70W Amplifier - Using RD16HHF1

[@XPCTECH]

So cool, love your work, keep it up, I appreciate the time you took to share this with everyone.

[@na5y]

Thank you for the kind comment!

[@M0NTVHomebrewing]

Hi! Thanks for the video and well done on the amplifier. I've built the Eamon Skelton (EI9GQ) 16W PA that uses 2 x RD16HHF1. I followed the modifications of VK3PE which uses standard toroids for the transformers. The most I ever squeezed out of mine was 14W on 80m, 12W on 40m and 7.5W on 20m. So it sounds like your results are in the similar ballpark. Skelton also used 500mA bias per MOSFET as you did. Keep up the great work! 73 Nick M0NTV

[@na5y]

Thank you for the kind comment - I'll have to check out Eamon's amp. On a deeper look with my amp using the RD16s I am seeing some nasty parasitics at around 158 Mhz. I might do another version of the board just for the RD16s.

[@vu2abs]

I suggest you try a diplexer type LPF where there is HPF which is connected to a dummy load. The effect of this on any solid state linear amp should be pretty significant and very ineresting experiment!!

[@ohmschool]

Very interesting difference between 2 types of FETs, thanks for sharing! I can’t remember if your output step-up xformer is 1:2 turns (1:4 impedance) or 1:3 turns (1:9Z). In theory, 1:3 should boost output voltage vs 1:2. Could also try boosting to 1:4 turns (1:16Z) however harmonic content is likely to degrade further at this higher ratio.

[@na5y]

Output transformer is 0.5T : 3T. Although based on Rob's comment above 10-11W for the pair is typical for class AB

[@ohmschool]

Thank you so much for the info on output xformer 👍I think I mistakenly assumed you were targeting class C with 0.5A bias, but seeing Robs comment I agree with his suggestion to lower bias closer to 100ma which is the bias I use for some of my LMDOS FETs

[@na5y]

@Rob's Blog Sorry for the confusion - its a single turn with a center tap and I was referring to each side as "0.5 turn"

[@R2AUK]

Thanks for sharing. Looking forward to the next episode 👍

[@na5y]

Thank you Alex. Stay tuned!

[@germanjohn5626]

You can get rid of the iso pad beneath the transistors that will help with cooling.

[@na5y]

Agree with the RD16HHF1 because the metal back of the mosfet is connected to the Source (assuming your Source is connected to ground). You couldn't do that on an IRF510, say, because the body is connected to Drain. Thank you for the comment John!

[@KennethScharf]

1 amp quesant current? That sounds a bit high for class AB1. About 13-14 watts resting power.

[@na5y]

Yeah you are right - I need to do some more playing around with this amp. I'll get back to it one of these days!

[@stevew9945]

Nice video, keep posting

[@na5y]

Thank you Steve - lots of great suggestions coming in on next steps/experiments ;)

[@tobykearns7782]

Hey man, did you ever get to the bottom of your low power issue with these FET's? I've been playing around with a single ended class AB1 amp with an RD16HHF1 and am struggling too with getting anywhere near what the specs says out of it. IIn the past I was in communication with a seriously respected and well known QRP DIY'er and writer that built a 28MHz single ended amp and got 14W out with 250mW in with an RD16.. so like you I'm baffled also. Very strange indeed. I'm starting to think my FETs may be fake so I'm going to setup a simple jig to measure static Ids vs gv for a given Vds and see if the results track the spec sheet. I've messed with my output matching, used broaband xformers , 9:1 and 4:1 and also L-C-C matching networks. Squeezing blood out of a stone.. Keen to know if you ever got any further with this amp. Thanks for the video, great stuff.

[@na5y]

I never did get to the bottom of it and need to revisit. I know my rd16s were genuine because I got them from rfparts.com. My suspicion was that adapter board I had. Certainly most of the rd16s on ebay are fakes with wrong pinouts even

[@Davidsmith218]

Have you tested or tried a irfp260n I feel you having more knowledge than me that you would be able to get some good power out of a cheap mosfet say up to 600 watts + depending on in put voltage

[@na5y]

I actually have a pile of those - but haven't got around to them yet. I agree they look promising. Unfortunately they have a different footprint entirely - so I'll have to do a new board

[@technishn]

Is that a parasitic happening at lowdrive? If you look w/o a LPF you might see something in wideband on the spec an . it may be due to the swapping of leads on the FETs?

[@na5y]

Someone else also suggested that Kevin - I am definitely going to remove the LPF and remeasure. Thank you for the comment!

[@M0XFXUK]

Hi I am really enjoying these videos as I am a bit of a tinkerer myself. Re the RD16's I imagine that you should get better performance within the limits of the devices opposed to the IRF MOSFETs as the RD's were designed specificity for this purpose. I think that you may achieve better performance regarding harmonics and a balancing the waveform during the 2 tone test by biasing the devices separately and same goes for the IRF devices, just an idea. Regarding the power output level I would have thought that you should be able to should be able to achieve at least 25/30W in this configuration. Mitsubitshi used to be very good with example circuit designs for balanced RF power stages however; I am unable to find much in the way of design examples on these devices as everything in the datasheet is single ended. Still may be worth a look.

[@na5y]

Glad you are enjoying the videos John. I didn't have as much success with the RD16 as I thought I might - probably because of that separate adapter board

[@M0XFXUK]

@@na5y I think the problem is all to do with the impedance of the transformers.

[@Masirah1]

Don't waste your time on this old transistor. New mosfet have better performance and market availability.

[@na5y]

Interested in your suggestions for other FETs to try

[@Masirah1]

@@na5y what is the maximum Vds voltage and required RF output power?

[@na5y]

@@Masirah1 Around 14V and 30-40W class AB would be a target - something with a low IMD

[@4youian]

Hi, I've been using the RD16HHF1 and they are great for HF. I think Mitsubishi are discontinuing them from early next year but I don't see a replacement. Can you please suggest an alternative? 73, VK7IAN

[@Masirah1]

@@na5y I think IPS70R1K4P7S CoolMOS is fitting the criteria. gate voltage should be in the range from 4.7-6.2 volts. so the bias point is around 5.45 volts. I don't think bias current is relevant for FET as it is voltage controlled device not like BJT.