Quick questions about diodes

[ThreeChordWonder]

My 1966 onwards* Mojotone JTM45 kit and my Mojotone DR103 kit both use a single 1N4007 diode in the power tube bias supply.

The rectifier diodes on the DR103 are 1N5408s.

My next project, a Mojotone Super Lead 100 also uses 1N4007s in the rectifier.

It's my understanding, correct me if I'm wrong, that the 1N5406 van handle more current than the 1N4007. 3 amps vs 1 amp. There's a slightly higher forward voltage drop, 1.0 to 1.5 volts vs 0.6 to 1.0 volts.

So the questions are simply:

1. would 1N5406 diodes be better in the rectifier section?
2. Can I use 1N5406s in the bias feeds in place of 1N4007s?

* The original JTM45 used KT66s IIRC, but post 1966 Marshall switched to EL34s, so the Mojotone is a copy of the post switch version. I expect Marshall made the switch because the EL34s were cheaper and or easier to find in the 1960s UK.

 

[PelliX]

Well, there are a few differences if you look at the specsheets:

https://www.mouser.com/datasheet/2/149/1N5408-888344.pdf

https://www.mouser.com/ds/2/149/1n4007-888322.pdf

1. would 1N5406 diodes be better in the rectifier section?

You wouldn't "need" the higher current rating, but it would be advisable to spec higher, indeed. Simply put.

2. Can I use 1N5406s in the bias feeds in place of 1N4007s?

Yes, but given the very limited draw here it doesn't make much sense. Again, simply put.

A couple of suggestions/notes; the JTM45 typically has a valve rectifier. Are you referring to the solid state rectifier some offer as a switchable option, or protection diodes for in case the rectifier valve fails in a bad way?

Another suggestion for the (solid state) rectifier would be a UF (or UltraFast) switching diode. Less noise, slightly better efficiency, etc:

https://www.vishay.com/docs/88756/uf5400.pdf

Take a look at for example the UF5408 if you want something beefy.

 

[ThreeChordWonder]

<snip>

A couple of suggestions/notes; the JTM45 typically has a valve rectifier. Are you referring to the solid state rectifier some offer as a switchable option, or protection diodes for in case the rectifier valve fails in a bad way?

<snip>
I'm sticking with the tube rectifier on the JTM45 unless and until it gives me problems. A touch of authenticity and all that.

The question was about the Super Lead 100 kit, which uses a diode rectifier, which I will be starting shortly.

Take a look at for example the UF5408 if you want something beefy.

Thanks. Just ordered some.

 

[Jon Snell]

1N4007 is a 1000volt PIV device. 1.0volt forward drop. 1A
1N5406 is a only 600volt PIV. 1.2volt forward drop. 3A
1N5408 however is 1000volt. 1.2volt forward drop. 3A

 

[ThreeChordWonder]

1N4007 is a 1000volt PIV device. 1.0volt forward drop. 1A
1N5406 is a only 600volt PIV. 1.2volt forward drop. 3A
1N5408 however is 1000volt. 1.2volt forward drop. 3A

Thanks.

I presume the 5406s in the Hiwatt copy are okay, as the DC HT is about 470 V or lower.

The 4007s in the Marshall 100 watt should be okay too, but I'm not averse to spending a few extra pennies up-rating stuff. I replace the 0.5 watt resistors with 1 watt ones, the 2s with 5s, etc., for example. So it makes sense to do it with the diodes as well.

 

[Jon Snell]

There is absolutely no benefit in using a 3Amp diode for a 250mA supply.

 

[Pete Farrington]

I'm not averse to spending a few extra pennies up-rating stuff. I replace the 0.5 watt resistors with 1 watt ones, the 2s with 5s, etc., for example. So it makes sense to do it with the diodes as well.

This approach can end up making a transformer winding the weakest link in the chain.

 

[ThreeChordWonder]

This approach can end up making a transformer winding the weakest link in the chain.

Except for the fuses.

 

[arthur.lowery]

Thanks.

I presume the 5406s in the Hiwatt copy are okay, as the DC HT is about 470 V or lower.

The 4007s in the Marshall 100 watt should be okay too, but I'm not averse to spending a few extra pennies up-rating stuff. I replace the 0.5 watt resistors with 1 watt ones, the 2s with 5s, etc., for example. So it makes sense to do it with the diodes as well.

The diode has to take more than just the HT voltage when off (reverse biased) - indeed double the HT voltage (because the cathode is at HT, and the anode at -ve HT which is where the transformer swings to, thus 1000 V is necessary.

 

[2L man]

I believe using "ultra fast" rectifier diodes is worth the money because there can come less double the mains frequency "switching Off-noise spikes"

Rule of thumb is that four diodes are used for both Bridge and Center tab PS. Functioning bridge need four and CT supply make "the reverse peak voltage" double!

 

[PelliX]

I believe using "ultra fast" rectifier diodes is worth the money because there can come less double the mains frequency "spike-noise"

And often less heat dissipation/energy loss, too.

 

[william vogel]

Always use 1000 volt diodes just for KISS. The 5408 diodes are fine except the lead wire is so thick that it prevents being able to feed two inside turret holes and it’s difficult to wrap them around turrets. I’ve used them before and stopped because the lead wire thickness. I buy diodes in bulk and use uf4007 but never heard the difference between them and 1n types.

 

[ThreeChordWonder]

The thickness of the UF5408 wires is why I stepped away from the arrangement shown on the Mojotone pictorial. That has back-to-back diode pairs connected to central turrets and the HT DC comes off those. That's fine for 4007s with thinner wires, but not the 5408s.

They must use that turret board for a different model also, or thought far enough ahead, because there are (were in my case) two extra / unused pairs on that end of the turret board.

So, after wasting quite a few UF5408s, I removed the two central turrets and soldered in the diodes across individual turret pairs, four diodes, four pairs of turrets. Then I wrapped 18 gauge solid around the turrets as needed to make the necessary jumpers.

On a related note, anyone building a Hiwatt copy may find that modern diode leads aren't long enough to span the gap between the turret  strips used. I solved that by simply moving one strip half an inch closer to the other, drilling extra holes in the chassis to suit.