Buck/Booster converter problem?

Hey Long story short I am building a simple power supply. When I add diodes on the output of the buck boost converter I lose regulation on the output UNLESS I have a small load attached. I dont really lose regualtion but voltages start at different levels. Example supposed to be 12-65v, after diodes its 48-99v. but its not a real voltage its like floating. ( BUT if i measure before the diodes on the converter itself regulation and voltages is fine)

So with diodes on the output the Buck Boost converter has improper regulation unless there is a small load. With a load all Regulation is restored normally with correct voltages. I have tried many converters and many types of diodes including schottky diodes. Why is this happening? are my diodes not switching fast enough? I just want diodes on the output for spikes/reverse charge. I just don't get how adding diodes on the output is effecting the regulation. The only thing i can think of is I am messing with the DUTY cycle feed back or something? I was thinking of just adding the meter on the converters output before the didoes to see the proper regulation without a load? please help lol

Can you post a schematic ?
Many converters require a small load.

This converter:
https://www.ebay.ca/itm/DC-DC-Converter-15A-400W-Boost-Converter-Step-up-Voltage-Buck-Boost/132841718224?_trkparms=aid=111001&algo=REC.SEED&ao=1&asc=20160908105057&meid=2ad0efeb057d4788827bb1336af73faa&pid=100675&rk=4&rkt=15&sd=321576507977&itm=132841718224&_trksid=p2481888.c100675.m4236&_trkparms=pageci:05be912a-53e2-11e9-bd54-74dbd1802f30|parentrq:d4fd995b1690a9cb57031844ffece18a|iid:1

The input is straight 12v. The output one diode one way and another the other way. The Diode choice was a 10 amp 1000v diode. I tried other ones too.

About the small load? sorry for my newbie question but What Load can I put that would variate with the voltage and would not overheat? I have no problem putting a load, I can't think of anything that would work with the voltage ranges. ( that is easy to install and doesn't take up space). I intially thought a power resistor but wouldnt I be getting tones of heat on higher voltages or vise versa on the lower end depending on what resistor i use?

Its a poor diagram but maybe it will help what I am trying to say.

Are the didoes supposed to effect the output variation? are they not the right ones? is that even a thing lol

I think your getting a measurement anomaly.
The diodes might be allowing peaks through which your meter is reading.
Why do you have diodes on the o/p?
Its not uncommon for an smps to have a minium load, some control ic's have a minimum pulse width so you need to load the supply enough to shunt this much energy.

The link above describes it as a "boost constant current module". Is it constant current or voltage?

Mike.

Dr pepper you nailed it. I used a different meter and all works fine. In the end it was my meter only picking up the peaks. Thank you !
To answer your question I am using diodes on the output to make sure I dont get any spikes/unwanted voltages when charging or using a motor. Is this not the right thing to do? I am genuinely asking.

Next question is ok to use a FSU10A60 as a blocking diode on the output. Or should I use the 10 amp 1000v Diodes. I dont have much options in the moment lol. I figured the FSU10A60 would work well because there fast switching/High Surge Capability .

Lets say I wanted to use a load that can go from 12v to 65 without getting too hot what would I even use lol?

P.S Why would my meter do that>? Its a Agilent 34401a? I dont get how a 15 dollar meter works but the expensive on throws out a false reading? weird

You dont need the diodes.
But if you want you can put a diode across the motor in reverse bias, ie the cathode (silver stripe) goes to the + end, this will clamp any noise to 0.6v.
Your expensive meter is telling the truth, where as the cheaper one is ignoring the peaks, this is to be expected.
I use a cheapo meter and rubbish 'scope all the time, the good stuff only comes out when its needed.