I am gathering parts to make a stereo PP 2A3 amp. Running the 2A3s at 60ma each, the amp will require 250ma of B+ current. Hammond has available in 10H, a 200ma (193J) and a 300ma (193M) model. The larger model is twice the size, weight and expense! Can I push 250ma through the smaller choke safely? Can someone please estimate the lose in induction that would result? Thanks.
Always give components some headroom 🙄
It’ll end in tears if you don’t, it’s sloppy to say the least. 🙂
Do you run 2watt resistors at 3watts ?…..
It’ll end in tears if you don’t, it’s sloppy to say the least. 🙂
Do you run 2watt resistors at 3watts ?…..
Member
Joined 2009
Paid Member
depending on the PSU design you may be able to just put a 193J into each channel, that's two total. I use the 193 models when I want to mount above the chassis.
What country? If Canada, a1parts.ca has the best pricing on Hammond. In the USA it's Arrow.
I have a 193Q from my first amp build. I no longer use filter chokes for power amplifiers. I find good filtering for the front end is all that's needed as a couple of volts of ripple on B+ doesn't make it to the speakers due to PP and CMRR.
I have a 193Q from my first amp build. I no longer use filter chokes for power amplifiers. I find good filtering for the front end is all that's needed as a couple of volts of ripple on B+ doesn't make it to the speakers due to PP and CMRR.
Another way is to separate the output section from the input section. One HT for the output and one for the input. I do this because I like to use an AZ1/AZ11 mesh rectifier for the input section. It also takes a bit of the load off the main HT.
Thanks for the replies. Regarding the suggestions:
Two lower H chokes in series is an interesting idea but I wonder if regulation might suffer with the added PS resistance.
Separate chokes for each channel is very possible and would allow the use of the 193G (10H@150ma). But two of these are roughly the same footprint, cost and weight as one 193M, so I am not sure what will be gained.
Pushing 250ma through a 200ma rated choke is only going to cost inductance as far as I can see. If it cost half the induction, then I might as well forget it. But if it cost 10%...
Two lower H chokes in series is an interesting idea but I wonder if regulation might suffer with the added PS resistance.
Separate chokes for each channel is very possible and would allow the use of the 193G (10H@150ma). But two of these are roughly the same footprint, cost and weight as one 193M, so I am not sure what will be gained.
Pushing 250ma through a 200ma rated choke is only going to cost inductance as far as I can see. If it cost half the induction, then I might as well forget it. But if it cost 10%...
Member
Joined 2009
Paid Member
I would hazard a guess that you won't have a big issue pushing the extra 25% current through it but it maybe borderline. Once the core starts to saturate the inductance will drop fast.
As others say, I wouldn't exceed the current rating, but instead leave as much headroom as possible. How about the 5H/300mA 193L? It would reduce the size and weight, and also the series resistance. 5 henries is plenty for a PP amp. You don't need much inductance because of the PS noise cancellation PP affords. Just my two cents.
Silvered mica - how did you derive the 250mA requirement, and is that continuous? I suggest that will be a major factor in assessment, as the damage issue is related to choke winding heating (rather than transient distortion to your ears).
OP said stereo push pull with 60mA idle... That's 240mA already and doesn't include the front end.
Why don’t you lower the H and increase the current rating and just use larger caps? People often go overboard with the chokes inductance with little to no real world gain. Caps are cheap too. You could also split the supply in two after the rectifier one for each channel.
A power supply filter choke has the role of impeding the change of AC, but letting filtered DC through. To work with DC current, it has mostly often a gapped core.
The airgap to some extent and point linearizes the magnetizing curve, but despite of that, after exceeding some amount of flux density, be it DC or AC, the choke core starts saturation, which decreases permeability, which decreases inductance.
The good thing is, the choke does not saturate sharply, it tends to saturate linearly with increasing flux density, due to Bdc = Inductance x Idc / Core area x Turns.
The thing is, when inductance drops due to saturation, flux density also drops linearly, which allows for more DC current to reach the same flux density.
A choke manufacturer should publish a saturation vs Bdc and Bac graph, which gives a bit more of info of the behavior or chokes. For my chokes I do not have such a graph, but publish the maximum allowable Bdc + Bac working ranges in the measure choke linear region. Which is 1T for amorphous and nanocrystalline cores and 1.4T for HiB, slightly gapped.
Lundahl publish something similar in their datasheets.
The airgap to some extent and point linearizes the magnetizing curve, but despite of that, after exceeding some amount of flux density, be it DC or AC, the choke core starts saturation, which decreases permeability, which decreases inductance.
The good thing is, the choke does not saturate sharply, it tends to saturate linearly with increasing flux density, due to Bdc = Inductance x Idc / Core area x Turns.
The thing is, when inductance drops due to saturation, flux density also drops linearly, which allows for more DC current to reach the same flux density.
A choke manufacturer should publish a saturation vs Bdc and Bac graph, which gives a bit more of info of the behavior or chokes. For my chokes I do not have such a graph, but publish the maximum allowable Bdc + Bac working ranges in the measure choke linear region. Which is 1T for amorphous and nanocrystalline cores and 1.4T for HiB, slightly gapped.
Lundahl publish something similar in their datasheets.
Why not use 5H of choke (193L is 5H, 0.3A)? The more current an amp uses the lower the choke value needs to be (and the larger the caps).
I've seen the implementation (including the maths) of a power supply that works far better with two 3H chokes than one 9H choke. The PS was a cap input rather than needing a swinging choke. Natural thing is to split the left and right channels if you're pulling too much current for a single PS.
The math is not that intuitive.
For example, a single 10H 200mA for both channels.
If you use 2 inductors instead, you need 20H 100mA for each channel, in order to get same voltage regulation.
For example, a single 10H 200mA for both channels.
If you use 2 inductors instead, you need 20H 100mA for each channel, in order to get same voltage regulation.
Last edited:
I am no expert, nor do I play one on TV, but given this is a PP amp it seems that the output stage doesn't really need heroic filtering. Why not Run RC up to the output stage and then put the choke after the output stage B+ tap and before the driver and VAS taps?
