1:20 am	#3
Refugee Dekatron Join Date: Apr 2012 Location: Worksop, Nottinghamshire, UK. Posts: 5,553	Re: Transistor equivalent Is it not a re-branded one of these? http://alltransistors.com/transistor...ransistor=4616

14th Oct 2014, 3:30 am	#4
Maarten Dekatron Join Date: Dec 2007 Location: Haarlem, Netherlands Posts: 4,199	Re: Transistor equivalent ON numbers are custom made or custom selected transistors from Philips Semiconductors (NXP). Equivalents or near equivalents often exist but are hard to find out (I maintain a list at http://www.circuitsonline.net/forum/view/113864 but this one isn't in it - well as of now it is, but without relevant data other than a link to this forum topic). Could you provide a context, such as brand name and model of the equipment? Sometimes a search by equipment turns up an equivalent. Otherwise some values can be assumed from its position in the circuit. Last edited by Maarten; 14th Oct 2014 at 3:39 am.

14th Oct 2014, 6:34 pm	#6
Al (astral highway) Dekatron Join Date: Nov 2006 Location: London, UK. Posts: 3,496	Re: Transistor equivalent Hey, this has very similar characteristics and is a TO220 device, easily available. __________________ Al

14th Oct 2014, 9:14 pm	#7
Maarten Dekatron Join Date: Dec 2007 Location: Haarlem, Netherlands Posts: 4,199	Re: Transistor equivalent Similar characteristics to what? I haven't seen any context on the ON5194 device, so I'm afraid your guess is as good as mine. TO220 MOSFETs range roughly from 50V 100A to 900V 2A.

14th Oct 2014, 10:11 pm	#8
Biggles Rest in Peace Join Date: Feb 2004 Location: Hexham, Northumberland, UK. Posts: 2,234	Re: Transistor equivalent Thanks for your replies. The device is a power switch on a module driving a motor at 12 volts in an automotive application but had failed short circuit causing the motor to go full speed all the time. I measured the current at 16A which I thought was a bit heavy so that may have caused the original to fail. I put an IRF640 in rated at 18A as that is what I had available. This blew after a short while leaving two possibilities; either the motor is drawing too much current or the device I used was under rated, hence the request for info on the original component. I still think the motor current may be the problem but I was going to try say a 25 or 30 amp rated transistor to see if it survived. I don't have much info on power fets but the replacement would have to be in a TO220 package to fit the board correctly. Cheers. Alan.

14th Oct 2014, 10:54 pm	#10
Radio Wrangler Moderator Join Date: Mar 2012 Location: Fife, Scotland, UK. Posts: 22,874	Re: Transistor equivalent 16A on an 18A rated device is definitely pushing it too far, but the most popular way to dispatch mosfets to the land of their ancestors is to exceed voltage capabilities, even for brief transients. You need to look carefully at how that transistor is protected from its inductive load. David __________________ Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done

14th Oct 2014, 10:54 pm	#11
DangerMan Heptode Join Date: Jul 2009 Location: Stoke-on-Trent, Staffordshire, UK. Posts: 719	Re: Transistor equivalent If it's automotive it may be a very low Rdson "Trench" FET. Quite a specific family of devices.

14th Oct 2014, 11:18 pm	#12
Tyso_Bl Hexode Join Date: Jul 2014 Location: Magor, Monmouthshire, Wales, UK. Posts: 436	Re: Transistor equivalent The motor may be on its way out, arcing causing high back EMF across the FET, the snubber protection network may have failed so leaving any replacement vulnerable. __________________ Adapt, Improvise, Oh Bother.....

14th Oct 2014, 11:50 pm	#13
Biggles Rest in Peace Join Date: Feb 2004 Location: Hexham, Northumberland, UK. Posts: 2,234	Re: Transistor equivalent I checked the schottky diode protection (another TO220 package) and that is fine. As RW says, I was probably pushing it using a device rated at 18A, when start up surges will most likely be quite a bit higher than that. At least it proved that the module will work if a suitably rated FET is used.

15th Oct 2014, 7:30 am	#15
crackle Rest in Peace Join Date: Mar 2011 Location: Basildon, Essex, UK. Posts: 4,100	Re: Transistor equivalent Too much friction making a DC motor run slow will cause it to draw excessive current. Could this be the cause? Mike

15th Oct 2014, 10:00 am	#16
Silicon Nonode Join Date: Jun 2007 Location: Coulsdon, London, UK. Posts: 2,163	Re: Transistor equivalent Alan, you need to do a bit of maths to find out what is going on. The heat generated in the transistor can be estimated using the formula Watts = Amps squared times RDS(on). The IRF640 has an RDS(on) of 0.18 Ohms. 16 amps will generate 46 Watts of heat. Is the transistor firmly attached to a large heatsink with a good flow of air? If not, you need to find a transistor with a very low RDS(on). The transistor identified in post 6 had an RDS(on) of 5 milliOhms. It will dissipate 1.28 Watts when passing 16Amps.

15th Oct 2014, 11:00 am	#17
Maarten Dekatron Join Date: Dec 2007 Location: Haarlem, Netherlands Posts: 4,199	Re: Transistor equivalent The RDS(on) of 0.18 ohms is only reached at a gate tension of 10V (at room temperature) so it will be way worse in practice, and faillure can occur even with cooling and when the load is behaving exactly as it should.

15th Oct 2014, 4:54 pm	#18
Biggles Rest in Peace Join Date: Feb 2004 Location: Hexham, Northumberland, UK. Posts: 2,234	Re: Transistor equivalent As the unit is a variable speed controller I suspect that the FET is controlled by a simple variation of the gate voltage. I measured that with a multimeter. I don't think it's complicated enough to use PWM technology i.e. I haven't scoped it. So really if an uprated equivalent is available as previous posts suggest then I will go for that. In the mean time the current drawn by the fan is going to be compared with an identical unit. I am awaiting the results. The circuit is fused at 40A which would point towards a high expected current but of course this is not guaranteed.

15th Oct 2014, 7:02 pm	#19
Silicon Nonode Join Date: Jun 2007 Location: Coulsdon, London, UK. Posts: 2,163	Re: Transistor equivalent If it is not using PWM technology and is acting like a big variable resistor it will need a big heatsink and a good flow of air. Suppose it drops the voltage by 6 Volts and limits the current to 8Amps. It will dissipate 48Watts of Power as heat.

15th Oct 2014, 7:10 pm	#20
Biggles Rest in Peace Join Date: Feb 2004 Location: Hexham, Northumberland, UK. Posts: 2,234	Re: Transistor equivalent The more I think about it the more it seems likely that the motor is drawing too much current and that killed the original FET. Off the top of my head I would say that a fan motor would draw about 5A?