C7042 The Power Thread

Here are the numbers on the 4 Car Powerbase, both the original and the improved version with the cooling vents.



As you can see they are all very similar because the are esentially the same, with minor tweeks. The new 4 Car is however more robust because of the addition of heatsinks.

Rich

Interesting to see the results for the 4 car powerbase as that is what I will be upgrading from. Now I really can't wait for the boat from China to arrive!

Thanks Rich!

I will take a proper look round what's in the Powerbase when I do the Look Inside thread, for the moment let's concentrate on the Power path and where the volts go.



So this is what it looks like. You can see two sets of 4 Mosfets, there being an H Bridge for each lane in Analogue. The relay is used to connect the two pairs together in digital mode, so we are paralleling up the Power stages, which is one of the reasons for the good performance.

Just above you can see the chokes. Four single wound ones in each of the feeds to the track and two bifilar ones between them to get rid of common mode noise. These will have been necessary to meet the EMC regulations.

So now going back to the numbers we see that at 5 Amps we are losing 12.95 - 11.4 Volts, so about 1.6 Volts.

About 1/4 of this, 0.4V is lost in the output chokes and the wiring to the track. Now looking at deadpedal's question, replacing them with wire links and beefing up the track wiring would increase the output voltage at 5 Amps by 0.4 Volt, a usefull gain, the current that could be drawn before overload would however not change.

So where do the rest of the volts go? Well I have measured another 1/4, due to variances in the input voltage, in this case not due to the Power supplies but due to the internal regulators that sit under the heatsink and fan. So replacing the regulators & probably the input resettable fuses with links could give us another 0.4 Volts at 5 Amps. Also at this point we would be able to add a variable aftermarket supply.

The final 0.8 Volts I have not yet measured but will almost certaily be in the RDS on of the mosfets, and is actually a good number.

More detail to follow later when I have a production Powerbase to play with and can try the changes in practice.

Rich

YES, I vote for the add-on of fuel management. PLG has sense to me only as Pit-Pro.
Happy to see the new PB is reaallllyyy good!
Credit to Scaley and Pro guys

Cheers,
JamieG

QUOTE (RichG @ 9 Aug 2010, 02:13) <{POST_SNAPBACK}>So straight in with a table of numbers and then some explanation.


...
You can see how well C7042 with 2 PSU's compares to S-H At 5 Amps S-H has dropped 0.43 Volt which is 3% whereas C7042 has dropped 1.55 Volts which is 12% S-H will of course go on to 20 Amps plus.

Thanks for the starting point.... A few thoughts:

- when comparing S-H and C7042, one must note on the table that the 'starting' (no-load) voltage between the two is not the same. The C7042 starts at around 13v, whereas the S-H starts at around 12.4v If both PSUs were the same higher voltage, then the output from S-H at 5A would be .6v lost to get to something more like 12.9v, while C7042 has lost 1.5v to 11.4
- the RDS of the mosfets seems high - since in digital they are in parallel (when using 2 PSUs) they calculate out as 0.32 ohms split between both top and bottom of each bridge. Many P & N matched TO220 mosfets total less than 0.2 ohms, which makes me wonder about the drive here
- the mosfets are required to dissipate some 4 watts which is quite low for a TO220 device

What are the output part numbers, and how are they driven - is that the 16pin device to their left?

Ian

I appreciate that the starting voltages are different, it's just difficult to contrive them to all be the same, particularily when they are done with different PSU's and at different times. To a first approximation the best thing to do is to look at the delta change in voltage at a common current, and then express as a % change, which is what I have done.

In this case 5 Amps. C7042 sees a 12% reduction, and S-H sees a 3% reduction. I think this is a fair comparison.

To then make it even fairer if you were to subtract the volt drop in the output chokes from C7042, as they are removed in the S-H upgrade, then the result for C7042 is 9%

I was not saying that the remaining 0.8 Volt was entirely due to RDS ON, just that that was the result of the non accounted for voltage, RDS ON will of course be a chunk of that, however there is likely to be something else I have missed.

The mosfets are FQP20N06L 60V 20A

RDS On is quoted at 0.06 Ohm which at 5 amps should be 0.3V * 2 / 2 for the parallel stages = 0.3V, so yes you are right, much less than the 0.8 V unnacounted for. However bear in mind my measurements made a bit quickly and I was concentrating on the losses after the Mosfets, and due to the PSU regulation.

Yes the driver chips are the 16 pin devices, much better than the direct drive in C7030. They are ISL83202

More work to be done.

Rich

These cables found at the link below are working very well with my aftermarket power supply!
http://www.mcmelectronics.com/product/23-7...ELAID=220275341


I checked the polarity before giving it a go. The red banana plug end matches the red on my Pyramid PX-26KS.


If you happen to not already have any of the Scaley wall warts, aftermarket PSU works great! (and who would pay $120 for wall warts?)
I will combine two cables to hit both outlets shortly... after I stop playing

After combining two cables I've got more amps than I know what to do with (for now). Running with several pace cars is very fun and the wife and I ran some analog together for quite a while (well... for as long as I could keep her interested )