TFS7708 high idle temperature. V2
Hi All.
The previous topic on this seems to have been hijacked by spammers ( read morons).
Carrying on here.
Pi Wrench , your comments about the output inductor saturating were spot on. I'm using a Sendust toroid , which has a nice
soft saturation characteristic. As the load went above about 15A , it's inductance obviously became too small.
I stacked 2 cores together and the problem mostly went away. The drain voltage still creaps up a little , but nowhere near
what it was doing. I obviously need to source a slightly bigger core for the output choke.
Cheers
Rob
Comments
I am going to do that , it's just good to know what causes certain issues in these designs so that you know how to "debug" designs that don't work first time.
Thanks for the help.
Cheers
Rob
P.S all the spam messages that were in the previous thread are now gone , so someone got to it. Good stuff!
Hi Pi-Wrench.
I've started with the PIXIs designer spreadsheet and was wondering about a few things.
I've attached the design so far. When I get down to line 171 (Fill factor and losses) I'm starting go get lots of fields that won't populate.(#N/A) Not sure if that means "Not applicable" or "Not available".
Have I forgotten to do something further up the design?
I've also got a Mag inductance of only 10mH and it's asking to go even lower?
Cheers
Rob
| Attachment | Size |
|---|---|
| HiperTFS2_Forward_Design2.pixls | 78 KB |
Please send me a PDF version of your spreadsheet - this forum doesn't like active attachments...
Sorry , here you go
| Attachment | Size |
|---|---|
| HiperTFS2_Forward_Design2.pdf | 1.11 MB |
I ran the spreadsheet for continuous output of 27V/13A. If you are actually running lower continuous power with a peak rating, you should use the continuous rating at the top of the spreadsheet, then fill in the peak rating lower down. Attached is a PDF of the spreadsheet I ran using your parameters.
| Attachment | Size |
|---|---|
| HiperTFS2_cust_27V_13A.pdf | 1.11 MB |
I didn't see any bogus input in your original thread - maybe someone else got to it before me. We detete spam as soon as we find it
Anyway, at the risk of sounding like a broken record, if you run your design parameters using the PI software, you should get a workable design for the output inductor.