Suhr Reactive Load  [schematic]

Tube or solid-state, this section goes to eleven!
User avatar
AndrewGallios
Information
Posts: 33
Joined: 20 Oct 2020, 07:27
Has thanked: 6 times
Been thanked: 1 time

Post by AndrewGallios »

floris wrote: 09 Apr 2021, 14:02 Btw, is that electrolitic capacitor you bought a bipolar one?
no it is polarised...oops

User avatar
AndrewGallios
Information
Posts: 33
Joined: 20 Oct 2020, 07:27
Has thanked: 6 times
Been thanked: 1 time

Post by AndrewGallios »

thank you Darkwood

User avatar
AndrewGallios
Information
Posts: 33
Joined: 20 Oct 2020, 07:27
Has thanked: 6 times
Been thanked: 1 time

Post by AndrewGallios »


User avatar
floris
Cap Cooler
Information
Posts: 462
Joined: 14 Oct 2007, 12:05
Has thanked: 207 times
Been thanked: 59 times

Post by floris »

That will work but imo is overkill and way too expensive. MKP film caps are very expensive for large capacitance values. You need a Bipolar Non-Polarized electrolitic capacitor which will be much cheaper.

I guess two of these in parallel will be good enough (don't know about the ripple current spec of these):
http://www.banzaimusic.com/100uF-100V-A ... polar.html

User avatar
floris
Cap Cooler
Information
Posts: 462
Joined: 14 Oct 2007, 12:05
Has thanked: 207 times
Been thanked: 59 times

Post by floris »

The full signal out of a distorting amp will flow through the cap at frequencies above the low resonance peak, so that can be quite a current, for a 100 W amp with a 8 Ohm speaker load this can be around 3,5 A.

Is there anybody who checked the ripple current or ESR specs of the bipolar electrolitic capacitor(s) that they used? Or is any bipolar cap with enough voltage spec good enough?

I worry about this because if the ripple current is too high, the cap will become too hot and that leads to breakdown of the cap, perhaps leading to a malfunctioning load box and a defect output transformer. When the cap shorts, the low resonance part no longer works, but the load box keeps functioning. If the cap malfunctions as an open connection, the resistance to higher frequencies of the load box increases and that might lead problems in the amp.
Perhaps I'm overthinking this...

User avatar
bmxguitarsbmx
Resistor Ronker
Information
Posts: 296
Joined: 25 Mar 2010, 21:15
Has thanked: 364 times
Been thanked: 113 times

Post by bmxguitarsbmx »

I think those are some good points Floris.

User avatar
AndrewGallios
Information
Posts: 33
Joined: 20 Oct 2020, 07:27
Has thanked: 6 times
Been thanked: 1 time

Post by AndrewGallios »


User avatar
floris
Cap Cooler
Information
Posts: 462
Joined: 14 Oct 2007, 12:05
Has thanked: 207 times
Been thanked: 59 times

Post by floris »

These 100VDC F&T bipolar caps have a 35VAC spec. A 100W amp on a 8 Ohm load yields 28VAC. Should be ok I guess.

User avatar
floris
Cap Cooler
Information
Posts: 462
Joined: 14 Oct 2007, 12:05
Has thanked: 207 times
Been thanked: 59 times

Post by floris »

But on a 16 Ohm load, that same 100W amp yields 40VAC. That will not work with those caps.

User avatar
AndrewGallios
Information
Posts: 33
Joined: 20 Oct 2020, 07:27
Has thanked: 6 times
Been thanked: 1 time

Post by AndrewGallios »

im on 8ohm reactive load build so i will never conect the 16 ohm tap on the amp

only the 8ohm tap
maybe the 4ohm tap

User avatar
AndrewGallios
Information
Posts: 33
Joined: 20 Oct 2020, 07:27
Has thanked: 6 times
Been thanked: 1 time

Post by AndrewGallios »

!

User avatar
AndrewGallios
Information
Posts: 33
Joined: 20 Oct 2020, 07:27
Has thanked: 6 times
Been thanked: 1 time

Post by AndrewGallios »

ready! its killing!!
A549F142-BD6B-4DDB-8116-68F08CF77D08.jpeg
DB9B0DB6-70AC-48F0-A841-B000B43CF466.jpeg
48550994-6B51-4381-8DFC-C915F6A818E4.jpeg

User avatar
floris
Cap Cooler
Information
Posts: 462
Joined: 14 Oct 2007, 12:05
Has thanked: 207 times
Been thanked: 59 times

Post by floris »

Wow that looks great, good work!

User avatar
Optical
Breadboard Brother
Information
Posts: 124
Joined: 28 Jul 2016, 22:44
Location: Auckland
Has thanked: 15 times
Been thanked: 14 times

Post by Optical »

On mine the part that gets the hottest is the air-core inductor. - with a 100W tube amp
I'm using the nichicon caps that someone mentioned here earlier. They get a little warm but nothing to worry about.

Image

User avatar
bmxguitarsbmx
Resistor Ronker
Information
Posts: 296
Joined: 25 Mar 2010, 21:15
Has thanked: 364 times
Been thanked: 113 times

Post by bmxguitarsbmx »

Excessive heating of your air coil is likely from eddy currents in the chassis. Try turning the coil 90degrees. As is, the currents are making loops out of the plane of the chassis, at 90 degrees the currents will make loops that are limited in size by the thickness of the chassis. Big difference! The losses may be so huge that your top end inductive impedance is nuked. Eddy losses are like putting a resistor* in parallel with the coil. There is a whole lot of interesting stuff here...

User avatar
Optical
Breadboard Brother
Information
Posts: 124
Joined: 28 Jul 2016, 22:44
Location: Auckland
Has thanked: 15 times
Been thanked: 14 times

Post by Optical »

Interesting, I'll try that.
I should mention it only gets warm when the amp is absolutely cranking. But I'll see if there is any audible effect from the position!

User avatar
bmxguitarsbmx
Resistor Ronker
Information
Posts: 296
Joined: 25 Mar 2010, 21:15
Has thanked: 364 times
Been thanked: 113 times

Post by bmxguitarsbmx »

See if the heat is actually coming from the chassis. Thats where the eddy currents are. And it just seems like the coil is hot.

I had the idea that I could "tune" the top end rise by threading a bolt into a plastic core inside my air coil. For a few millimeters it works, then eddy losses take over and the top end impedance falls to the dc resistance. (eddy losses being modeled as a parallel resistor) Then you can try different alloys of bolts and you can take a slice down the side of the bolt to cut the current loops into smaller sections. Because the magnetic lines run perpendicular to the eddy currents, a slice down the length of the bolt only cuts the path of currents and leaves your permeable mass largely untouched. In the cases of inserting the bolt, it is the bolt that gets hot as that is where the currents are. Thats why I am guessing your chassis is the thing that is hot. Put that in your pipe and smoke it! It was a long time ago I did these experiments so, I'm curious to how your experiments go!

Post Reply