Heat sink for full wave rectifier? (metalworking content)

Yup, and in the spirit of reasonable excess for cheap, I'd suggest picking up one or two of these babies for $2.35 each quantity 1.

formatting link

Best regards, Spehro Pefhany

Reply to
Spehro Pefhany
Loading thread data ...

Thanks to everyone for their input. I have a bigass finned aluminum heat sink, about 10" square, that will go on the rectifier. And will bolt the heat sink to the enclosure for good measure.

And I also appreciate the education on calculating heat dissipation. My knowledge of electronics is none too good. I understood the five-tube radio I built in high school, but these confounded solid-state thingies just mess up my head. ;-)

-- Best -- Terry

Reply to
Terry

ok, have it your way - here is a reference showing that what I said is right

formatting link
there are at least 10,000 other references - perhaps you have found some new kind of silicon, with new semiconductor properties. Go ahead and give more bad information to the OP, this is not my problem. In fact, maybe YOU should take some measurements and check your facts.

and no, I will not respond, I have no need to get into a screaming contest for no money with people who have no information. I will let the OP do his own research and determine what he wishes to do. I've offered my recommendation, YOU choose to tell me I am wrong.

Reply to
Bill Noble

Guys, can I ask an ignorant question. I thought that voltage drop depends on the maximum reverse voltage that the diode would withstand? Is that true or not?

In other words, is the voltage drop on a 1,000V rated diode, the same as on a 30v rated diode?

i
Reply to
Ignoramus25553

there you have it folks- the man cited electronics-tutorials.ws

he's right, anybody else is wrong.

case closed.

Reply to
Cydrome Leader
[ ... ]

For a single junction -- no. Pretty much the same range as a function of the current through the device.

However -- high voltage rectifiers can be multi-junction devices

-- junctions in series -- to get sufficient PRV from a device which can handle the current needed.

These multi-junction devices will drop a nominal FV for each junction in series. Yes, it is about 0.700 V at some nominal current and temperature, but yes, the Forward voltage drop will increase as the current increases, and as the temperature increases. (And the nominal

0.700 V is for silicon junctions. Germanium ones are down around 0.150 V, but you are not likely to find germanium diodes in use these days. Various other semiconductor materials (e.g. GaAs) will each have their own nominal forward voltage drop

Now -- when you stack a bunch of diodes in series -- you can't just grab any old handful from a bin and stick them in series. They need to be individually tested and matched for junction capacitance and for leakage at various reverse voltages so when there is a high voltage (steady state or a spike) the voltage is evenly spread across the diodes. Back in the old days, such things often had a set of resistors and capacitors in parallel with each diode, to swamp the individual leakage resistance and junction capacitance variations. Sometimes, these assemblies were potted to make them look like (and fit in the sockets for) high voltage rectifier tubes. Four pins on one end (only one of which matters here, since there is no filament to heat up) and a cylindrical cap on the other end for the high voltage connection.

Enjoy, DoN.

Reply to
DoN. Nichols

(...)

Yup. Forward voltage is largely independent of Peak Reverse Voltage. Here is a part available in PIVs from

50 V to 1 KV:
formatting link
Notice that Vf at 1A is the same for all flavors at 1.1 V in 'non-pulse' service, largely due to the material spec of each side of the junction.

Like all good rules, this one has an exception, for 'rectifiers' made up of series diode elements to withstand very high peak inverse voltages will have one diode drop for each diode in the string. This device for example:

formatting link
..has 22 diodes in series to provide rectification at 220 KV. We can expect forward voltage to be upwards of

17-24 V at maximum current for this device because we have to saturate all 22 junctions before the device turns on.

--Winston

Reply to
Winston

formatting link
down to the bottom to see the large jump in complexity when you consider a diode's actual properties instead of simply calling the drop 0.7V. It doesn't mention that the positive and negative scales on the diode curve graph are very different to compress the image. +Y could be Amps while -Y is microAmps.

Look at Figure 3:

formatting link
drop is about 0.7V between 0.5A and 1A, but between 1 and 2V at normal operating currents. I used to build the type of equipment that made those measurements on the production line. The note in the bottom of Fig.3 shows how they were done without heating the diode.

jsw

Reply to
Jim Wilkins

No it is not. Which is why it is not "smart" to massively over-spec the voltage of a diode.

Reply to
clare

I should have said "No, it is not, necessarilly"

Reply to
clare

There is a lot of miss understanding on diodes.

The very high voltage diodes are lower quality in many respects than lower ones. The Peak Inverse volage of 1000 means a 1000v reverse voltage can be applied without a lot of current flowing.

The leakage in both directions is higher than say a 100v model. The 100v one would vaporize on high voltage so each have their job.

If you are running motors in 240 or 375v range, the 1000v would be ok. Magnetic loads can create double voltage spikes.

Mart> >>> Bill Noble wrote:

Reply to
Martin Eastburn

The forward voltage of a diode depends on quite a few variables including geometry, doping levels, current and temperature. If Vf is important to a design, it's always best to consult the data sheet to see what Vf will be with a particular part at the anticipated current level and temperature.

Those packaged 25 to 35 amp bridges tend to have Vf of about 1.1 volt per diode at rated current, and this is essentially independent of the peak inverse voltage rating. Vf of 0.7 volts is seen at lower current, perhaps a couple of amps. Vf varies about logarithmically with If.

Reply to
Don Foreman

"Misunderstanding" is one word.

And the convention on USENET is bottom-posting.

Rich Grise, Self-Appointed Chief, Internet Grammar Police ;-)

Reply to
Rich Grise

FWIW, I concur with Don here, but I've only been an electronics tech (and sometime de facto "engineer") for about forty years now. ;-)

Cheers! Rich

Reply to
Rich Grise

In terms of checking junctions while troubleshooting, a typical DMM diode test function (go/no go check only), the indicated voltage drop will basically be the same for most common rectifiers. Different brands or models of DMMs will likely have different readings.

Rectifiers of all types have various working parameters, one of which is leakage (not the kind you see visually) which is internal current leakage. Most rectifiers fail by shorting, some will open, and some exhibit leakage. Acceptable leakage limits are generally stated in uA micro-amps. Measuring leakage is usually performed with a leakage tester or a test circuit which will apply the rated voltage determined by the manufacturer's specs (or the working voltage of the intended circuit).

Checking high voltage rectifiers with a DMM will very likely result in open junction test readings because the voltage available at the DMM test leads is too low (many meters will only have about 2 volts at the test leads).

Checking zener diodes with a DMM may result in slightly different junction voltage drop readings than ordinary rectifiers. Many of the DMMs I've used over the years have had this characteristic, and a slightly different reading (of about 0.1V) is generally nothing to be concerned about.

Other types of diodes will will exhibit somewhat odd-looking DMM readings, so knowing what specific type of diode is necessary to knowing how to interpret the DMM readings (or lack of).

Reply to
Wild_Bill

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.