Humax 9150 & 9300 freezing etc - Caps replace

Beavor-Bob

New Member
As they age, these boxes are showing various fault symptoms - provided you can work safely & solder correctly, these are easily fixable.
On the power supply board, there are 3 electrolytic notorious "Capxon" capacitors that fail with time. Bit surprised Humax used these rubbish caps!
C21 1000uf 25v
C12 & 14, both 2200uf 16v or 25v
These are cheaply obtainable from Ebay, normally the 105deg temp rating type will be supplied. Try to keep the one nearest the heat-sink, as far away from it as possible.
The PSU, which is probably used in other Humax boxes is type: PW718s
The large smoothing electrolytic & smaller ones rarely fail - it seems low voltage 680uf - 2200uf Capxon ones fail in many pieces of equipment.

Once the caps are replaced, problems like
A. reluctance to change channels
B. slow population of EPG
C. smearing of movement within recordings, all vanish.
Very probably caused by reduced voltage or hash as the Capacitors age.
Other than the fact these boxes are not HD, they are quite good performers. Though why they hid some of the remote functions under a cover, beats me!
The 9150 usually needs a factory reset to wipe the old channel storage away & other errors, following Caps replacement. strangely the 9300 works with just a rescan!
 
Last edited:
The 9150 usually needs a factory reset to wipe the old channel storage away, strangely the 9300 works with just a rescan!
Earlier versions of both the PVR-9150T and PVR-9300T software did not wipe the old channel storage away during a factory reset rescan.
The latest version for both is available from Humax though my preference is for the latest but one (1.00.23).
http://beta.humaxonline.co.uk/downloads
 
Last edited:
Earlier versions of both the PVR-9150T and PVR-9300T software did not wipe the old channel storage away during a factory reset.
The latest version for both is available from Humax though my prference is for the latest but one (1.00.23).
http://beta.humaxonline.co.uk/downloads
Did you mean Automatic Search rather than factory reset (I assume you mean Default Setting). A Default Setting has always cleared the previously stored channels on all these early Humax boxes whereas an Automatic Search has only cleared previously stored channels with later versions of software, introduced in version 1.00.23 I think. All previously stored channels can also be cleared using the Edit Channels section of the menu.
 
A Default Setting has always cleared the previously stored channels on all these early Humax boxes whereas an Automatic Search has only cleared previously stored channels with later versions of software,
Is that not the same thing?
 
Twas quite interesting that this article has coincided with my 9150T exhibitting the aforesaid symptoms! I ordered (ebay) the recommended caps but without taking into account the dimensions and prior to taking the lid of the machine. Consequently the 2200uf are too fat (16mm) and would necessitate subjecting them to yoga. I can source (RS) 16v 20mm h x 10mm w or 25v 25mm x 10mm w. Which would be recommended - and why - are both 16v and 25v acceptable?

PS The existing caps are not showing obvious external signs of failure.
 
Last edited:
What is the voltage rating printed on the existing capacitors? Without knowing the ins and outs of the engineering design choices, it is safest to meet or exceed the voltage rating of the existing component - or, if that is not possible, the higher the voltage rating the better.

As the capacitors in question are probably in the PSU circuit, it is not a good idea to make guesses whether 16V would be enough.
 
For a given capacitance and type, a capacitor with a higher voltage rating will usually be physically larger. The amount of space available will be the constraint on voltage rating.
 
Thanks for the replies.

Yes there seems to be some correlation between capacitance, voltage, size, life expectancy and price (just look at RS filtered to 2200uf and 16v ot 25v)!

Within that I've narrowed my choice to 16v 0r 25v both 20h x 10w which dimensionally about matches what's fitted (actually 16v). My query mainly arose because of the "or" in OP.

However, not being an electronics expert, also wondered about the 2000 hr life expectancy. Is that accmulsating the whole time the machine is on (excluding standby) or just the accumulated microsecs involved in charge/discharge (eg epg usage, channel changes etc)?
 
I don't know where that life expectancy figure comes from. It can only have a bearing under specified stress conditions (which are probably irrelevant to your application), and I see no reason to take any notice of it.

My query mainly arose because of the "or" in OP.
I see now. In terms of capacitor voltage ratings, higher is better - but in this case 16V appears to be good enough.
 
What voltage rating are the original capacitors? If they are 16v then they are probably part of a 12v supply or less so 16v is fine, don't put in a lower voltage than the originals. They are also temperature rated, 85°C and 105°C from memory, go for the higher temperature in a power supply situation and again note what the originals are. Going for the higher temperature rating will not be a problem anyway. Lifetime will have nothing to do with epg usage, channel changes etc but will be a lot longer than that specified in the spec, I think that figure will be an absolute minimum running fully stressed. Usually these big electrolytic capacitors have a 'weak link', like a score line in the top of the case. Failing capacitors of this type may show bulging on the top and can break open the 'weak link' oozing ..... stuff. :eek:
 
I don't know where that life expectancy figure comes from. It can only have a bearing under specified stress conditions (which are probably irrelevant to your application), and I see no reason to take any notice of it.

Wet electrolytic capacitors are well known for having rapid loss of life at higher temperatures, which is why using 105C rated caps is recommended.

IIRC the decrease in life is something like 10% for every 10C above ambient, or some temperature mentioned in the data sheet for the item involved. I know it is quite significant.
 
I have replaced all the electrolytic capacitors including the biggest high voltage cap on the power supply board. So far the 9150 has been working fine for a few days. I bought the caps from a private seller in taobao website when I was in China. The caps are well known brand with rating of 105 degC.

In 2012 I replaced the caps of my other Humax 9200, I bought the unbranded caps from the local Maplin store. After cap replacement, the 9200 only worked for 1 year. The freezing symptom has returned. I may buy some better caps and change the caps on the power supply board again, even though the 9200 worthes nothing now. I just want to keep the 9200 as a spare.
 
In 2012 I replaced the caps of my other Humax 9200, I bought the unbranded caps from the local Maplin store. After cap replacement, the 9200 only worked for 1 year. The freezing symptom has returned. I may buy some better caps and change the caps on the power supply board again, even though the 9200 worthes nothing now. I just want to keep the 9200 as a spare.
In 2012 the number of programme events broadcast from a 5 SD mux freeview transmitter exceeded the number that the 9200T can remember at one time. The strain on the 9200T only kicked in after a retune when it became aware of any extra channels on the mux since the last retune. It will try and store the programme events in its storage even if you deleted those channels at retune or after retune, overwriting some events which it has already stored. As there will be still gaps in the epg it is forever trying to populate the epg. If a manual retune is actioned without one of the larger muxes (SDN Arq A or Arq B) then the freezing normally stops.

Have you tried tuning to less multiplexes to check whether that is the trigger for the freezing on your 9200T?
 
The capacitors of the 9200T power supply board were changed in 2012. My relative did not use it until 2016. The 9200T was alright until a few months ago, the picture started to freeze. She had to turn the 9200T on using the switch at the back in order to get it working again.

Do you think it is worth changing the caps on the power supply board of the 9200T again?
 
Changing the caps (Electrolytic 105deg from RS) made no difference to the symptoms (as decribed in OP) on mine.

My feeling is that it could be the hdd - either moving to failure or because of fragmentation issues - but don't want to loose the backlog of recordings by reformatting at this stage and 9150T seems to be the only model that encyption cannot be overcome by copying. I realise that the hdd is formatted as ext2 which shouldn't suffer fragmentation issues (unless approaching capacity) in linux but if Humax use their own propriety o/s with areas reserved for epg and recording info then this may not be the case. Anyway, I've a spare hdd that can be used to test the hypothesis....
 
My feeling is that it could be the hdd - either moving to failure or because of fragmentation issues - but don't want to loose the backlog of recordings by reformatting at this stage and 9150T seems to be the only model that encyption cannot be overcome by copying.
Content on the 9150 is not encrypted; you can download the content to a PC using humaxrw and play it if you want to satisfy yourself of the accuracy of my statement.
I realise that the hdd is formatted as ext2 which shouldn't suffer fragmentation issues (unless approaching capacity) in linux but if Humax use their own propriety o/s with areas reserved for epg and recording info then this may not be the case.
I have no idea why you think the file system is EXT2; it is a proprietary FAT based file system unique to Humax. The EPG cache is on the hard drive but is a single file. The recording schedule is held in NVRAM on the motherboard.
 
Back
Top