goldenreference high voltage power supply (GRHV)

[johnwmclean]

Another candidate for C2M1000170D replacement could be G3R450MT17D.

[jamesmking]

On 5/18/2021 at 10:52 PM, johnwmclean said:

 

 

The pin out is NOT the same...

G3R450MT17D is G S D looking from front left to right

C2M1000170D is G D S looking from front left to right.

don't worry the C2M1000170D will be back in stock in a few months... maybe.... 

Update: well I though I had derped but it turns out there are at least two versions of the G3R450MT17D datasheet and they don't even agree on the printout or current handling...

 

Edited June 5, 2021 by jamesmking

[iemphile]

The Cree's I ordered from digikey in January arrived a few weeks ago, site still shows 0 in stock though. Maybe people preordered the entire batch?

[Pars]

On 5/11/2021 at 1:54 PM, Kerry said:

They are the Phoenix Mate-N-Lok-2 connectors.  They are rated at 600V which is nice. 

 

On 5/13/2021 at 2:06 PM, starcat said:

Also 5.08 pitch which makes it possible to use them instead the usual phoenix block type connectors. Thanks, Kerry.

Any part numbers for these? Any search I do on Mouser brings up a TE Mate-N-Lok, and isn't 5.08mm pitch..

 

[starcat]

I was looking into and not all are 5.08. The ones with more pins and double row aren't. Maybe @Kerry can sched some light.

In the mean time I was looking at the Molex SPOX. All 5.08, directional, various pins available as well. Can be used wire-to-board as direct replacement for the terminal blocks on Kevin's PCBs:

Wire connectors

Board connectors

 

 

Edited May 19, 2021 by starcat

[johnwmclean]

I think SPOX are limited to 250V, in saying that they are great connectors from my experience. 

[Kerry]

Here are the part numbers for the 8 pin connectors and pins...

1-794065-0

770579-1

770988-1

[starcat]

8 minutes ago, Kerry said:

Here are the part numbers for the 8 pin connectors and pins...

1-794065-0

770579-1

770988-1

Thanks, Kerry. Looks like the pitch is 4.14mm on these.

20 minutes ago, johnwmclean said:

I think SPOX are limited to 250V, in saying that they are great connectors from my experience. 

Aren't the phoenix terminal blocks as well 250V? 

[johnwmclean]

They vary in voltage depending on which series, I have used these 651-1730159, it’s has a 400V rating.

[Pars]

Thanks guys. Yes, I use the 5.08mm SPOX connectors on my own (lower voltage) builds, as I detest the screw down connectors for my own use.

I'll need to take a closer look at these, but Molex makes a Micro-Fit 3.0 series, which is 3mm pitch, and rated for 600V apparently. 3 pin PN: 43650-0315. Might be something to plan for on PCB designs if they actually work out to their ratings. I have some of them for some Salas shunts for a UGS preamp project.

[starcat]

10 hours ago, johnwmclean said:

They vary in voltage depending on which series, I have used these 651-1730159, it’s has a 400V rating.

I use the same. The specs show different ratings 250, 400, 630V. For the small currents in the stat amps should be more than fine I guess. 

 

[waltzingbear]

you might look at Souriau Quik lock connectors (SMS series) for HV application, haave used them in equipment for 40 years, great stuff. Not all the same configurations as Molex. 5.08 pitch and excellent pins.

 

Performance characteristics

Operating temperature: -55°C to +125°C

Insulation resistance: 5000 MΩmin.

Test potential: 2000 VAC

Current rating: 5 AMP

Durability: 500 matings and unmatings.

 

Cheers

Alan

[starcat]

Thanks, very nice.

Posting here the catalog link to the Souriau SMS series quick lock connector and pdf catalogue as well.

 

Edited May 22, 2021 by starcat

[johnwmclean]

On 5/19/2021 at 8:04 AM, jamesmking said:

The pin out is NOT the same...

G3R450MT17D is G S D looking from front left to right

C2M1000170D is G D S looking from front left to right.

don't worry the C2M1000170D will be back in stock in a few months... maybe.... 

I had a another look from the datasheets they are both G D S viewed from front left to right.

[starcat]

John, have you already tried out the G3R450MT17D? 

[johnwmclean]

On 5/31/2021 at 10:09 PM, starcat said:

John, have you already tried out the G3R450MT17D? 

I haven’t.

However I did pass this by Kevin and he said the G3R450MT17D part should work, but he has not used it.

DigiKey still have good stock, I secured a few pcs for myself just in case.

[chinsettawong]

I use SCT3080KLGC11 in my Grounded Grid power supply and it works well.

 

[starcat]

The SCT3080KLGC11 is three times the price of the G3R450MT17D.

Is it that much "better"?

Edited June 2, 2021 by starcat

[chinsettawong]

When I went and bought, I didn’t check the differences in price and quality.  It’s what’s available in my area and I needed it quickly.  

[jamesmking]

On 5/31/2021 at 12:45 PM, johnwmclean said:

 

 

I thought I was going insane. Here's the data sheet from mouser: source and drain are reversed compared to your datasheet for the same product also the current ratings are different WTF GeneSic..... 

https://www.mouser.co.uk/datasheet/2/169/G3R450MT17D-1919148.pdf

Edited June 2, 2021 by jamesmking

[Pars]

Hmm, the datasheet right off of Genesic's site shows GDS: https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&ved=2ahUKEwi5usK-kvnwAhXKZs0KHYmTCvMQFjABegQIAxAD&url=https%3A%2F%2Fwww.genesicsemi.com%2Fsic-mosfet%2FG3R450MT17D%2FG3R450MT17D.pdf&usg=AOvVaw1sVqBXJV-AnV_PZztKUbPF

Now this is confidence inspiring!

Newark's is GDS

Arrow's is GSD

Edited June 2, 2021 by Pars

[Beefy]

If you check the revision number and file properties, the datasheet from Mouser is from September 2020, and the one from GeneSiC direct is May 2021. It also seems that in the new datasheet, current has been de-rated.

 

Whether it was a mistake on the datasheet, or a change in the part...? Either way, I certainly wouldn't put it in a board without testing it first!

 

[EDIT] ninja'ed by jamesmking's edit!

Edited June 2, 2021 by Beefy

[johnwmclean]

On 6/3/2021 at 12:16 AM, jamesmking said:

I thought I was going insane.

Teehee...
I was kind of thinking this was completely out of character for James.

Meanwhile I bought some C2M1000170Ds from eBay and kind of resigned them to be fakes...

They’re looking a bit beat up from being jumbled around, but I think the gamble paid off.

Edited June 4, 2021 by johnwmclean

[starcat]

John, how do you tell they are not fakes? Have you tested them and could you describe your test procedure? 

Found here some more on ebay https://www.ebay.de/itm/184068141515

[jamesmking]

On 6/4/2021 at 10:36 AM, starcat said:

John, how do you tell they are not fakes? Have you tested them and could you describe your test procedure? 

Found here some more on ebay https://www.ebay.de/itm/184068141515

starcat THE EBAY TRANSISTORS LOOK FAKE the sellers negative feedback is also full of warnings of fake transistors, not as described, measured and out of spec, arrived not working etc etc. and many many cases of just not shipping anything at all.

My known genuine have:

1.  numbers and letters inside the circular indents. there should be left circular indent moulded words china and a single letter code, right circular indent a three digit code again moulded in my case 1 letter followed by 2 numbers.

2. all text should be moulded not printed or etched

3. the manufacturers logo is different

4. the main writing on the lower half of the body should be inside a indented rectangle.

All of which agrees with the photos in the posting from John just above and is significantly different from the ebay listing you link to.

Production may have changed over the years but I highly doubt the ebay devices are even cree.

 

To attempt to determine if genuine.

least reliable visual inspection of case, markings, <- some what reliable but production markings change over the years and some are easier to fake than others. However some fakers are so lazy simple things like the length of the pins, shape of the case are wrong. But the more practiced fakes will take another device with the same case and legs and rub off any markings and print new markings or laser etch new markings. I don't think many if any fakes would go as far as to remould cases.

better visual inspection and then transistor identifier like a peak dca75. can identify pin out and type but not breakdown voltage which is important for power supplies and valve amps etc. I have received fakes that don't even have the correct pin out or are not even the correct type e.g. npn when they should be pnp etc etc. But again the more experienced and more prolific fakers will at least provide a fake with the correct type and pinout even if the breakdown voltages, wattages etc are very different.

even better  a device capable of testing breakdown voltages like a DY294... can provide breakdown voltages but not characteristic curves but this would not necessarily spot a fake that has similar pin out and breakdown voltages but wildly different characteristic curves.

even better proper high voltage curve tracer and measure against spec sheet curves and or a known genuine parts

P.S. I have reported multiple serial transistor fakes to ebay and ebay do not give a damn. I have tried explaining that not only is it fraud but the voltages involve constitute a potential risk to life and fire hazard... ebay don't give a damn. I went to pay pal... don't give a damn.

The issue is many people don't realise about fakes and if they buy and don't test or use them they will probably leave good feedback about the transaction reinforcing the sellers reputation.... I fell into this trap myself when I first started I just did not know about fakes.

As an experiment I even purchased over a period of more than a year a few transistors of the same type from the same seller  and every time they were fakes. I used ebay returns each time providing evidence from the DCA and DY294 that they where fakes and the seller continued to sell the same fakes year in year out with no problems. I bet they even resold the ones I returned.

Edited June 5, 2021 by jamesmking